perm filename CISCO.MSG[1,LES] blob
sn#859863 filedate 1988-07-21 generic text, type C, neo UTF8
COMMENT ⊗ VALID 00131 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00017 00002 Software licensing with Cisco
C00019 00003 ∂08-May-86 1236 BOSACK@SU-SCORE.ARPA Rough draft of the gateway letter
C00024 00004 ∂21-May-86 2058 CS.EAB@Lindy draft license with cisco
C00027 00005 ∂24-May-86 1730 LES Draft License with Cisco
C00037 00006 ∂23-May-86 2122 GD.WHY@Lindy Software licensing with Cisco
C00040 00007 ∂24-May-86 1817 LES Cisco licensing
C00051 00008 ∂27-May-86 1529 GD.WHY@Lindy Cisco licensing
C00052 00009 ∂03-Jun-86 1426 CS.EAB@Lindy
C00057 00010 ∂18-Jun-86 1339 RINDFLEISCH@SUMEX-AIM.ARPA CISCO Software
C00059 00011 ∂25-Jun-86 1545 RINDFLEISCH@SUMEX-AIM.ARPA Re: CISCO Software
C00061 00012 ∂25-Jun-86 1551 NILSSON@SU-SCORE.ARPA Re: CISCO Software
C00063 00013 ∂25-Jun-86 1630 LES Bosack/cisco
C00067 00014 ∂26-Jun-86 1442 RINDFLEISCH@SUMEX-AIM.ARPA Re: CISCO Software
C00069 00015 ∂01-Jul-86 1732 HK.PLD@Lindy issues to be resolved
C00074 00016 ∂01-Jul-86 1927 LES re: issues to be resolved
C00083 00017 ∂02-Jul-86 2300 NILSSON@SU-SCORE.ARPA [Nils Nilsson <NILSSON@SU-SCORE.ARPA>: Matters of Concern]
C00093 00018 ∂09-Jul-86 1526 LES Devaney message
C00094 00019 ∂10-Jul-86 1940 LES Computer Masks taken by cisco
C00109 00020 ∂10-Jul-86 2043 LES Computer Design Thefts
C00112 00021 ∂11-Jul-86 1507 NILSSON@SU-SCORE.ARPA CSD-CF Change
C00115 00022 ∂01-Aug-86 0955 HANSEN@Sierra.Stanford.EDU Re: Sources
C00118 00023 ∂05-Aug-86 1129 LES Pseudo-Sun boards
C00119 00024 .require "memo.pub" source_file
C00131 00025 ∂13-Aug-86 2255 HK.AXA@Forsythe.Stanford.Edu An experiment in electronic messaging
C00135 00026 ∂14-Aug-86 1629 LES re: An experiment in electronic messaging
C00138 00027 ∂22-Aug-86 1736 HK.AXA@Forsythe.Stanford.Edu Copyrighting Tip/Gateway software
C00144 00028 ∂26-Aug-86 1258 HK.AXA@Forsythe.Stanford.Edu Cisco stuff
C00145 00029 Return-Path: <hedrick@topaz.rutgers.edu>
C00148 00030 ∂29-Sep-86 1640 TOM@SU-SCORE.ARPA [Bill <Yeager@SUMEX-AIM.ARPA>: Re: 68k CPU]
C00150 00031 ∂05-Nov-86 2005 LES re: Phone call
C00153 00032 ∂12-Nov-86 1914 LES Bosack Visit
C00155 00033 ∂14-Nov-86 1317 HK.PLD@forsythe.stanford.edu Bosack Visit
C00157 00034 ∂14-Nov-86 1445 LES re: Bosack Visit
C00160 00035 ∂19-Nov-86 2018 GD.WHY@forsythe.stanford.edu re: Bosack Visit
C00163 00036 ∂20-Nov-86 0029 LES re: re: Bosack Visit
C00164 00037 ∂21-Nov-86 0021 LES Protection of Printed Circuit Designs
C00169 00038 ∂21-Nov-86 1230 CS.EJK@forsythe.stanford.edu
C00172 00039 ∂25-Nov-86 0818 CS.EJK@forsythe.stanford.edu
C00173 00040 ∂27-Nov-86 0823 JJW Bridge and Cisco
C00225 00041 ∂09-Dec-86 1356 KHOWARD@Score.Stanford.EDU message
C00226 00042 ∂10-Dec-86 0327 CS.EJK@forsythe.stanford.edu LEN BOSACK MEETING
C00228 00043 ∂16-Dec-86 1928 LES Licensable Materials
C00230 00044 ∂17-Dec-86 1425 GD.WHY@forsythe.stanford.edu Licensable Materials
C00233 00045 ∂19-Dec-86 1851 A.ERIC@GSB-HOW.STANFORD.EDU Ampex memory situation
C00237 00046 ∂21-Dec-86 1715 LES re: Licensable Materials
C00239 00047 ∂24-Dec-86 0900 CS.EJK@forsythe.stanford.edu LEN BOSACK MEETING
C00242 00048 ∂26-Dec-86 1242 HK.JXW@forsythe.stanford.edu LEN BOSACK MEETING
C00244 00049 ∂25-Dec-86 1114 YEAGER@SUMEX-AIM.STANFORD.EDU cisco
C00246 00050 ∂20-Jan-87 1507 CS.EJK@forsythe.stanford.edu STATUS OF CISCO LICENSE/ SUN PATENT
C00251 00051 ∂27-Jan-87 0848 CS.EJK@forsythe.stanford.edu CISCO NEGOTIATIONS STATUS
C00255 00052 ∂28-Jan-87 0828 HK.JXW@forsythe.stanford.edu CISCO NEGOTIATIONS STATUS
C00261 00053 ∂28-Jan-87 1147 hansen@talbots.STANFORD.EDU Stanford Tip/Gateway Software Description.
C00265 00054 ∂28-Jan-87 1427 LES re: CISCO NEGOTIATIONS STATUS
C00267 00055 ∂29-Jan-87 1531 CS.EJK@forsythe.stanford.edu CISCO STATUS REPORT
C00271 00056 ∂29-Jan-87 2015 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU [Bill <Yeager@SUMEX-AIM.STANFORD.EDU>: OTL licensing]
C00274 00057 ∂30-Jan-87 1250 LES re: [Bill <Yeager@SUMEX-AIM.STANFORD.EDU>: OTL licensing]
C00277 00058 ∂01-Feb-87 0932 CSL.ALLISON@Sierra.Stanford.EDU Re: Memory Management
C00279 00059 ∂02-Feb-87 1352 CSL.ALLISON@Sierra.Stanford.EDU re: Memory Management
C00281 00060 ∂30-Jan-87 2152 YEAGER@SUMEX-AIM.STANFORD.EDU gateway code
C00285 00061 ∂03-Feb-87 1357 CS.EJK@forsythe.stanford.edu TIP/GATEWAY ETC. ROYALTY-SHARING AGREEMENT
C00288 00062 ∂03-Feb-87 1924 LES TIP/Gateway Software Royalties
C00291 00063 ∂03-Feb-87 2155 YEAGER@SUMEX-AIM.STANFORD.EDU Re: TIP/Gateway Software Royalties
C00296 00064 ∂04-Feb-87 1015 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU Network Software Royalties
C00300 00065 ∂04-Feb-87 1229 LES re: TIP/Gateway Software Royalties
C00303 00066 ∂06-Feb-87 1525 HK.JXW@forsythe.stanford.edu
C00306 00067 ∂06-Feb-87 1825 CS.EJK@forsythe.stanford.edu ROYALTY SHARING POLICY
C00311 00068 ∂07-Feb-87 0045 CS.EJK@forsythe.stanford.edu CISCO NEGOTIATIONS
C00314 00069 ∂09-Feb-87 0824 YEAGER@SUMEX-AIM.STANFORD.EDU Schnurel duart
C00316 00070 ∂10-Feb-87 1633 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU Re: ROYALTY SHARING POLICY
C00320 00071 ∂11-Feb-87 1025 YEAGER@SUMEX-AIM.STANFORD.EDU Duart
C00322 00072 ∂13-Feb-87 1217 CS.EJK@forsythe.stanford.edu STATUS OF CISCO NEGOTIATIONS
C00324 00073 ∂13-Feb-87 1321 LES re: STATUS OF CISCO NEGOTIATIONS
C00325 00074 ∂13-Feb-87 1250 CS.EJK@forsythe.stanford.edu Ethernet boards, UART boards
C00328 00075 ∂13-Feb-87 1556 LES re: Ethernet boards, UART boards
C00332 00076 ∂17-Feb-87 0908 CS.EJK@forsythe.stanford.edu SUN Boards
C00334 00077 ∂18-Feb-87 1610 CS.EJK@forsythe.stanford.edu cisco questions and comments
C00337 00078 ∂18-Feb-87 1927 LES re: cisco questions and comments
C00339 00079 ∂18-Feb-87 1618 HK.JXW@forsythe.stanford.edu SUN Boards
C00340 00080 ∂19-Feb-87 0837 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU re: cisco questions and comments
C00342 00081 ∂19-Feb-87 0834 CS.EJK@forsythe.stanford.edu
C00345 00082 ∂20-Feb-87 0803 TOM@Score.Stanford.EDU meis/030
C00348 00083 ∂20-Feb-87 1336 TOM@Score.Stanford.EDU re: meis/030
C00350 00084 ∂20-Feb-87 1403 LES re: cisco agreements
C00352 00085 ∂23-Feb-87 1028 CS.EJK@forsythe.stanford.edu
C00357 00086 ∂23-Feb-87 1507 LES reply to message
C00358 00087 ∂23-Feb-87 1747 LES "Cisco" UART board
C00361 00088 ∂26-Feb-87 1927 LES Sun & Ethernet Board Support
C00363 00089 ∂26-Feb-87 0051 CS.EJK@forsythe.stanford.edu cisco
C00369 00090 ∂26-Feb-87 2134 LES re: cisco
C00381 00091 ∂27-Feb-87 1240 LES Modified Sun boards
C00383 00092 ∂27-Feb-87 1429 HK.NJR@forsythe.stanford.edu
C00384 00093 ∂03-Mar-87 1055 CS.EJK@forsythe.stanford.edu cisco meeting
C00389 00094 ∂03-Mar-87 1559 LES re: cisco meeting
C00400 00095 ∂04-Mar-87 1639 HK.NJR@forsythe.stanford.edu
C00402 00096 ∂04-Mar-87 1731 LES Cisco Fiasco
C00404 00097 ∂04-Mar-87 1837 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU Re: Cisco Mess
C00409 00098 ∂05-Mar-87 1407 LES CPU Version P-20
C00412 00099 ∂10-Mar-87 1758 CS.EAB@forsythe.stanford.edu
C00414 00100 ∂24-Mar-87 1353 CS.EJK@forsythe.stanford.edu Sun Status
C00416 00101 ∂31-Mar-87 0848 CS.EJK@forsythe.stanford.edu
C00417 00102 ∂31-Mar-87 1140 LES Baskett's Feelings
C00419 00103 ∂31-Mar-87 1624 LES Cisco Ethernet Board
C00421 00104 ∂02-Apr-87 0852 CS.EJK@forsythe.stanford.edu Ethernet boards: cisco
C00423 00105 ∂02-Apr-87 1216 LES re: Ethernet boards: cisco
C00427 00106 ∂03-Apr-87 0857 HK.JXW@forsythe.stanford.edu Ethernet boards
C00428 00107 ∂10-Apr-87 0847 CS.EJK@forsythe.stanford.edu sun
C00430 00108 ∂11-Apr-87 1232 HK.JXW@forsythe.stanford.edu ethernet
C00433 00109 ∂11-Apr-87 2311 LES re: ethernet
C00436 00110 ∂12-Apr-87 2135 HK.JXW@forsythe.stanford.edu re: ethernet
C00437 00111 ∂17-Apr-87 1657 LES Ethernet Board Burn-in
C00438 00112 ∂21-Apr-87 1122 CS.EJK@forsythe.stanford.edu cisco license negotiations
C00440 00113 ∂21-Apr-87 1352 LES re: cisco license negotiations
C00443 00114 ∂22-Apr-87 1155 GOTELLI@Score.Stanford.EDU Cisco Computer Account
C00445 00115 ∂18-May-87 1858 LES Schnurle Letter
C00448 00116 ∂27-Jul-87 1258 HK.JXW@forsythe.stanford.edu Returned mail: User unknown
C00450 00117 ∂06-Aug-87 1236 HK.JXW@forsythe.stanford.edu cisco
C00454 00118 ∂07-Aug-87 1600 LES re: cisco
C00456 00119 ∂11-Aug-87 1020 TOM@Score.Stanford.EDU CISCO
C00458 00120 ∂13-Aug-87 1105 GD.WHY@forsythe.stanford.edu cisco
C00460 00121 ∂19-Aug-87 1707 GD.WHY@forsythe.stanford.edu cisco settlement and related request
C00463 00122 ∂20-Aug-87 0856 HK.JXW@forsythe.stanford.edu cisco settlement and related request
C00465 00123 ∂27-Oct-87 0850 AR.PMB@forsythe.stanford.edu Future Business With Cisco Systems
C00467 00124 ∂27-Oct-87 1141 LES re: Future Business With Cisco Systems
C00469 00125 ∂28-Oct-87 0552 HK.JXW@forsythe.stanford.edu Future Business With Cisco Systems
C00471 00126 ∂29-Oct-87 1103 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU Re: Royalty sharing for tip/gateway
C00474 00127 ∂29-Oct-87 1552 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU CISCO Royalties for TIP/Gateway Software
C00478 00128 ∂29-Oct-87 1713 nilsson@Tenaya.Stanford.EDU CISCO Royalties for TIP/Gateway Software
C00481 00129 ∂30-Oct-87 1714 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU Re: CISCO Royalties for TIP/Gateway Software
C00485 00130 ∂11-Nov-87 1205 ARK gfs
C00486 00131 ∂23-Feb-88 1402 AIR
C00487 ENDMK
C⊗;
Software licensing with Cisco
As you may know, we need a new communications gateway between ARPAnet
and the Stanford Ethernet system. Len Bosack, wearing his hat as head
of cisco Systems, offers to provide software for such a gateway at no
cost and to maintain it for a fixed period in return for access to certain
software and facilities at Stanford.
I asked Len Bosack for a draft agreement covering the proposed transaction
and received the following. I understand that an agreement between cisco
and OTL is in the offing. This clearly should be scrutinized by legal counsel.
I invite your comments on both the structure and content of the agreement.
Perhaps there should be only one agreement instead of two.
Les Earnest
∂08-May-86 1236 BOSACK@SU-SCORE.ARPA Rough draft of the gateway letter
Received: from SU-SCORE.ARPA by SU-AI.ARPA with TCP; 8 May 86 12:36:42 PDT
Date: Thu 8 May 86 12:37:08-PDT
From: Len Bosack <BOSACK@SU-SCORE.ARPA>
Subject: Rough draft of the gateway letter
To: LES@SU-AI.ARPA
Message-ID: <12205114643.11.BOSACK@SU-SCORE.ARPA>
**DRAFT**DRAFT**DRAFT**DRAFT**DRAFT**DRAFT**
This letter is a summary of the discussions we have had over the past
few weeks with regard to cisco Systems and the Stanford Arpanet gateway,
including the recitations and agreement we believe appropriate.
I am a Director and Officer of cisco Systems, a California corporation.
As a conflict-of-interest may exist, you have been asked to handle
these matters as Chairman of the Computer Facilities committee.
cisco Systems is negotiating with the Stanford Office of Technolgy Licensing
a reciprocal license covering non-Arpa gateways and communications servers
at Stanford.
For several years, the Arpanet node known as Stanford-Gateway or Golden
has acted as a gateway between the Arpanet and the Stanford network. Golden
is a LSI11/23 and due to limited code address space is becoming unsuitable
for continued service as a gateway. cisco Systems is developing extensions
to its ASM/AGS products which will enable the AGS to function as a full-service
Arpanet gateway. This product will be quite suitable as a replacement for
Golden.
By running the extended AGS software on hardware now owned by Stanford,
Stanford can avoid an additional expenditure of roughly $20000. cisco Systems
would benefit from having access to certain facilities at Stanford in the
development and debugging of the extended AGS functions.
To achieve this mutual benefit we agree as follows:
1. cisco Systems will be granted access to the technical facilties needed
for the work at the Stanford site. Generally, any 4.2BSD or later system
will suffice. We expect the work at Stanford to take roughly one month until
initial production trials.
2. cisco Systems grants Stanford the nonexclusive right to use the extended
AGS software in any Arpanet gateway operated by Stanford for its own use.
3. cisco Systems will make any additional improvements made to the extended AGS
software available to Stanford until 1-July-1990. After that time, Stanford
may continue to use the 1-July-1990 version as provided in 2. or negotiate
a further agreement with cisco Systems for additional updates.
4. Stanford recognizes that the software is the valuable proprietary property
of cisco Systems and will take reasonable steps to safeguard this property.
Any student, faculty, staff, consultant or other person granted access to this
software by Stanford will be informed of the valuable nature of the property
and their obligation to safeguard it.
5. As the nature of this agreement is cooperation for mutual benefit, each
party holds the other harmless with regard to any use of the software.
6. All express and implied warranty is disclaimed. Stanford acknowledges that
it is to be considered an expert in accepting the disclaimer of all implied
warranty.
-------
∂21-May-86 2058 CS.EAB@Lindy draft license with cisco
Received: from SU-LINDY.ARPA by SU-AI.ARPA with TCP; 21 May 86 20:57:51 PDT
Date: Wed, 21 May 86 20:59:43 PDT
From: Elizabeth A. Batson <CS.EAB@SU-Forsythe.ARPA>
To: LES@SU-AI
Subject: draft license with cisco
Les,
A few comments (questions, really) about the cisco proposal
for the Arpanet gateway.
1. Who is actually going to be doing the
work? Who will supervise it, if anyone?
If they are Stanford employees, is this time
commitment realistic? Will there be a possible conflict
as to what related software belongs to Stanford vs. cisco?
2. What is the extent of the technical
facilities required? What is the dollar value
of that access?
3. Will Stanford be providing any people time/
resources to the project? If so, then will those people
be willing to see their work commercialized by
cisco?
4. What is the total contribution of each party?
How does it compare to the option of Stanford doing
it ourselves and then receiving royalties from
a commercial license.
Les, if I were you, my main concern would the conflict
of interest implies by Question 1. Questions 2-4 are
basically asking if Stanford is getting a good deal.
I don't see any reason to tie the two agreement together,
the development agreement should be relatively short and
simple if you decide to go ahead. You would want to add
a development schedule to make sure that cisco proceeds with
the work in a timely manner. If you want help with the
agreement itself, I would be happy to contribute.
Please let me know what you decide to do.
Elizabeth Batson
∂24-May-86 1730 LES Draft License with Cisco
To: cs.eab@SU-FORSYTHE.ARPA
CC: Nilsson@SU-SCORE.ARPA
Elizabeth,
Here are my responses to your questions about the Cisco proposal
for the Arpanet gateway.
1. Who is actually going to be doing the work?
Who will supervise it, if anyone?
As I understand it, the principal work, which is software development,
is being done by Kirk Lougheed of EE under the supervision of Len Bosack.
If they are Stanford employees, is this time commitment realistic?
If you mean by this "Will their time commitment to Cisco not conflict
with their Stanford responsibilities?" my answer is "There probably is
some conflict, particularly for Lougheed." I tried to reach his supervisor
(whose name now escapes me) several months ago to discuss this issue but
didn't succeed.
Will there be a possible conflict as to what related software belongs to
Stanford vs. cisco?
There was some contention over this issue earlier, based on issues raised
by Bill Yeager of SUMEX, who wrote most of the programs that Cisco based
their development on. My impression is that these issues went away after
Yeager was invited to do some consulting for Cisco. Yeager (wearing his
Stanford hat) now plans to write some Stanford-specific supplementary
programs that will run within the Cisco environment and that, presumably,
will belong to Stanford.
2. What is the extent of the technical facilities required?
What is the dollar value of that access?
They need (and are using) a Unix environment. I don't know how much
computer time this project is gobbling, but would guess that the dollar
value at University rates is on the order of $20,000. They also are using
a small specialized computer that was built by the Computer Facilities
Group here in Computer Science specifically for use as a gateway. The
parts cost of this device is on the order of $15,000. The engineering
cost is hard to estimate, but Stanford will presumably continue use it to
provide ARPAnet gateway service. (Cisco has "independently" developed an
essentially identical computer that they will fabricate as part of their
product.)
3. Will Stanford be providing any people time/ resources to the project?
If so, then will those people be willing to see their work commercialized
by cisco?
As mentioned earlier, Bill Yeager seems to be voluntarily involved now
in a way that would apparently benefit Stanford.
4. What is the total contribution of each party?
Hard to say without carefully examining both the programs originally
written by Yeager and the derivative work done by Lougheed. One unclean
thing about this project is that the derivative work was originally
undertaken as "Stanford" project and became a Cisco project only after I
started questioning whether it made sense for us to underwrite such a
development given that there would soon be a commercial product (from
Proteon) that could do the job.
How does it compare to the option of Stanford doing it ourselves and then
receiving royalties from a commercial license.
In general, Stanford doesn't do very well at this sort of thing. The
standards of documentation and idiot-proofing needed for commercial
products seem to be beyond the capabilities of most programmers who
work here.
In case I have given the impression that I am uncomfortable with this
entire undertaking, I would have to say that is an accurate conclusion.
The main things that make me uncomfortable are as follows.
(1) Cisco is using its Stanford connections in ways that no outside
company could do. Indeed, it is unlikely that an outside company could
have learned of this opportunity unless some of the principals left
Stanford to form or join such a company;
(2) There still are no clear boundaries between network development
projects being done for Stanford and those being done for Cisco.
This leads other Stanford employees to be (understandably) suspicious
about why certain things are being done and to speculate about improper
motives.
(3) The most serious problem, as I see it, is that the available
administrative tools do not appear to be adequate for keeping projects
like this under control. We could, of course, try to suppress it,
but if we consider the best interests of Stanford it appears to be
better to make a deal. Unfortunately, we do not have much leverage
over the terms, given that the work has been nearly completed (using
our resources) without any formal agreement.
The clean way to have handled this, as I see it, would have been to make a
deal in advance spelling out the rights, obligations and benefits of both
parties, including a duration limit on any access to Stanford's
facilities. If they didn't like the terms, of course, they would be free
to leave Stanford and do it on their own. Unfortunately, that would be
easier to do than it should be because of an apparent (to me) flaw in
Stanford's position -- we decline to use trade secret laws to protect
software.
If we used trade secret protection, we could prohibit our employees
from taking software written by them or others, reorganizing it to avoid
copyright protection and selling it as their product. Without trade
secret law, I don't see much chance of controlling the flow. So we might
as well declare that ripping off the University is one of the fringe
benefits of working here.
Coming back to the current situation, there is noticeable distress among
some of our staff who work for Len Bosack, know of his outside business
interests and think that many of the things being undertaken by him
ostensibly as Stanford projects may actually be a part of Cisco's
business plan. My goals are:
(1) to get a new ARPAnet gateway at no direct cost to Stanford;
(2) see you get whatever other benefits Stanford can reasonably expect
from a business arrangement with Cisco, such as royalties from their
gateway product;
(3) to clearly separate Cisco's R&D efforts from Stanford's development
projects, if necessary by inviting the principals to either go away
or terminate their business.
Les Earnest
∂23-May-86 2122 GD.WHY@Lindy Software licensing with Cisco
Received: from SU-LINDY.ARPA by SU-AI.ARPA with TCP; 23 May 86 20:30:50 PDT
Date: Fri, 23 May 86 20:32:45 PDT
From: Bill Yundt <GD.WHY@SU-Forsythe.ARPA>
To: LES@SU-AI
Subject: Software licensing with Cisco
REPLY TO 05/12/86 16:42 FROM LES@SU-AI.ARPA: Software licensing with Cisco
Les
Sorry for delayed reply.
The agreement proposed by Len seems fair to me and a good
business proposition. I don't see any particular reason
to bundle it into a broader OTL agreement but they may.
I am not sure what the legal view will be. I would argue
strongly that the use of Stanford facilities would, in this
case be for the development of a product to our specifications,
albeit by a commercial enterprise for their subsequent use.
I don't know if this runs afoul of the tax and policy
restrictions prohibiting commercial use of Stanford owned
property/facilities, but would hope not under the circumstances
that the work is commissioned for our use and it is merely
the payment for it that is being provided by use of Stanford
facilities.
There may be a little stickier problem with the use of a
Unix system, per se, as the Unix license itself (covering
BSD....not third party supplied object-only licensed versions
like SUN Unix or ULTRIX) absolutely prohibits uses other
than purely educational and institutional administration.
I am forwarding this note with a copy of the draft agreement
to Adrian Arima for review and consideration of the foregoing.
Let me know if you hear from OTL....you may need to send a
hard copy to Elizabeth Batson.
....Bill
To: LES@SU-AI.ARPA
∂24-May-86 1817 LES Cisco licensing
To: gd.why@SU-FORSYTHE.ARPA
Bill,
Thanks for your note. I also just heard from Elizabeth Batson; the
attached response to her is built around her questions.
I see enough alligators crawling around in this swamp without getting
into Unix licensing or tax restrictions. My goal is to get what we
can out of the current situation and end it as soon as possible.
Les
------------------------------------------------
To: cs.eab@SU-FORSYTHE.ARPA
CC: Nilsson@SU-SCORE.ARPA
Subject: Draft License with Cisco
Elizabeth,
Here are my responses to your questions about the Cisco proposal
for the Arpanet gateway.
1. Who is actually going to be doing the work?
Who will supervise it, if anyone?
As I understand it, the principal work, which is software development,
is being done by Kirk Lougheed of EE under the supervision of Len Bosack.
If they are Stanford employees, is this time commitment realistic?
If you mean by this "Will their time commitment to Cisco not conflict
with their Stanford responsibilities?" my answer is "There probably is
some conflict, particularly for Lougheed." I tried to reach his supervisor
(whose name now escapes me) several months ago to discuss this issue but
didn't succeed.
Will there be a possible conflict as to what related software belongs to
Stanford vs. cisco?
There was some contention over this issue earlier, based on issues raised
by Bill Yeager of SUMEX, who wrote most of the programs that Cisco based
their development on. My impression is that these issues went away after
Yeager was invited to do some consulting for Cisco. Yeager (wearing his
Stanford hat) now plans to write some Stanford-specific supplementary
programs that will run within the Cisco environment and that, presumably,
will belong to Stanford.
2. What is the extent of the technical facilities required?
What is the dollar value of that access?
They need (and are using) a Unix environment. I don't know how much
computer time this project is gobbling, but would guess that the dollar
value at University rates is on the order of $20,000. They also are using
a small specialized computer that was built by the Computer Facilities
Group here in Computer Science specifically for use as a gateway. The
parts cost of this device is on the order of $15,000. The engineering
cost is hard to estimate, but Stanford will presumably continue use it to
provide ARPAnet gateway service. (Cisco has "independently" developed an
essentially identical computer that they will fabricate as part of their
product.)
3. Will Stanford be providing any people time/ resources to the project?
If so, then will those people be willing to see their work commercialized
by cisco?
As mentioned earlier, Bill Yeager seems to be voluntarily involved now
in a way that would apparently benefit Stanford.
4. What is the total contribution of each party?
Hard to say without carefully examining both the programs originally
written by Yeager and the derivative work done by Lougheed. One unclean
thing about this project is that the derivative work was originally
undertaken as "Stanford" project and became a Cisco project only after I
started questioning whether it made sense for us to underwrite such a
development given that there would soon be a commercial product (from
Proteon) that could do the job.
How does it compare to the option of Stanford doing it ourselves and then
receiving royalties from a commercial license.
In general, Stanford doesn't do very well at this sort of thing. The
standards of documentation and idiot-proofing needed for commercial
products seem to be beyond the capabilities of most programmers who
work here.
In case I have given the impression that I am uncomfortable with this
entire undertaking, I would have to say that is an accurate conclusion.
The main things that make me uncomfortable are as follows.
(1) Cisco is using its Stanford connections in ways that no outside
company could do. Indeed, it is unlikely that an outside company could
have learned of this opportunity unless some of the principals left
Stanford to form or join such a company;
(2) There still are no clear boundaries between network development
projects being done for Stanford and those being done for Cisco.
This leads other Stanford employees to be (understandably) suspicious
about why certain things are being done and to speculate about improper
motives.
(3) The most serious problem, as I see it, is that the available
administrative tools do not appear to be adequate for keeping projects
like this under control. We could, of course, try to suppress it,
but if we consider the best interests of Stanford it appears to be
better to make a deal. Unfortunately, we do not have much leverage
over the terms, given that the work has been nearly completed (using
our resources) without any formal agreement.
The clean way to have handled this, as I see it, would have been to make a
deal in advance spelling out the rights, obligations and benefits of both
parties, including a duration limit on any access to Stanford's
facilities. If they didn't like the terms, of course, they would be free
to leave Stanford and do it on their own. Unfortunately, that would be
easier to do than it should be because of an apparent (to me) flaw in
Stanford's position -- we decline to use trade secret laws to protect
software.
If we used trade secret protection, we could prohibit our employees
from taking software written by them or others, reorganizing it to avoid
copyright protection and selling it as their product. Without trade
secret law, I don't see much chance of controlling the flow. So we might
as well declare that ripping off the University is one of the fringe
benefits of working here.
Coming back to the current situation, there is noticeable distress among
some of our staff who work for Len Bosack, know of his outside business
interests and think that many of the things being undertaken by him
ostensibly as Stanford projects may actually be a part of Cisco's
business plan. My goals are:
(1) to get a new ARPAnet gateway at no direct cost to Stanford;
(2) see you get whatever other benefits Stanford can reasonably expect
from a business arrangement with Cisco, such as royalties from their
gateway product;
(3) to clearly separate Cisco's R&D efforts from Stanford's development
projects, if necessary by inviting the principals to either go away
or terminate their business.
Les Earnest
∂27-May-86 1529 GD.WHY@Lindy Cisco licensing
Received: from SU-LINDY.ARPA by SU-AI.ARPA with TCP; 27 May 86 15:29:03 PDT
Date: Tue, 27 May 86 15:29:12 PDT
From: Bill Yundt <GD.WHY@SU-Forsythe.ARPA>
To: LES@SU-AI
Subject: Cisco licensing
REPLY TO 05/24/86 18:19 FROM LES@SU-AI.ARPA: Cisco licensing
Les, I sympathize with (and share) your goal, having little
desire to become alligator bait. Also agree with the responses
you gave to Elizabeth....
Thanks....Bill
To: LES@SU-AI.ARPA
∂03-Jun-86 1426 CS.EAB@Lindy
Received: from SU-LINDY.ARPA by SU-AI.ARPA with TCP; 3 Jun 86 14:26:06 PDT
Date: Tue, 3 Jun 86 14:25:38 PDT
From: Elizabeth A. Batson <CS.EAB@SU-Forsythe.ARPA>
To: LES@SAIL
Pat,
I thought that you might be able to use a little bit of
background on the cisco Systems situation from the OTL
perspective.
The principals in cisco are Len Bosack, director of computer
facilities for the Comp. Sci dept and Kirk Lougheed, a programmer
who works in Bob Dutton's group in Elec. Eng. Sandy Lerner,
Bosack's wife, also works for the company. Sandy managed
the computing facilities for the Business School but at present
is not employed by Stanford.
This company, in its short history, has raised
questions of conflict of interest in at least three different
arenas.
1. Cisco has taken a hardware product designed by
the Department of Computer Science and sold it to
the Business School and Operations Research. Bosack
claims that the product uses is just know-how and what would
be trade secrets if we had them and thus does not need
a license. Jerry Chehok of Internal Audit is looking into
these sales and I believe that he will be contacting you
about his findings.
2. Cisco has developed some software used for
Ethernet gateways and communications servers.
It is clearly based on work done by Bill Yeager at
SUMEX, and they have approached OTL for a non-exclusive
license to Yeager's work. They have also promised Yaeger
some direct compensation. There are additional enhancements
that some believed were being done by the Dept. of
Computer Science but which Bosack claims were done
independently and thus belong to cisco. I don't have
enough information yet to determine the ownership of these
enhancements.
3. Stanford needs a new Arpanet node. The cost of purchase
from an outside vendor is about $20,000. Cisco has offered to
the work for free (using Stanford facilities) provided that
they would own the copyright (with STanford having a perpetual
license to use it). Cisco would also provide free updates
for 4 years. The work would be done primarily by Lougheed
under the direction of Bosack. The project includes
the disputed enhancements mentioned above.
Les Ernest is reviewing this proposal for the Dept. of
Computer Science and he can discuss the details with you.
From my narrow perspective,
if Bosack is involved in a serious conflict of interest,
I don't want to validate his position by giving him a
software license. How do you feel about this?
Should you be involved in assessing the propriety of
such a license? If it weren't for this problem, we
would routinely grant a non-exclusive license the company.
I have been stalling Bosack while trying to clearer
picture of the situation. I will wait to hear from
you before getting involved in any serious licensing
discussions.
Elizabeth Batson
∂18-Jun-86 1339 RINDFLEISCH@SUMEX-AIM.ARPA CISCO Software
Received: from SUMEX-AIM.ARPA by SU-AI.ARPA with TCP; 18 Jun 86 13:38:11 PDT
Date: Tue 17 Jun 86 18:03:55-PDT
From: T. C. Rindfleisch <Rindfleisch@SUMEX-AIM.ARPA>
Subject: CISCO Software
To: Bosack@SU-SCORE.ARPA
cc: Rindfleisch@SUMEX-AIM.ARPA, Devaney@SUMEX-AIM.ARPA, Nilsson@SU-SCORE.ARPA,
LES@SU-AI.ARPA
Message-ID: <12215659892.44.RINDFLEISCH@SUMEX-AIM.ARPA>
Len, I've been asked by Pat Devaney in the Provost's Office to try to formulate
a judgement about the possible relationship between the network gateway
software you want to market through CISCO and Stanford software that may have
served as a precursor. I'd like to arrange with you to get on-line access to
a) your CISCO source code and b) the code running currently in CSD and EE
gateways and TIP's for review and comparison. I will arrange for obtaining
copies of the SSRG codes from Bill Yeager.
Let me know what would be the best way to proceed. Thanks, Tom R.
-------
∂25-Jun-86 1545 RINDFLEISCH@SUMEX-AIM.ARPA Re: CISCO Software
Received: from SUMEX-AIM.ARPA by SU-AI.ARPA with TCP; 25 Jun 86 15:44:21 PDT
Date: Wed 25 Jun 86 15:12:49-PDT
From: T. C. Rindfleisch <Rindfleisch@SUMEX-AIM.ARPA>
Subject: Re: CISCO Software
To: Bosack@SU-SCORE.ARPA
cc: Devaney@SUMEX-AIM.ARPA, Nilsson@SU-SCORE.ARPA, LES@SU-AI.ARPA,
Rindfleisch@SUMEX-AIM.ARPA
In-Reply-To: <12215659892.44.RINDFLEISCH@SUMEX-AIM.ARPA>
Message-ID: <12217725897.10.RINDFLEISCH@SUMEX-AIM.ARPA>
Len, I haven't seen a reply from you to my message of the 17th about CISCO
software. I will be leaving this Saturday for two weeks and am trying to pull
together loose ends so I can let Pat Devaney know where things stand. Could
you check one of the following boxes:
--- Never saw your message
--- Saw your message and sources will be ready by ------
--- Other -------
Thanks, Tom R.
-------
∂25-Jun-86 1551 NILSSON@SU-SCORE.ARPA Re: CISCO Software
Received: from SU-SCORE.ARPA by SU-AI.ARPA with TCP; 25 Jun 86 15:51:09 PDT
Date: Wed 25 Jun 86 15:49:22-PDT
From: Nils Nilsson <NILSSON@SU-SCORE.ARPA>
Subject: Re: CISCO Software
To: Rindfleisch@SUMEX-AIM.ARPA
cc: Bosack@SU-SCORE.ARPA, Devaney@SUMEX-AIM.ARPA, LES@SU-AI.ARPA,
NILSSON@SU-SCORE.ARPA
In-Reply-To: <12217725897.10.RINDFLEISCH@SUMEX-AIM.ARPA>
Message-ID: <12217732549.13.NILSSON@SU-SCORE.ARPA>
I'd very much like to see the CISCO software matter brought to a conclusion,
and will involve myself in all of this if expediting looks like it is needed.
I presume Tom R. and Len B. will be in touch either in person or by phone
in the next few hours to figure out how progress can be achieved. If there
is no progress in the next day I'd like to know that. Thanks, -Nils
-------
∂25-Jun-86 1630 LES Bosack/cisco
To: nilsson@SU-SCORE.ARPA
I had a talk with Len awhile ago, as planned. Len says that he will get
the cisco disclosure to you and me by tomorrow.
He says that he had not responded to Tom Rindfleisch's request for source
code on the gateway software because they (cisco) are unwilling to provide
machine source; they are willing to provide hard copy. I asked why he had
not said that to Tom instead of not responding. He offered no explanation.
I also said that, based on a copyright court case that I just learned of,
the gateway software is rather clearly derived from the Stanford software.
He said his lawyer thought otherwise but expressed a willingness to act as
if that were true and to negotiate a royalty licensing agreement with
Stanford.
We discussed the need for a rate adjustment in the cost centers. He said
that things are not as bad as they appear because the new SAIL terminals
should have been capitalized and depreciated over several years instead of
being treated as current expenses. He said that he will give priority
attention to making the needed rate adjustments.
I grumbled about the fact that the Facilities Group has still not produced
bills in the form of electronic mail that is sent to all users for
verification, even though he promised that they would do it by last
February. I pointed out that without verification we will continue to
have undetected blunders, such as Ginsbergs SAIL account -- he switched
from McCarthy's project to Genesereth's in July 1984 but continued to be
charged to McCarthy. We can now repair that damage only back to the
beginning of the current fiscal year (Sept. 1). I also noted that we
still do not have subaccount billing, which I requested last Fall because
it is required to track courtesy accounts. Len said he would give priority
attention to these problems.
It appears to me that Len is not in control of his group, probably for
lack of putting in enough time on it. Consequently, they are close to
mutiny. I think that we should continue to press hardest for a resolution
of the most immediate problem: cisco's relations with Stanford. Assuming
that we get through that satisfactorily, I propose that we ask Len to do
some longer range planning for his group, including milestones and dates,
put it in writing, and ensure that it is carried out as planned. As near
as I can tell, their current planning on a given morning does not extend
past noon.
Les
∂26-Jun-86 1442 RINDFLEISCH@SUMEX-AIM.ARPA Re: CISCO Software
Received: from SUMEX-AIM.ARPA by SU-AI.ARPA with TCP; 26 Jun 86 14:41:32 PDT
Date: Thu 26 Jun 86 14:42:14-PDT
From: T. C. Rindfleisch <Rindfleisch@SUMEX-AIM.ARPA>
Subject: Re: CISCO Software
To: NILSSON@SU-SCORE.ARPA
cc: Devaney@SUMEX-AIM.ARPA, LES@SU-AI.ARPA, Rindfleisch@SUMEX-AIM.ARPA
In-Reply-To: <12217732549.13.NILSSON@SU-SCORE.ARPA>
Message-ID: <12217982473.10.RINDFLEISCH@SUMEX-AIM.ARPA>
Nils, this is to let you know that I still haven't heard from Len, by any
medium. Even if he were to respond now, I have no time left to devote to the
CISCO matter before I leave. Since I do not have any supervisory
responsibility for Len, I feel a bit like an outsider in this increasingly
untenable situation. Let me know how you all would like to proceed and how I
can help when I get back on the 13th.
Good luck! Tom R.
-------
∂01-Jul-86 1732 HK.PLD@Lindy issues to be resolved
Received: from SU-LINDY.ARPA by SU-AI.ARPA with TCP; 1 Jul 86 17:32:40 PDT
Date: Tue, 1 Jul 86 17:32:35 PDT
From: Patricia L. Devaney <HK.PLD@SU-Forsythe.ARPA>
To: LES@SAIL
Subject: issues to be resolved
Dear Les and Len: Here's hopefully a more legible list of
issues which yet need to be resolved, as discussed at our
meeting yesturday. Holler if my recollection of the issues
and their resolution doesn't match yours. Thanks, Pat
l) Len needs to submit a written disclosure to Nils re the level
and nature of his outside consulting activities, including a
declaration that either he is not using any University resources
for non-University purposes, or, if he is, the schedule by
which he will cease such use;
2) Nils needs to write a written approval re the terms and conditions
under which Len has permission to consult; futhermore, Nils/
Les needs to clarify Len's reporting relationship within the
Department. If there is any question as to Len's scope of
University duties, these should also be clarified.
3) The Computer Science Department and our office need to work
together to develop a policy statement re unrelated business
activities in CS--i.e., conditions under which any Department
resources may be made available for non-University purposes;
Les has drafted a position paper which Pat needs to review;
4) Any requests to CS made by other University departments for
services/products which are outside the scope of the Department's
normal activities and which involve resources which might be
provided by outside vendors, should be referred to Les
for resolution;
5) Acquisition of goods/services from Cisco Systems by Stanford
must be approved on a case-by-case basis from this office.
6) Agreement needs to be reached re the extent of Stanford's
ownership of the software involved in the communication
servers being developed/commercialized by Cisco Systems.
Sufficient detail involved in this agreement needs to be
reached so as to clarify Stanford's rights to license
the original or derivative works to others without
accountability to Cisco. Following resolution of this
issue, terms of a mutually-agreeable license (which may
also include cross licenses with Cisco-owned works) may
be worked out by OTL, Bill Yundt, Cisco, et al.
Les, Len, and others will work together to resolve the
these issues, hopefully within the next week.
Anything I've missed here?
Thanks to you both for your continued cooperation and help.
Cheers, Pat
To: LES@SAIL, BOSACK@SCORE
cc: HK.PLD
∂01-Jul-86 1927 LES re: issues to be resolved
To: HK.PLD@SU-FORSYTHE.ARPA
CC: Nilsson@SU-SCORE.ARPA
[In reply to message sent Tue, 1 Jul 86 17:32:35 PDT.]
Your list of action items in the Bosack/cisco case fits my recollection.
I have talked to Bill Yeager and he is making a copy of his gateway
software and an explanatory report that will come to me. I have not
reached either Bob Dutton or Steve Hansen yet, so I have not tried to get
in touch witk Kirk Lougheed.
For your information, I stumbled across another issue this afternoon that
I have not yet had a chance to discuss with Len, but I want to tell you
because it will almost certainly cause more problems in our negotiations
with cisco. This involves a computer board that was developed at Stanford
and that apparently has migrated to cisco. The background of this board
is a bit complicated. I will tell you more than you probably want to
know.
During 1980-81, the so-called "SUN" (for Stanford University Network)
computer board was developed here. It used the then-new Motorola 68000
computer chip and 64k RAM chips. The principal designer was a student
named Andy Bechtolsheim, but many people contributed to the design,
including me. Andy asked in advance of undertaking this project whether
he would be able to develop it commercially and was told (by me) that he
could.
Many copies of the original SUN board were made at Stanford and it was
used both as the central element in a family of workstations assembled her
and as the active element in a number of gateways used in the burgeoning
Stanford local area networks, based on ethernet technology.
Andy then went to OTL and asked whether a license was required in order
for him to develop the board commercially. I was not a party to that
negotiation, but I understand that he was told that he was free to pursue
it without a license, given that there was apparently nothing patentable
in the device.
Andy found that a number of companies were interested in using the SUN
board and signed a half-dozen or more licensing agreements. He provided
each licensee with circuit board artwork and documentation (both developed
and produced at Stanford, of course) and collected both up-front fees and
royalties.
I left Stanford about that time and my company, Imagen, became his first
licensee. Imagen paid Andy $10,000 in royalties; some later licensees
paid as much as $40,000 each. Andy provided the same artwork and
documentation to Stanford at no cost and these boards continued to be
fabricated at Stanford for some time.
(I also negotiated on behalf of Imagen a separate licensing agreement with
Stanford for an invention that my partner had made as a Stanford staff
member. This subsequently yielded Stanford something over $400,000 in
equipment discounts and royalty payments.)
A little later, Andy left Stanford to become a founder of Sun Microsystems,
without finishing his PhD, alas. He further refined the design of the SUN
board and ceased licensing either the older design or the new one. Sun
sold many of the improved computer boards to Stanford on a heavily
discounted basis. (In the four years since then, Sun has become an
enormous commercial success -- I understand that Andy's shares now have a
market value of about $25 million.)
[The following is a hearsay account from a single reliable source. I have
not yet cross-checked it with other sources.]
Early this year, the CSD facilities group headed by Len Bosack found that
they needed some more Sun-like computer boards for building gateways.
Sun had stopped selling such boards in the small quantites that Stanford
needed. Len and others figured out that a relatively minor change in the
old SUN board would permit it to use the newer and more economical 256k
RAM chips. They experimentally rewired a board here at Stanford, verified
that it worked, then fabricated 10 of the improved boards for Stanford.
There was one funny thing about the fabrication project. While the parts
were paid for by Stanford, the new artwork for the printed circuit card
was paid for by cisco. It is possible that cisco also paid for the
fabrication of some number of these boards for their own use as part of
the same transactions.
Conclusion
We now find that, even though the development work for this improved
computer board was done at Stanford by Stanford employees, we do not have
a copy of the PC card artwork and so are not in a position to build more
such boards when we need them.
Unless the apparent facts of this case are refuted, I believe that we
should demand that cisco turn over the PC artwork to Stanford so that we
will be in a position to fabricate more of these computer boards if we
need them. We should be willing to pay the modest bill for modifying the
original Stanford artwork, which was apparently contracted out.
We should also demand that cisco negotiate a licensing agreement with
Stanford for the use of this computer board. I believe that they have
already delivered products to customers that use it. I believe that the
design has substantial value -- on the order of $50,000.
The more I look, the messier it gets.
Les Earnest
∂02-Jul-86 2300 NILSSON@SU-SCORE.ARPA [Nils Nilsson <NILSSON@SU-SCORE.ARPA>: Matters of Concern]
Received: from SU-SCORE.ARPA by SU-AI.ARPA with TCP; 2 Jul 86 23:00:09 PDT
Date: Wed 2 Jul 86 22:59:26-PDT
From: Nils Nilsson <NILSSON@SU-SCORE.ARPA>
Subject: [Nils Nilsson <NILSSON@SU-SCORE.ARPA>: Matters of Concern]
To: les@SU-AI.ARPA
Message-ID: <12219645848.17.NILSSON@SU-SCORE.ARPA>
Here is the version I sent Len. Thanks for your help on this.
I made just a few changes here and there. -Nils
---------------
Mail-From: NILSSON created at 2-Jul-86 22:56:22
Date: Wed 2 Jul 86 22:56:22-PDT
From: Nils Nilsson <NILSSON@SU-SCORE.ARPA>
Subject: Matters of Concern
To: bosack@SU-SCORE.ARPA
cc: nilsson@SU-SCORE.ARPA
Message-ID: <12219645290.17.NILSSON@SU-SCORE.ARPA>
Len, There are a number of matters that we need to talk about as soon as
possible. Unfortunately you haven't been around when I have, and I am
going to be away until Wednesday, July 9. I did schedule a time on your
calendar at 10:00 a.m. on Wednesday, so I hope we will be able to talk
then.
The items that are causing me concern involve evidence of conflict
between your relationship with cisco and your job as mgr. of CSD-CF. I
don't think I know as much about these possible conflicts as I need to
know, and I would like to hear directly from you about them. If,
indeed, there are actual or potential conflicts of interest we need to
decide Wednesday how to resolve them.
Here are some specific problems that are worrisome:
1. There has been mounting evidence that your activities on behalf of
your company, cisco systems, have been intermingled with your Stanford
responsibilities and that you have not been as careful as you should
have been about avoiding conflicts of interest. Instead of disclosing
fully your outside activities to me in advance, you proceeded to
entangle yourself in a number of problems, some of which apparently go
back a number of months before I heard of these activities.
2. I have heard that about one month ago you were asked by Pat Devaney
to give me a written disclosure of your consulting activities, including
their nature and extent, and a statement about the extent to which
Stanford facilities may have been used in support of cisco or other
outside activities. You apparently promised to provide such a
disclosure promptly but I have not received it.
3. You have reportedly been careless about fulfilling commitments and
meeting other obligations. Tom Rindfleisch tells me that on June 17 he
sent you a request for a copy of the gateway software---to be used as a
basis of comparison with gateway software developed at Stanford. Les
Earnest tells me that on June 25 he asked you why you had not responded
to Tom's request. You reportedly said that you would do so immediately
but that you would not provide machine-readable code, only hard copy.
Tom sent you a reminder later that day pointing out that he would soon
be leaving for two weeks and needed a prompt answer to his request.
Still he received no reply even after I sent a message to you both
emphasizing that I expected a reply to Tom's request.
4. Les Earnest says that he reminded you several times (on my behalf)
about providing the disclosure of your consulting activities and that
each time you promised to do it promptly, but so far I do not have such a
disclosure. For example, on June 25 you reportedly promised to complete
it by the end of the next day but didn't. On June 30, in your most
recent meeting with Pat Devaney, you apparently agreed to complete it
not later than July 1. Where is the disclosure?
5. In the course of the earlier meeting with Devaney you were reported
to have disclosed that you are consulting for cisco about 20 hours per
week. You apparently were not aware that this is well beyond the
maximum permitted under Stanford policies. Les Earnest says that you
subsequently remarked to him that you were going to reduce your Stanford
appointment to less than full time so as to accommodate your outside
responsibilities. I have received no such proposal from you and,
frankly, believe that it would be impractical for you to fulfill the
responsibilities of your position with less than a full time effort.
6. It appears that you have, in fact, reduced your level of effort on
Stanford responsibilities. While we do not have fixed working hours,
all of our full-time people are expected to be here on the order of 40
hours per week, and much of that time must be in normal daytime hours so
that you can interact with other staff members. In fact, I seldom find
you here whenever I come looking. Others report similar experiences.
It appears that you are spending much too much of your time somewhere
else.
7. Finally, and most important, I am concerned whether you are really
fulfilling your responsibilities as head of the Computer Facilities
Group. For example, financial planning and control seems to be
inadequate--I understand that the computer cost centers are running at a
substantial deficit that must be made up by the end of the fiscal year
in August. This will likely require a substantial rate change that will
cause budgeting problems for research projects. This problem should
have been foreseen earlier and fixed with the smaller rate changes that
would have then been possible. There also seems to be some distress
among your staff over both a lack of direction and, in some cases, a
sense that they are being drawn into what they perceive as your
conflicts of interest.
I would like to think that each of these problems can be addressed and
resolved satisfactorily. But such a resolution may require a choice on
your part either to substantially disengage yourself from outside
consulting and pay much more attention to your responsibilities here or
to leave Stanford and proceed with the development of your company. I
would like you to consider these points and discuss them with me next
Wednesday.
-Nils
-------
-------
∂09-Jul-86 1526 LES Devaney message
To: Nilsson@SU-SCORE.ARPA
Here, for your information, is what Pat Devaney asked for a week ago.
As you can see, it is not very explicit about disclosing past uses of
Stanford facilities or intellectual property.
I talked to Pat just now and told her what is happening. She confirmed
my impression that a leave of absence is inappropriate in this case.
I have since read Guide Memo 22.7, paragraph 2.b.(2) and conclude that
she is correct.
Les
------------------------------------
∂10-Jul-86 1940 LES Computer Masks taken by cisco
To: cs.eab@SU-FORSYTHE.ARPA
This note describes some issues to be resolved between Stanford, cisco
Systems/Len Bosack and Sun Microsystems. It now appears that there are
two more thefts of property to worry about.
1. A computer design that belongs to Sun Microsystems and is probably
protected both by copyright and trade secret laws was taken without
authorization from a protected computer file at Stanford.
2. An improved version of this computer design, developed by Stanford staff
members based on the Sun design and using Stanford facilities exclusively
is being used at Stanford but also has been taken by cisco Systems and
is being sold by them. The board masks for this computer have been removed
from the Stanford campus, so we are not in a position to make more when
we need them.
The background of this case is a bit complicated.
During 1980-81, the so-called "SUN" (for Stanford University Network)
computer board was developed here. It used the then-new Motorola 68000
computer chip and 64k RAM chips. The principal designer was a student
named Andy Bechtolsheim, but many people contributed to the design,
including me. Andy asked in advance of undertaking this project whether
he would be able to develop it commercially and was told (by me) that he
could.
Many copies of the original SUN board were made at Stanford and it was
used both as the central element in a family of workstations assembled here
and as the active element in a number of gateways used in the burgeoning
Stanford local area networks, based on ethernet technology.
Andy then went to OTL and asked whether a license was required in order
for him to develop the board commercially. I was not a party to that
negotiation, but I understand that he was told that he was free to pursue
it without a license, given that there was apparently nothing patentable
in the device.
Andy found that a number of companies were interested in using the SUN
board and signed a half-dozen or more licensing agreements. He provided
each licensee with circuit board artwork and documentation (both developed
and produced at Stanford, of course) and collected both up-front fees and
royalties.
I left Stanford about that time and my company, Imagen, became his first
licensee. Imagen paid Andy $10,000 in royalties; some later licensees
paid as much as $40,000 each. Andy provided the same artwork and
documentation to Stanford at no cost and these boards continued to be
fabricated at Stanford for some time.
(I also negotiated on behalf of Imagen a separate licensing agreement with
Stanford for an invention that my partner had made as a Stanford staff
member. This subsequently yielded Stanford something over $400,000 in
equipment discounts and royalty payments.)
A little later, Andy left Stanford to become a founder of Sun
Microsystems, without finishing his PhD, alas. He further refined the
design of the SUN board, still using the SUDS design program on the SAIL
computer at Stanford, and ceased licensing either the older design or the
new one. The improved computer board had four layers rather than the two
that had been used in the version that was licensed to various companies.
Sun sold many of the improved computer boards to Stanford on a heavily
discounted basis. (In the four years since then, Sun has become an
enormous commercial success -- I understand that Andy's shares now have a
market value of about $25 million.)
Early this year, the CSD facilities group headed by Len Bosack found that
they needed some more Sun-like computer boards for building gateways.
Sun had stopped selling such boards in the small quantites that Stanford
needed. Len and others figured out that a relatively minor change in the
old SUN board would permit it to use the newer and more economical 256k
RAM chips. They experimentally rewired a board here at Stanford, verified
that it worked, then developed a printed-circuit card version and
fabricated 10 of the improved boards for Stanford.
There was one funny thing about the fabrication project. While the parts
were paid for by Stanford, the new mask for the printed circuit card disappeared
from campus. It also appears that some number of these computers were
concurrently fabricated for cisco, possibly as part of the same
transactions, at cisco's expense. The entire project was managed by Bosack,
but other Stanford staff members participated in various ways, including
prototype development and testing.
Having learned of this situation from a staff member, on July 10 I asked
Len Bosack for his views on what had happened and who he thought owned the
computer board. He said he believed that it belonged to cisco because he
had done the actual design work, which was not a part of his normal Stanford
job responsibilities. I asked him how the design work had been done.
He said that it had been done using the SUDS design program on the SAIL
computer and that it was developed as a modification to the original
SUN board. I pointed out to him that cisco does not and never has had a
computer account on SAIL, so the computer time for that project must have
been paid for by Stanford.
Bosack agreed and also said that he thought Stanford should have the right
to make copies of this board for its purposes, but that he believed that
cisco owns the commercial rights. I invited further explanations for
how cisco came to own the rights to this computer. He said that it was
"essentially the same as the original Sun board case" -- if it was alright
for Andy Bechtolsheim to take his design and sell it, then it was alright
for Len to do it.
I knew that the original SUN computer board was a two-layer board whereas
the one he had developed was a four-layer board, like the one that Andy
Bechtosheim had developed for Sun Microsystems but had never licensed
outside the company. I asked Len how he had managed to get a four-layer
version. He said that Andy Bechtolsheim had provided the source files
for his design.
I knew that Andy had developed this board at SAIL, so I asked Len how the
transfer had taken place. Len said that Andy placed a copy of the software
in a place where he could get it. I asked whether in his negotiation with
Andy, he said that he wanted the design for use by Stanford or whether he
had disclosed his intention to use it to develop a cisco product. Len
said that Andy knew it was for use by cisco.
This afternoon, I called Andy Bechtolsheim and asked for his recollection
of the negotiation with Bosack and whether Len approached him as a
representative of Stanford or of cisco. Andy said that there had been no
such negotiation and that he had never authorized the use of the design.
Andy speculated that it had beed copied from his protected area. We both
knew that there was no way to protect such files from access by a system
administrator, or any other programmer with substantial knowledge of the
system.
Andy also said that as far as he is concerned, the four-layer board design
is the property of Sun Microsystems.
Conclusions
It appears that the basic design of a number of computers now in use at
Stanford, cisco Systems and some of cisco's customers was taken by Len
Bosack without authorization from a protected computer file at Stanford
that belonged to Sun Microsystems, then was modified by him and other
Stanford staff members under Bosack's management using Stanford facilities
exclusively. Furthermore the other staff members apparently were not
aware of where the basic design came from or that the improved version was
to be a cisco product.
I note also that we do not have a copy of the printed circuit card masks
for the improved computer; they are reportedly in the hands of cisco.
I believe that both the apparent theft of the Sun design and the removal
of the improved design from Stanford represent thefts of intellectual
property with substantial value. I suspect that we can negotiate
something with Sun Microsystems that will clean up the first theft.
I also believe that cisco should not be able to use or sell the modified
computer without at least a license from Stanford and, most likely, Sun
Microsystems, depending on what they have to say about what happened.
Let me point out that the mass bus-ethernet board that was sold to GSB by
cisco involves similar issues, but in a less complicated way. It was
developed exclusively at Stanford for Stanford use (principally by Len
Bosack) and in that case too, the PC masks were taken away by cisco and
the product has been sold to various customers, including Stanford.
I believe that this should not have happened except with a royalty-bearing
license from Stanford. My understanding is that someone at Stanford
has decided differently. This makes no sense to me.
The more I look, the messier it gets.
Les Earnest
∂10-Jul-86 2043 LES Computer Design Thefts
To: hk.pld@SU-FORSYTHE.ARPA
[Here is a message that I just sent to Elizabeth Batson. It is a revision
of the one I sent you earlier, reflecting additional information obtained
from conversations with Len Bosack and Andy Bechtolsheim of Sun Microsystems.
I regret to report that this case now apparently involves both theft of
commercial trade secrets and deliberate lies in an attempted coverup.
It gets uglier and uglier.
As I mentioned earlier, I have a tape containing the gateway software
developed by Bill Yeager and corresponding documentation. I plan to turn
it over to Elisabeth Batson.
Late this afternoon I heard from Steve Hansen, who said that he had been given
a copy of the gateway software by Kirk Lougheed, as requested. He also said
that he had made a dump tape from Kirk's file area a little earlier as a
precaution. When he compared them, he found that the files Kirk had given
him were from a much earlier time period; in other words, they represented
a much earlier stage of development than Lougheed actually carried out at
Stanford. When he subsequently rechecked Kirk's file area, he found that
all the newer versions of the files had been removed.
It appears, then, that a systematic attempt at evasion is underway.
-Les]
------------------------
∂11-Jul-86 1507 NILSSON@SU-SCORE.ARPA CSD-CF Change
Received: from SU-SCORE.ARPA by SU-AI.ARPA with TCP; 11 Jul 86 15:07:05 PDT
Date: Fri 11 Jul 86 15:00:21-PDT
From: Nils Nilsson <NILSSON@SU-SCORE.ARPA>
Subject: CSD-CF Change
To: faculty@SU-SCORE.ARPA, staff@SU-SCORE.ARPA, csd@SU-SCORE.ARPA
Message-ID: <12221917930.50.NILSSON@SU-SCORE.ARPA>
Len Bosack has informed me that he would like to leave Stanford (at
least temporarily) to spend more time with a company he helped
start--cisco systems. He will be leaving officially at the close of
work on Wednesday, July 15, 1986. I have asked Les Earnest to take on
the additional responsibility of Acting Director, CSD-CF effective
immediately and he has agreed. I want Les to start functioning in this
position even before Len leaves in order to effect a smooth transition.
Len has graciously agreed to be available for consultation, etc. during
the transition and beyond.
I want to thank Len especially for his hard work in making CSD-CF
an effective and efficient organization. We will miss him, and I'm
sure we all want to wish him success in his new venture.
We will soon post a job announcement in order to find a permanent
director for CSD-CF. Suggestions about whom we might contact about
this position are welcome. Please direct these suggestions to
Les Earnest.
-Nils
-------
∂01-Aug-86 0955 HANSEN@Sierra.Stanford.EDU Re: Sources
Received: from SIERRA.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 1 Aug 86 09:54:55 PDT
Date: Fri 1 Aug 86 09:55:50-PDT
From: Stephen Hansen <Hansen@Sierra.Stanford.EDU>
Subject: Re: Sources
To: Yeager@SUMEX-AIM.Stanford.EDU
cc: Hansen@Sierra.Stanford.EDU, tom@Score.Stanford.EDU, les@Sail.Stanford.EDU,
g.yundt@Score.Stanford.EDU
In-Reply-To: <12227359084.26.YEAGER@SUMEX-AIM.ARPA>
Message-ID: <12227367519.26.HANSEN@Sierra.Stanford.EDU>
Bill,
I have some good news. While trying to determine just when the
Helens and Mojave archive tapes were stolen, we discovered that on
July 6 the operator had mistakenly dumped the /ps file system, where
Kirk had his files, on the /class file system tape. The information
on this tape, when combined with the tape that I made on June 27 gives
us the most recent version of the EE-CF Tip/Gateway software sources.
I talked with Adrian Arima of the University's General Counsul's
Office about your concerns with the copyrights. He affirmed what I
had been told before, that since the work was done as part of Kirk's
job, it is considered 'work for hire' and Stanford holds all rights
to the results. The fact that the work was done at Stanford, using
Stanford equipment, only strengthens the claim.
We are fixing up the copyright notices in the files and as soon
as it is done I'll make a copy available to you.
Stephen Hansen
-------
∂05-Aug-86 1129 LES Pseudo-Sun boards
To: Tom@SCORE.STANFORD.EDU
Proto Engineering confirms that they have the artwork for the computer
boards and can make more when we want. They also say that cisco had a
batch of 30 made in a separate run at a later time.
.require "memo.pub" source_file;
.csm 1986 August 8,"Adrian Arima, Staff Counsel","Les Earnest, CSD Assoc. Chair."
.cc "Nils Nilsson (CSD Chair.), Donn Martin (Internal Audit), Tom Dienstbier"
.su "Len Bosack and Cisco Systems"
.fi
.cb CONFIDENTIAL
This note reports recent results of my inquiries into certain transactions
involving Len Bosack and his company, Cisco Systems.
.cb Modified Sun Computer Boards
I now believe that there was not, in fact, any misappropriation of Sun
Microsystems' proprietary information by Stanford.
As you know, the account of Len Bosack on how he acquired the Sun-1
computer board design did not mesh with Andy Bechtolsheim's statement that he
had not made the design files available.
After two more rounds of telephone interviews with each of them, I conclude that
both were distorting the truth and that Bosack's account was mostly
correct.
According to our computer records, on November 9 at 11:57 AM
someone running on the SAIL computer as SMI
(Sun Microsystems Inc., Andy Bechtolsheim's account) copied five files
called P1.DRW, P2.DRW, P3.DRW, P4.DRW and P5.DRW into file area [1,3],
which is the public file area in which system programs are stored.
About three minutes later, these same files were copied into protected
file area [PS,LB] by someone logged in as LB (Len Bosack).
According to both Andy Bechtolsheim and Len Bosack, these files describe
the Sun-1 computer design.
They are still resting in file area [1,3], which is publicly accessible.
Len says that this transaction was the result of a chance meeting
with Andy in the Display Dungeon of Jacks Hall. Len says that Andy
offered the files as an aid to developing an improved version.
Len had also stated earlier that Andy made these files available to him
as a representative of Cisco Systems rather than as a Stanford Employee.
Andy earlier stated than no such transaction took place and that he
did not know of Cisco Systems' existence.
Last night a bit after midnight I found Andy running on SAIL and
reviewed with him Len's specific recollections and the date and time
information from our computer records.
I then asked him if such an incident could have occurred.
This time he said that it could have, but that any such disclosure
would have been covered by the nondisclosure agreement that Tom Dienstbier
had signed earlier.
I already knew of this nondisclosure agreement (copy enclosed) and knew
that it covered only Sun 2/50 equipment and that it terefore did not apply
to the Sun-1 board.
I did not discuss this point with Andy, however.
My impression from this exchange was that Andy had not been forthcoming
earlier because he was worried that his action in making the Sun-1
design available to Stanford was probably improper, so
he did not confirm that it had happened until after he had invented a
rationalization for his action.
Though we will probably never know for sure what Andy and Len said to
each other on November 9, 1985, I believe Andy's statement that there
was no discussion of Cisco at that time.
Len was not being open about his business activities with others
at Stanford then, so it would have been surprising for him to
have mentioned it to Andy at a time when he was trying to get him to
make the Sun-1 design files available.
As I mentioned on the phone, we reviewed our records and found that
10 modified Sun boards were made for us by Proto Engineering
(telephone 408 738-0693).
I called them, talked to their sales person (Bob) and learned that they
had copies of the board artwork on file and will make more when we want.
He also stated that Cisco had them make 30 copies of this board in a later batch.
Following up on your recommendation that we file for copyright on the
board, Tom Dienstbier contacted Proto Engineering and picked up
the artwork and drillboard.
They were on file there as belonging to Cisco rather than Stanford
and also carry a Cisco copyright notice.
Tom noted that the first batch of boards that were made for Stanford also
carried a Cisco copyright notice, giving the year as 1985, whereas the
artwork we retrieved carries a 1986 copyright notice.
It is likely that the only difference is the copyright notice, though
we would have to do a careful comparison to be sure.
In any case, the later version also appears to have been produced using the SUDS
design system on the SAIL computer.
We plan to remove the Cisco copyright notice and place a Stanford notice
on the artwork, preparatory to filing for copyright.
Incidentally, I believe that Len Bosack still has the original artwork
for the modified Sun board, so our taking the copies from Proto Engineering
will not prevent him from making more.
He probably would incur a retooling charge for making a new drill plate,
however.
.cb MEIS Boards
Tom Dienstbier says that we still have MEIS board artwork, but he believes
it carries no copyright notice.
We will plan to apply one in preparation for filing.
Tom also informs me that the principal contributors to the software that
runs the MEIS board were David Eppstein (a former student), David Todd (a visiting
faculty member), Kirk Lougheed and Len Bosack.
We have not located the source files yet, but they are believe to reside
on the SCORE computer as well as some other machines at Stanford.
.cb 3 Megabit Ethernet Boards
I earlier reported that Len Bosack had arranged to fabricate and
sell 35 Ethernet boards
to Cisco late last year at a price of $425 each, for a total of $14,875,
but that Cisco still had not paid Stanford in the intervening 9 months or
so.
We also knew that these boards had been resold to Xerox PARC.
After going through a six-level search at PARC, on August 7 I finally found
a person who confirmed the terms of the sale: Norb Beyer at 415 494-4070.
He said that they had purchased 35 boards at $800 each, or $28,000 total.
I asked for copies of the purchase order and cancelled check.
He wanted to know what was going on, so I outlined the situation.
He then agreed to forward copies, which should arrive shortly.
I have asked Donn Martin to review the cost to Stanford of fabricating the
ethernet boards.
.cb Other Matters
When Donn Martin requested access to the residual computer files of Len Bosack,
Kirk Lougheed and Sandy Lerner, I expressed concern about privacy
issues both to Donn and to you and asked you to review the
appropriateness of such access.
Today you replied that such files are supposedly for official
Stanford use only and that they were therefore subject to audit
and other use in support of Stanford operations.
I then invited Donn Martin to review the contents of these files.
Responding to your earlier requests, enclosed please find a copy of the
Personnel Requisition for the CSD-CF Director's position that led to Len
Bosack's selection for that position and also Len's signed SU-18 Patent Agreement
form.
.skip 2
.indent 0,0; nofill
Enclosures:
Sun Microsystems nondisclosure form signed by Tom Dienstbier
Personnel Requisition for University Manager III
Stanford SU-18 form signed by Len Bosack
∂13-Aug-86 2255 HK.AXA@Forsythe.Stanford.Edu An experiment in electronic messaging
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 13 Aug 86 22:55:17 PDT
Received: by lindy.stanford.edu with Sendmail; Wed, 13 Aug 86 22:55:34 pdt
Date: Wed, 13 Aug 86 22:57:35 PDT
From: Adrian Arima <HK.AXA@Forsythe.Stanford.Edu>
To: LES@SAIL.STANFORD.EDU
Subject: An experiment in electronic messaging
Les -- What follows are a lot of lines for a simple message. Hope
this one makes it through. aa
*********************************************************************
22:40:10 08/13/86 FROM GG.JDN "Jim Nisbet":
Adrian (HK.AXA),
This message was returned by the MIT "AI" host because the userid
was unknown. For some reason it was returned to "general delivery"
at Forsythe, so it came to me.
I'm guessing that you probably meant to send the mail to the
Stanford AI lab. Why don't you try "LES@SAIL"; according the SAIL
directory that is Les Earnest.
/j
(Forsythe Postmaster)
--------
From daemon Wed Aug 13 17:17:34 1986
Received: by lindy.stanford.edu with Sendmail; Wed, 13 Aug 86 17:17:31 pdt
Itsmail-To: hk.axa@forsythe
Itsmail-Date: Wed, 13 Aug 86 17:16:54 pdt
Date: Wed, 13 Aug 86 19:30:13 EDT
From: Communications Satellite <COMSAT@AI.AI.MIT.EDU>
Subject: Msg of Wednesday, 13 August 1986 19:30-EDT
To: "HK.AXA@Forsythe.Stanford.Edu"@LINDY.STANFORD.EDU
Message-Id: <[AI.AI.MIT.EDU].83234.860813>
============ A copy of your message is being returned, because: ============
"LES@AI.ARPA" at AI.AI.MIT.EDU is an unknown recipient.
============ Failed message follows: ============
Received: from lindy.stanford.edu by AI.AI.MIT.EDU 13 Aug 86 19:05:18 EDT
Received: by lindy.stanford.edu with Sendmail; Wed, 13 Aug 86 16:04:31 pdt
Date: Wed, 13 Aug 86 16:06:37 PDT
From: Adrian Arima <HK.AXA@Forsythe.Stanford.Edu>
To: LES@AI.ARPA
Les -- Please let me know if this message gets through to you
(I'm not sure of your electronic mail address).
Anyway, the AEA convention was at end of June or early July.
aa
To: LES@SAIL
∂14-Aug-86 1629 LES re: An experiment in electronic messaging
To: HK.AXA@FORSYTHE.STANFORD.EDU
[In reply to message sent Wed, 13 Aug 86 22:57:35 PDT.]
Adrian:
Yup, it worked.
I trust that you have received the documentation from Xerox PARC by now.
I was puzzled by the dates on the invoice and check. The February billing
date indicates that the fabrication of the ethernet boards must have
started in late 1984. I had been told earlier by Lynn Gotelli that she
billed Cisco for the first time in November 1985, some nine months after
Bosack had taken the boards and delivered them to Xerox.
I went to Lynn this morning and asked why Cisco had not been billed earlier.
She said that Len had told her to delay the billing.
This afternoon I received a call from Jeffery Blatt (213 550-8282), who is
an attorney for Sun Microsystems and who was responding to my earlier
conversation with Andy Bechtolsheim. I reviewed the general situation with
him and aimed him at you.
Blatt remarked that one element of the Sun-1 computer board is a memory
management unit that has been patented. He is offering us a license to
use the design at no cost, which I guess would be nice to have. Blatt
plans to send a threatening letter to Bosack but will discuss tactics with
you before acting.
Cheers,
Les
∂22-Aug-86 1736 HK.AXA@Forsythe.Stanford.Edu Copyrighting Tip/Gateway software
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 22 Aug 86 17:36:22 PDT
Received: by lindy.stanford.edu with Sendmail; Fri, 22 Aug 86 17:36:07 pdt
Date: Fri, 22 Aug 86 17:36:44 PDT
From: Adrian Arima <HK.AXA@Forsythe.Stanford.Edu>
To: LES@SAIL.STANFORD.EDU
Subject: Copyrighting Tip/Gateway software
Steve/Bill/Les/Iris -- Hard copy to follow. aa
* * * * * * * * *
MEMO TO STEVE HANSEN
FROM ADRIAN ARIMA
DATE: AUGUST 22, 1986
RE: COPYRIGHTING TIP/GATEWAY & SUMEX SOFTWARE
Attached is a blank official copyright form TX, and a copy of the
form with entries written in. In order to fill out the forms
properly, we need to know (numbers keyed to item numbers on the
TX form):
1. What is the name of the software?
What was the name of any previous version? (E.g., was the
ethertip/gateway software considered just a new version of
the SUMEX gateway?
3. What was the year that the version of the source code being
registered was completed? (E.g., was version 4.400 completed in
1985?)
I mentioned that the distribution to the other universities
might be deemed a "publication". If there is documentation
or other facts indicating that they got the software under
restrictions against further distribution, then the software
could be registered as unpublished. It would be useful if
the networking group would discuss this and come up with a
summary and any documents that are relevant. Otherwise,
please give the best guess of the date the software was
first distributed outside Stanford?
5. I assume there was no prior copyright registration of any
previous versions of this software.
6. Was the code based on any pre-existing other software?
(E.g., from other universities, from DEC, etc.dD)
(Note that for works made for hire, the employer is the "author",
not the employee.)
COPIES FOR DEPOSIT
The copies of the manuals you sent are appropriate for copyright
registration. (Please provide the information requested above
for the manuals. Also, it's more likely that the manuals have
been published if they have been sent to outsiders or if
programmers/students have been allowed to take copies with them
when they left Stanford.)
I would like to register copyrights in the following versions of
the software:
1. The original SUMEX code as transfered by Bill Yeager to EE.
2. Current version of SUMEX gateware.
3. Version 4.400 of the TIP gateware.
4. Version 5.2 of the TIP gateware, with Stanford copyright
notices only (no cisco notices) and with references to cisco
cleaned out.
I need 2 copies of printouts of the first 50 pages and the last
50 pages of each version the software. (The law only requires
deposit of the first 25 and the last 25 pages, but I'd like to
deposit more to avoid any arguments.)
I would appreciate your asking for this information from the
other networking participants on Monday. Again, the information
gathered should be considered confidential lawyer/client
communications.
cc w/ attachments:
Bill Yeager
Les Earnest
Les: Please provide the info for the MEIS software and
documentation. Does the packet you sent me include the
software as well? It looks like it.
Iris Brest
To: HANSEN@TALBOTS, LES@SAIL, YEAGER@SUMEX, HK.IXB
cc: GD.WHY
∂26-Aug-86 1258 HK.AXA@Forsythe.Stanford.Edu Cisco stuff
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 26 Aug 86 12:48:06 PDT
Received: by lindy.stanford.edu with Sendmail; Tue, 26 Aug 86 12:46:44 pdt
Date: Tue, 26 Aug 86 12:49:08 PDT
From: Adrian Arima <HK.AXA@Forsythe.Stanford.Edu>
To: LES@SAIL.STANFORD.EDU
Subject: Cisco stuff
Since I will be out of town for several weeks, please direct
your messages on the cisco matter to Iris Brest (HK.IXB).
aa
To: LES@SAIL, HANSEN@SIERRA, YEAGER@SUMEX, GD.WHY
cc: HK.IXB
Return-Path: <hedrick@topaz.rutgers.edu>
Received: from topaz.rutgers.edu by SUMEX-AIM.ARPA with TCP; Tue 23 Sep 86 18:13:43-PDT
Received: by topaz.rutgers.edu; Tue, 23 Sep 86 21:11:29 edt
Date: Tue, 23 Sep 86 21:11:29 edt
From: hedrick@topaz.rutgers.edu (Charles Hedrick)
Message-Id: <8609240111.AA11987@topaz.rutgers.edu>
To: Yeager@sumex-aim.arpa
Subject: Re: Gateway changes.
Thanks. Next time we do work on that gateway, I'll look into
integrating it. I assume that my version will work as well in our
situation, since we have only IP. You might be interested that we now
have a bunch of the Cisco boxes. We had some problems getting them
going. But they fixed them quite expeditiously, and they are now
working fine. I haven't had a chance to do any performance testing
yet. We are moving our Arpanet gateway to the Cisco gateway this
week. Our 11/23 gateway just can't cope. Proteon made us a very
generous offer to replace it with one of their new 68000 things. They
would accept the 11/23 as a tradein and charge only $3500 for a 68000
gateway with 1822 interface. Unfortunately, after the problems we
have had in getting software support from Proteon with the 68000 code,
I decided to go with people that I felt I could trust. This is
probably unfair to Proteon. Their 68000 gateway is a serious product,
whereas the 11/23 code was never very widely used. But I'm just not
willing to risk more problems with our Arpanet connection.
-------
-------
∂29-Sep-86 1640 TOM@SU-SCORE.ARPA [Bill <Yeager@SUMEX-AIM.ARPA>: Re: 68k CPU]
Received: from SCORE.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 29 Sep 86 16:40:25 PDT
Date: Mon 29 Sep 86 16:39:38-PDT
From: Thomas Dienstbier <TOM@SU-SCORE.ARPA>
Subject: [Bill <Yeager@SUMEX-AIM.ARPA>: Re: 68k CPU]
To: les@SU-AI.ARPA
Message-ID: <12242907525.12.TOM@Score.Stanford.EDU>
Well here's the reply..
tom
---------------
Return-Path: <YEAGER@SUMEX-AIM.ARPA>
Received: from SUMEX-AIM.ARPA by SU-SCORE.ARPA with TCP; Mon 29 Sep 86 16:36:09-PDT
Date: Mon 29 Sep 86 16:30:12-PDT
From: Bill <Yeager@SUMEX-AIM.ARPA>
Subject: Re: 68k CPU
To: TOM@SU-SCORE.ARPA
In-Reply-To: <12242871149.12.TOM@Score.Stanford.EDU>
Message-ID: <12242905808.35.YEAGER@SUMEX-AIM.ARPA>
I have NEVER seen any copyright on any of this stuff. We have shipped
copies of our PROMs to Rutgers, and they may in fact have copied them
and sent them elsewhere. I believe we also have sent copies to U. of
Chicago.
We have always been rather free with information in the spirit of
proliferating good work!
Bill
-------
-------
∂05-Nov-86 2005 LES re: Phone call
To: Yeager@SUMEX-AIM.STANFORD.EDU
[In reply to message sent Wed 5 Nov 86 19:44:33-PST.]
I have delivered the kick that you request and we are starting over with
a new lawyer late Friday afternoon.
Given that I need to get an ethernet-to-ethernet link via 56 kbaud modem
between MJH and Lucid in Menlo Park right now and given that there was no
plausible basis for business dealings between Stanford and Cisco, I asked
our subcontractor (Lucid) if they could make a deal with Cisco. Len
agreed to sell them the necessary software additions for $3k, so I
then set about to get governmenatal approval.
In separate discussions with Tom D., Len also agreed to return the
multibus communication cards that belong to Stanford that somehow migrated
to Cisco. Len then became unreachable and later renegged on both offers,
saying that he wouldn't do anything to help Stanford. So, alas, we seem
to be heading for hardball.
I am now seeking alternative ways of making the Lucid link work. Bridge
says that they will sell us bridge hardware and software that will do the
job for $17k (both ends). For a week I have been trying to find out from
Proteon whether they have a solution, but I am getting no response.
I called you to ask how much work would be involved in hacking the existing
network code to do the job, given that we already apparently have adequate
hardware to do the job. I would also appreciate any ideas you might have
on other potential sources. I have asked Tom D. to poke DEC.
Les
∂12-Nov-86 1914 LES Bosack Visit
To: HK.AXA@FORSYTHE.STANFORD.EDU, HK.PLD@FORSYTHE.STANFORD.EDU,
Nilsson@SCORE.STANFORD.EDU, GD.WHY@FORSYTHE.STANFORD.EDU
Len Bosack finally returned my telephone call today and agreed to bring
back the communications boards together with the documentation on how they
work. He brought the boards in a few minutes ago without the
documentation, saying that he couldn't find it quickly but would bring it
over soon.
We then discussed the outstanding issues briefly and I asked him why he
was not seeking licenses from Stanford. After the usual extensive
thrashing, he finally agreed that this was a good idea and that he would
get in touch with Niels Reimers in the next day or two. I will contact
Niels to prepare him.
I stated that the amount paid by Cisco on the ethernet transaction was
insufficient. He showed some willingness to negotiate on this issue,
while emitting the usual smokescreen. I said that I would try to bring
forth an appropriate invoice from Stanford.
In summary, we seem to be moving slightly. I'll do my best to keep it
rolling.
Les Earnest
∂14-Nov-86 1317 HK.PLD@forsythe.stanford.edu Bosack Visit
Received: from FORSYTHE.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 14 Nov 86 13:17:31 PST
Received: by lindy.STANFORD.EDU; Fri, 14 Nov 86 13:12:20 PST
Date: Fri, 14 Nov 86 13:14:47 PST
From: Patricia L. Devaney <HK.PLD@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: Bosack Visit
REPLY TO 11/12/86 19:22 FROM LES@SAIL.STANFORD.EDU: Bosack Visit
Les, does the outcome of your meeting with Len alter our
plans re a written response to him and his lawyers? I don't
think we should be bringing in OTL on this until we have
agreement on who owns what and until Len has returned/
reimbursed STanford for what he owes the University.
I suggest that you draft a suitable letter to him from
bob street, as agreed at our meeting last week, but you
may want to change it somewhat from our envisioned response
to capture the essence of your recent dealings with Len.
Then, let's discuss with Nils and the lawyers whether
we still want to send a letter to his lawyer. What
do you think? Pat
To: LES@SAIL.STANFORD.EDU
cc: NILLSON@SAIL, HK.AXA, Williams(HK.JXW), GD.WHY
∂14-Nov-86 1445 LES re: Bosack Visit
To: HK.PLD@FORSYTHE.STANFORD.EDU
CC: NILSSON@SCORE.STANFORD.EDU, HK.AXA@FORSYTHE.STANFORD.EDU,
HK.JXW@FORSYTHE.STANFORD.EDU, GD.WHY@FORSYTHE.STANFORD.EDU
[In reply to message sent Fri, 14 Nov 86 13:14:47 PST.]
If Cisco is willing to license all of the technology under discussion, as
appears to be the case, then they are acknowledging Stanford ownership.
Of course, it remains to be seen if they will bargain in good faith, but
if they do then I seen no point in continuing the discussion of those
items outside of the licensing negotiation.
My view is that our goal is to negotiate a suitable licensing agreement
and the more directly we approach that goal the sooner we are likely to
reach it. I agree that an additional letter will be needed at least for
the payment of the ethernet board bill and that should probably be signed
by Bob Street. I am working on getting a suitable invoice put together
and will draft a letter.
Incidentally, Len did call Niels Reimers yesterday. I have scheduled an
appointment with Niels and Lisa Kuutila, who will handle the negotiation,
at their office next Tuesday morning at 10:00am. If anyone else would
care to help fortify Lisa, please come then or negotiate a different time.
Les
∂19-Nov-86 2018 GD.WHY@forsythe.stanford.edu re: Bosack Visit
Received: from FORSYTHE.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 19 Nov 86 20:18:47 PST
Received: by lindy.STANFORD.EDU; Wed, 19 Nov 86 20:12:58 PST
Date: Wed, 19 Nov 86 20:15:55 PST
From: Bill Yundt <GD.WHY@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: re: Bosack Visit
REPLY TO 11/14/86 15:10 FROM LES@SAIL.STANFORD.EDU: re: Bosack Visit
Les...hope I did not offend as a result of my quasi-rotten
mood at meeting with Technology Licensing re Bosack. I was
being harrassed from multiple quarters that morning so was
feeling barely civil. In any case, I think Lisa seems
fairly tough and knowledgable so am glad we have her working
on the matter.
Appropos the last note from Devaney and your revelation to
me about the agreement regarding the letter(s)
....particularly the one from our Lawyers. I think we
should go back to them and get them to send a letter
notifying Len's attorneys that we have a rights claim and
conisder it important that it be worked out promptly. Two
reasons....it might bring some pressure from that quarter
and it informs them officially in case Len's game is buying
time while he gets venture money.
...Bill mod 10
To: LES@SAIL.STANFORD.EDU
∂20-Nov-86 0029 LES re: re: Bosack Visit
To: GD.WHY@FORSYTHE.STANFORD.EDU
[In reply to message sent Wed, 19 Nov 86 20:15:55 PST.]
No problem provided we get our tactics straight before serious
negotiations begin. As I left, Lisa said she thought she would have a
preliminary discussion with Len by herself. I said that sounded
reasonable to me. (She strikes me as a sensible person who will not
give away the store.)
Your point about formal notice to cisco's lawyers is well taken.
∂21-Nov-86 0021 LES Protection of Printed Circuit Designs
To: HK.AXA@FORSYTHE.STANFORD.EDU, HK.JXW@FORSYTHE.STANFORD.EDU
CC: HK.PLD@FORSYTHE.STANFORD.EDU, NILSSON@SCORE.STANFORD.EDU,
GD.WHY@FORSYTHE.STANFORD.EDU, HK.NJR@FORSYTHE.STANFORD.EDU,
cs.ejk@FORSYTHE.STANFORD.EDU
Bill Yundt and I showed up at OTL Tuesday morning and talked over some
of the cisco negotiation issues with Niels and Lisa. Assuming that
suitable licensing agreements can be negotiated and that cisco promptly
pays the full amount due Stanford from the infamous ethernet board "sale,"
there remains one important side issue: how to protect PC card designs
that are developed at Stanford. After wrestling with this issue for
awhile, I believe that there is a workable approach. This note outlines
it and invites comments.
We have assumed for many years that printed circuit designs are protected
by copyright and have generally placed copyright notices on the PC
artwork. This is also a standard industry practice. When I heard from
Adrian awhile back that PC designs are apparently not protected by
copyright, I was appalled. Inasmuch as we do not use trade secret law, it
appeared that we had no protection against misappropriation of designs,
such as happened in the cisco case.
I have since checked this question informally with several lawyer
acquaintenances and have received uniformly ambiguous responses.
(Of course, I was asking for free legal advice and you generally get what
you pay for).
Finally, I obtained copies of a bunch of government notices regarding
copyright, inluding the standard rejection letter that is sent to people
who apply for copyright on printed circuit cards. It clearly states that
copyright does not apply to the designs of useful devices such as
PC cards.
It appears to me, however, that the PC artwork itself can be copyrighted
and that this would provide adequate protection of Stanford's interests.
This belief seems to be confirmed by their rejection letter, which gives
as an example of something that CAN be copyrighted "a floral relief design
on silver flatware" as opposed to "the design of the shape of the flatware."
It appears to me that PC artwork is exactly analogous to a relief design
and therefore is eligible for copyright as a separate entity.
Bill Yundt says he thinks he has heard of this scheme being used in
industry. In any case, if it sounds plausible to you, I propose that we
try it out by filing for copyright on some artwork, such as the modified
Sun board.
Les
∂21-Nov-86 1230 CS.EJK@forsythe.stanford.edu
Received: from FORSYTHE.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 21 Nov 86 12:30:11 PST
Received: by lindy.STANFORD.EDU; Fri, 21 Nov 86 12:24:04 PST
Date: Fri, 21 Nov 86 12:30:15 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: les@sail.STANFORD.EDU
Pat:
As you have seen from the em from Les Earnest, we had a meeting
earlier this week to discuss the cisco situation.
My plan was to set up a meeting with Len to discuss in a
preliminary way what his proposal for a license is. After
that we will have a second meeting to propose our terms for a
license.
In discussing the situation with Adrian, he mentioned that he
was not sure if we wanted some "admission" from cisco that they
need a license and that in fact Stanford owns the boards
or that we will simply license whatever rights we have without
specifying exactly what they are.
What are your thoughts on this? I will wait to contact Len
until I hear back from you.
∂25-Nov-86 0818 CS.EJK@forsythe.stanford.edu
Received: from FORSYTHE.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 25 Nov 86 08:18:02 PST
Received: by lindy.STANFORD.EDU; Tue, 25 Nov 86 08:11:24 PST
Date: Tue, 25 Nov 86 08:18:43 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: les@sail.STANFORD.EDU
Les:
I got a message from Pat Devaney that she will talk to chairs of
EE and Computer Science prior to giving me the go-ahead to start
negotiating the license.
Lisa
∂27-Nov-86 0823 JJW Bridge and Cisco
To: LES@SAIL.STANFORD.EDU, Tom@SCORE.STANFORD.EDU
This message arrived just after I sent you the previous one. It looks like
Cisco is getting a good review.
Date: Thu, 27 Nov 86 06:58:45 est
From: hedrick@topaz.rutgers.edu (Charles Hedrick)
Message-Id: <8611271158.AA27820@topaz.rutgers.edu>
To: tcp-ip@sri-nic.arpa
Subject: terminal servers: minireview of Bridge CS/100 and Cisco ASM
A message that just appeared on this list gave a reasonably complete
description of the Encore terminal server. Since we have used Bridge
CS-100's extensively, and are now using cisco ASM's, I thought it
might be helpful to give a description of them as well. I hope the
following review isn't too long to read, but it seemed worth trying to
give a good feel for the products. There are a number of similarities
in what these two products do. Both of them implement telnet using
special-purpose software (i.e. they do not run Unix or a Unix-like
shell). The user interfaces look like a typical DEC command scanner:
keywords which can be abbreviated, you can type ? at any point to see
what is wanted there. (With Bridge you have to hit carriage return
after the ?. Cisco activates immediately.) A number of the keywords
are even similar: connect, resume, disconnect, and show (though the
things that show will show are different). In both cases, you can
have multiple sessions active. You switch by typing an escape
character to get back to the terminal server, and then resuming the
session that you want to go back to. Because they implement telnet
and not rlogin, they are not as optimized for use with Unix as it
sounds like Encore is. However they do implement telnet sync, so
things are not as bad as they might be. The big problem with terminal
servers is that ↑C ↑O ↑S and ↑Q tend to have very long delays unless
you do something special. The difficulty is that in general there are
large buffers at both ends, and several packets can be "in flight".
So after you type ↑S, ↑C or ↑O, you can still get several thousand
characters. Rlogin solves the problem by integrating control
character handling on the terminal server and the host. The host uses
TCP out of band messages to keep the terminal server apprised of what
mode the tty is supposed to be in. Thus the terminal server handles
↑S locally when appropriate. But when you are in Emacs, the host
tells it not to do ↑S, and so that character works as the search
command. ↑C and ↑O are handled by cooperation between the server and
host. These features give rlogin a big advantage over a naive telnet
implementation. However it turns out that if you are careful, it is
possible to do nearly as well using telnet. The telnet protocol
includes the same out of band features as rlogin for implementing ↑C
and ↑O. It's just that most hosts and telnets don't bother to
implement these features. (In particular, 4.2 telnet and telnetd do
not.) Both Bridge and Cisco implement them in the terminal server.
So one need only repair the telnetd on your host. We have done this
on our primary timesharing machines. (Pyramid 90x systems. Alas, our
code may not be easy to import, since our telnetd is in the kernel,
for performance reasons.) This leaves ↑S. It would be possible to
use telnet negotations to turn on and off local ↑S handling.
Unfortunately, no one seems to have defined such a negotiation. Thus
we simply pick a character that is not used by Emacs for anything very
important (by default ↑\ is used), and set up our terminal servers to
use that as a local XOFF. The same character is set as XON. I.e. it
is a toggle. [An editorial comment: Why did Berkeley define a new
proocol, rlogin, rather than simply implementing telnet fully, and
adding a negotiation to handle toggling XOFF.]
Enough of generic descriptions. Now down to details. I'll start with
Bridge, because that is what we have had for the longest. Bridge has
at least 3 different terminal servers: CS-1, CS-100, and CS-200. They
have different numbers of ports: 32 for the CS-1 (I have heard rumors
that they may allow more now), 14 for the CS-100, and some smaller
number for the CS-200. The CS-100 is the only one that I know, though
I think the CS-200 might be more attractive for a new installation.
The CS-100 uses several 68000's to get enough bandwidth to drive all
of the lines at full speed. It boots from a floppy (although it is
also possible to get diskless machines, which boot over the network
from a special server CS-100. This would be a nice idea for any
installation that intends to have a number of boxes. The server can
also be used to help monitor the network, and to diagnose problems.)
Major networking parameters are set via a sysgen, which must be done
standalone (i.e. the system is not in normal operation). This allows
you to set the network addresses, subnet masks, and servers used for
various purposes. The system uses IEN116 for name service. (This is
an older name server standard. There are Unix implementations
available.) You can use one or more of the boxes as name server --
they keep the name table on floppy, or set up one or more of your Unix
systems as a server. The terminal interface is highly configurable.
You can set up any characters to do character echoing, tailor the
prompts and greeting messges, etc. We set it up to have the same
control characters as TOPS-20 or VMS. It appears that they have
options oriented to half duplex, and every other conceivable kind of
terminal environment. Of course, you can choose parity, character
size, and all of that. The box is designed to allow you to support
machines that don't have their own TCP/IP implementations. You can
connect a group of ports from a CS-100 to ports on the host. You can
define those ports as a hunt group with its own Internet address.
Then anyone who telnets to that address will get the first free line
in the group. There is enough processing power in the box to be able
to handle 9600 baud output under normal circumstances. However the
CS-100 is short on memory, so certain combinations of uses can cause
trouble (e.g. if you are doing this, and using the same box to drive a
printer). I'll say a bit more about this below. The following
transcript will give you an idea of the configuration options, as well
as the available commands. (By the way, the mechanism that we used to
produce this transcript puts us into system manager mode from our
favorite Unix machine. We could do it to any box on the Internet,
without typing a password. Fortunately, it more than a simple telnet,
and the program does not appear to be widely distributed. There is
also a limit to the damage one could do, since serious configuration
changes have to be done with a sysgen.)
Remote: show parameters
...............................Global Parameters...............................
DATE = Thu Nov 27 05:52:54 1986
WelcomeString = "↑G↑J↑MBridge CS/100, Rutgers LCSR Ethernet, Node Hill-SYS, Version 1.2000↑J↑M"
PROmpt = "Hill-Sys> " NMPrompt = "SYS> "
LocalPassWord = ... GlobalPassWord = ...
CONNectAudit = OFF ERRorAudit = ON
Remote: show allsessions [NB: It's before dawn on Thanksgiving Day]
Port/session# state Port/session# state
! 0 LISTEN ! 1 LISTEN
! 2 LISTEN ! 3 LISTEN
! 4 LISTEN ! 5 LISTEN
! 6 LISTEN ! 7 LISTEN
! 8 LISTEN ! 9 CONCTD with 128.006.005.107
!10 LISTEN !11 LISTEN
!12 LISTEN !13 LISTEN
Remote: show (!9) parameters
Parameters for PortId !9, current session
...................Port Transmission and VTP Characteristics...................
BUffersize = 82 DeVice = ( Terminal, Glass )
InterAction = ( Verbose, Echo, NoMacroEcho, BroadcastON, NoLFInsert )
InitMacro = "motd" MaxSessions = 6 PRIvilege = User
.........................Port Physical Characteristics.........................
BAud = 9600 BSPad = None CRPad = None FFPad = None
LFPad = None TabPad = None DataBits = 8 DUplex = Full
LinePRotocol = ASynchronous PARIty = None StopBits = 1
UseDCDout = ( AlwaysAssert, NoToggle ) UseDTRin = Ignore
.................Session Transmission and VTP Characteristics..................
BReakAction = ( InBand ) BReakChar = Disabled
DIsconnectAction = None DataForward = None ECHOData = OFF
ECHOMask = ( AlphaNum, CR, Term, Punct ) ECMChar = ↑↑
EOM = Disabled FlowControlFrom = Xon←Xoff
FlowControlTo = Xon←Xoff FlushVC = OFF IdleTimer = 2
LongBReakAction = IGnore LFInsertion = None MOde = Transparent
XOFF = ↑\ XON = ↑\
..................Sess
Remote: ?
BRoadcast ( <addr> ) <string>
Connect ( <addr> ) <address> [ ECM ]
DEFine <macro-name> = ( <text> )
DisConnect ( <addr> ) [<session number>]
DO <macro-name>
Echo <string>
Listen ( <addr> )
ReaD ( <addr> ) <option> <parameter>
ROtary !<rotary> [+|-]= !<portid>[-!<portid>] , ...
SAve ( <addr> ) <option> <filename>
SET <param-name> = <value> ...
SETDefault ( <addr> ) [<param-name> = <value>] ...
SHow ( <addr> ) <argument> ...
UNDefine <macro-name>
UNSave <filename>
ZeroStats
<BREAK> (to leave remote mode)
Remote: sho (!9) stat [This was done Thanksgiving day. Not very busy...]
PORT # 000.000.000.000 STATISTICS REPORT: -------------
DAILY-AVERAGE: CALL/D PKT/S BYTE/S ERROR/D
0 0 0 0
BUSIEST-MINUTE: CALL/M PKT/S BYTE/S ERROR/M
0 0 0 0
BUSIEST-SAMPLE: MAX#SN PKT/S BYTE/S ERROR
1 0 0 0
HOUR CALL/H PKT/S BYTE/S ERR/H HOUR CALL/H PKT/S BYTE/S ERR/H
0 0 0 0 0 1 0 0 0 0
2 0 0 0 0 3 0 0 0 0
4 0 0 0 0 5 0 0 0 0
6 0 0 0 0 7 0 0 0 0
8 0 0 0 0 9 0 0 0 0
10 0 0 0 0 11 0 0 0 0
12 0 0 0 0 13 0 0 0 0
14 0 0 0 0 15 0 0 0 0
16 0 0 0 0 17 0 0 0 0
18 0 0 0 0 19 0 0 0 0
20 0 0 0 0 21 0 0 0 0
22 0 0 0 0 23 0 0 0 0
Remote: sho ?
SHow ADDRess
SHow AllSessions [ p ]
SHow CONFigurationS [<filename>]
SHow ( <addr> ) DefaultParameters [<param-name> ...]
SHow GLobalParameters
SHow InternetPorts
SHow InternetServers
SHow MACros [<macro-name>]
SHow NAmes [<host name>]
SHow NetMAP
SHow ( <addr> ) PARAmeterS [<param-name> ...]
SHow <param-name> ...
SHow ROtaries
SHow ( <addr> ) SESsions [ P ]
SHow ( <addr> ) STATisticS [ Sample | Min | <hour> | Day ]
SHow VERSion
SHow VirtualPorts
<BREAK> (to leave remote mode)
Remote: ↑C
The CS-100's are reasonably reliable when used as simple terminal
servers. However from time to time we have run into glitches in their
TCP, which make us wary of using it for anything unusual. For
example, at one point we ran a printer off a line on a CS-100. The
host would connect to that port in order to access the printer. The
box where we did this always seemed to crash more often than others.
Also, if there were problems, we had to reboot the box. Apparently,
they did not implement RST in TCP. If the host crashed, this could
lead the CS-100 to keep trying to send a character to a connection
that no longer existed. It would ignore the RST that was sent to tell
it to desist. At times, the rate of retry could be high enough to
noticably affect the performance of the machine being attacked.
Historically, we have had fairly long delays in getting TCP problems
of this sort fixed. TCP/IP was clearly a lower priority with them
than XNS. However over the last few months, they have apparently
raised the priority of TCP/IP, and we are now getting fixes to long-
standing problems. Whether the particular problem I just described
is now fixed, I don't know. The most serious problem is that the
boxes are short of memory.
- There is a limit to the number of sessions you can have active
at once. If you use all 14 ports on a CS-100, there are
only 4 extra sessions. I.e. 4 people can have two sessions,
or one person can have 4, but that's all.
- The TCP buffers are small. The packet sizes tend to be very
small. (I just telnetted to one, and got a send window
of 102.) This can increase the CPU overhead on the host
and the network traffic when doing such things as Emacs
screen refreshes.
This is a known problem, and may not be present with all of the models,
or even on newer CS-100's. So you should check with your salesman
to find out the current limits.
Note that Bridge has a large line of TCP/IP products. I don't know
the current product list, but it includes gateways of various sorts,
and I think also a box that can handle 3270's.
=============
The cisco ASM is one of a set of products that use the same chassis.
It is a standard Multibus backplane, into which various boards can be
inserted. Unlike Bridge, they don't actually have a finite set of
products. They have these things with product numbers, like ASM-32/S.
But in fact all they are is a certain set of boards plugged into the
box. You can add boards and produce objects that are a combination of
different announced products. (E.g. if you add a second Ethernet card
to a terminal server, you have a thing that works as both a terminal
server and a gateway.) There is a single CPU, a 68000 on a cisco
version of the SUN card. (No connection with Sun Microsystems. Sun
was one of several people who licensed the original SUN design from
Stanford. Cisco's card is most similar to the Forward Technology SUN
board.) The board has plenty of memory. (1MB, which is loads of
memory for this application.) Terminals are connected to terminal
interface cards that handle 16 ports. The 3Com Ethernet card is
currently used. Cisco supports up to 80 terminals on one box.
However if they are all active at once, there will not be enough CPU
power in the 68000. This would be used if you had lots of offices,
but you knew that not everybody was logged in at once. Cisco claims
that 32 ports can be in use at once with no problem. We have a number
of 32-port boxes, and have never seen any slowdowns. I think we are
probably going to start adding port cards so that we have 48-port
boxes. They have a packaging problem with all of these ports. How do
you put 80 RS232 connectors on the back of a box? Obviously, you
don't. Their preferred configuration uses 50-wire phone company
cables. They have the standard phone company connectors on the back
of the boxes. We run the cables to a board on the wall, where we fan
the wires out to phone company punch blocks. Other wires are then run
out to the terminals. We can then cross patch any terminal to any
port. This is certainly the best way to handle large installations.
It results in compact connectors and fewer cables. However if you
prefer RS232 connectors, they will put up to 32 on the back of their
box. I think they also have a kludge for putting extra ones near the
box, connected by ribbon cable. The boxes normally have their code in
ROM, though there are provisions to load it using TFTP from any host
system that supports that. (Cisco sends out new code by giving us new
ROM's.) When a machine comes up, it needs two services from some
server: (1) It has to find its Internet address. It sends out both
RARP and bootp requests. Bootp servers are available for 4.3. We
also run it on 4.2, but I think it needs one extra ioctl. RARP is
also widely available. However before buying one of these boxes, you
should verify that you can run one or the other of these. (2) It
attempts to load a configuration file using TFTP. This contains
information such as terminal speeds, host name, greeting message,
routing tables, etc. TFTP servers are widely available, and should
run on just about any system that supports TCP/IP. Host name can be
handled via either IEN116 (an older TCP/IP name server protocol), or a
domain server. It will broadcast, or you can give it a list of
servers to use.
System administration can be done using any terminal or an incoming
telnet connection. Simply enable (which requires a password). It
is possible to type any of the commands interactively that go in the
configuration file, though it would be more likely to update the
configuration file and request that it be reloaded. (This will not
disturb other users, as they won't change parameters of terminals
that are currently in use.) One slight security problem: the password
is defined in the configuration file. Most versions of TFTP will only
access files that are protected so that everyone can read them. This
means that some machine on your network will have the password in a
file that is readable by the world. (We have altered TFTP to allow
us to protect the file. It uses a slightly nonstandard interpretation
of .rhosts to let us limit access to just cisco terminal servers.)
The command language looks like a typical DEC command scanner. It's
modelled after TOPS-20, but would look familiar to any VMS user, and
indeed to most Unix users. It responds to rubout or backspace, ↑U or
↑X, ↑W [word delete], and ↑R [retype line]. If you just type a host
name, it will connect. We have not seen any limit to the number of
connections you can have active at once. The example below will give
a feeling for the commands and options available. (It was obtained by
a normal telnet connection.) One of the strong points of the system
is that they try to let you see as many internal tables and parameters
as possible. This can be very useful in debugging situations. Note
that no ? help is available for the configuration commands. However
the results of the "show" commands will give a good feeling for the
sorts of options that can be set. There are not quite as many
terminal options as with Bridge. In particular
- I don't think any attempt is made to handle half-duplex terminals
- You can't change the editing characters (backspace/rubout, etc)
- No padding is supported. (The host system is assumed to do that.)
However there are more network configuration options. There is also
access control. You can control incoming or outgoing connections on
any port. Access control lists can contain wildcards. You can also
use this mechanism in their gateways to control which hosts access a
given network. (We use it for Arpanet access control. If you list
individual hosts, a hash table is used, so it should not cause a
performance problem. A list with several wildcards might not be quite
so fast.) Cisco appears to have implemented the IEEE 802.whatever
encapsulation. In addition to the usual Ethernet encapsulation using
ARP's, they support two versions of IEEE/ISO encapsulation, one which
follows the newest proposed method, and the other which seems to be
peculiar to H-P.
Connected to hilltop.rutgers.edu.
Escape character is '↑]'.
cisco ASM-32, Rutgers LCSR Computer Science Network, Node Hilltop.
Please type name of machine you wish to connect to, followed by a <CR>.
hilltop>enable
Password:
hilltop#?
banner Change message of the day banner
clear Reinitialization functions, type "clear ?" for list
configure Configure from terminal or over network
connect <host> Connect to host - same as typing just a host name
disconnect <cn> Break the connection specified by name or number
disable Turn off privileged commands
enable Turn on privileged commands
exit, quit Exit from the EXEC
name-connection Give a connection a logical name
ping Send ICMP Echo message
reload Halt and reload system
resume Make the named connection be current
send <line>|* Send message to a terminal line or lines
set <option> Set an option, type "set ?" for list
show <cmd> Information commands, type "show ?" for list
systat Show terminal lines and users
terminal Change terminal's hardware parameters, type "terminal ?"
unset <option> Clear an option, type "unset ?" for list
where Show open connections
<cr> To resume connection
hilltop#set ?
download.mode Optimize settings for using Kermit, etc.
egp-tracing Detailed printout of EGP transactions
escape <ch> Local escape character
event-watching Display special gateway events
gateway Gateway processing activity
hold <ch> Local hold character
imp-loopback Put any ARPA-1822 interfaces into self loopback
line-debugging Helpful debugging printout for RS232 lines
notify Notification of data pending on idle connections
tcp-debugging Debugging printout for TCP connections
tracing Print datagram routing information
hilltop#sho ?
access-lists Access control lists
arp ARP cache
buffers Network buffer utilization
controllers Serial network interface statistics
egp EGP neighbors
hardware Hardware configuration
hosts Host/address cache
imp-hosts Active IMP hosts
interfaces Network interface statistics
line <line> Line information, may specify a line
memory Memory utilization statistics
options Options configured via "set" and "unset"
printers Parallel printer status
processes Active system proceses
redirects ICMP redirect cache
routes Network routing table
stacks Process and interrupt stack use
terminal Terminal parameters
tcp <line> TCP information, may specify a line
traffic Network protocol statistics
users Summary of active lines and connections
hilltop#term ?
databits 5|6|7|8
flowcontrol none|hardware|software [in|out]
length <length>
parity none|even|odd|space|mark
speed 300|1200|2400|4800|9600|19200
start-character <decimal-number>
stop-character <decimal-number>
stopbits 1|2
terminal-type <string>
hilltop#sh line 14
Tty Typ Tx/Rx A Mode Status Capab Roty AccO AccI Uses Noise
* 14 TTY 1200/1200 L modem 100448 1100 - 1 - 139 79
Location: "Dialup x2970", Type: "", Length: 24 lines
TX/RX speeds are 1200/1200, 8 databits, 1 stopbits, no parity
No flowcontrol in effect.
Status currently=0x100448, default=0x100020, permanent=0x40
Capability currently=0x1100, default=0x1100, permanent=0x0
Idle EXEC timeout is 5 minutes.
Idle session timeout is 120 minutes.
Escape character is ↑↑ (30), default is ↑↑ (30)
Hold character is ↑\ (28), default is ↑\ (28)
Disconnect character is not set
Activation character is ↑M (13)
hilltop#systat
TTY Host(s) Location
tty14 H008-19 Dialup x2970
tty22 TOPAZ Dialup x2976
tty34 TOPAZ Dialup x2954
*vty41 idle TOPAZ
vty42 idle 192.12.88.3
hilltop#sho traffic
IP statistics:
Rcvd: 3075505 total, 23 format errors, 79 checksum errors, 0 bad hop count
0 unknown protocol, 2998056 local destination, 77347 not a gateway
Frags: 0 reassembled, 0 timeouts, 0 fragmented, 0 couldn't fragment
Bcast: 974 received, 4 sent
Sent: 0 forwarded, 2395890 generated, 64 encapsulation failed, 0 no route
ICMP statistics:
Rcvd: 1 checksum errors, 85 redirects, 46 unreachable, 342 echo
0 echo reply, 35 mask requests, 20 mask replies, 0 other
Sent: 0 redirects, 0 unreachable, 0 echo, 342 echo reply
1 mask requests, 0 mask replies
UDP statistics:
Rcvd: 1526 total, 0 checksum errors, 1131 no port
Sent: 1653 total, 0 forwarded broadcasts
TCP statistics:
Rcvd: 2996037 total, 471 checksum errors, 716 no port
Sent: 2394255 total
π --More--
ARP statistics:
Rcvd: 47699 requests, 1177 replies, 78 reverse, 69 other
Sent: 289 requests, 533 replies (0 proxy), 1 reverse
Xerox ARP statistics:
Rcvd: 0 requests, 0 replies
Sent: 0 requests, 0 replies
Probe statistics:
Rcvd: 16217 address requests, 81 address replies, 0 other
Sent: 289 address requests, 4 address replies (0 proxy)
hilltop#sho interface
Ethernet #0 is up, hardware address 0260.8C02.5606, IP address 128.6.4.56
MTU is 1504 bytes, encapsulation is ARPA, no access checking
Address determined by Reverse ARP from host 128.6.4.194
Time since last input is 0:00:00.000
Time since last successful output is 0:00:00.016
No output failure has occurred
3601442 input, 4119 with errors, 541 no input buffers
2401375 output, 3443 with errors, 0 congestion drops
202 resets, 0 runts rcvd, 0 giants rcvd
hilltop#sho ?
access-lists Access control lists
arp ARP cache
buffers Network buffer utilization
controllers Serial network interface statistics
egp EGP neighbors
hardware Hardware configuration
hosts Host/address cache
imp-hosts Active IMP hosts
interfaces Network interface statistics
line <line> Line information, may specify a line
memory Memory utilization statistics
options Options configured via "set" and "unset"
printers Parallel printer status
processes Active system proceses
redirects ICMP redirect cache
routes Network routing table
stacks Process and interrupt stack use
terminal Terminal parameters
tcp <line> TCP information, may specify a line
traffic Network protocol statistics
users Summary of active lines and connections
hilltop#sho line
Tty Typ Tx/Rx A Mode Status Capab Roty AccO AccI Uses Noise
0 CTY - direct 40 0 - - - 1 1
1 TTY 4800/4800 - direct 400040 0 - 2 - 0 107
2 TTY 4800/4800 - direct 400040 0 - 2 - 70 2757
3 TTY 4800/4800 - direct 400040 0 - 2 - 83 147
4 TTY 4800/4800 - direct 400040 0 - 2 - 70 13
5 TTY 4800/4800 - direct 400040 0 - 2 - 33 46
6 TTY 4800/4800 - direct 400040 0 - 2 - 54 2036
7 TTY 4800/4800 - direct 400040 0 - 2 - 0 2322046
10 TTY 2400/2400 - direct 40 0 - 1 - 39 8
11 TTY 2400/2400 - direct 400040 0 - 1 - 0 0
12 TTY 2400/2400 - direct 40 0 - 1 - 30 41
13 TTY 2400/2400 - direct 40 0 - 1 - 57 8
* 14 TTY 1200/1200 L modem 100448 1100 - 1 - 139 79
15 TTY 1200/1200 L modem 100020 1100 - 1 - 112 23
16 TTY 1200/1200 L modem 100020 1100 - 1 - 103 107
17 TTY 300/300 L modem 100020 1100 - 1 - 81 13
20 TTY 1200/1200 L modem 100020 1100 - 1 - 70 21
21 TTY 1200/1200 L modem 100020 1100 - 1 - 41 3
* 22 TTY 300/300 L modem 500600 1100 - 1 - 2 166
23 TTY 1200/1200 L modem 100020 1100 - 1 - 22 1
π --More--
Tty Typ Tx/Rx A Mode Status Capab Roty AccO AccI Uses Noise
24 TTY 1200/1200 L modem 100020 1100 - 1 - 7 45
25 TTY 1200/1200 L modem 100020 1100 - 1 - 13 12
26 TTY 1200/1200 L modem 100020 1100 - 1 - 35 29
27 TTY 1200/1200 L modem 100020 1100 - 1 - 5 0
30 TTY 9600/9600 L modem 100020 1100 - 1 - 0 0
31 TTY 9600/9600 L modem 100020 1100 - 1 - 0 0
32 TTY 2400/2400 L modem 108000 1100 - 1 - 78 389
33 TTY 2400/2400 L modem 100020 1100 - 1 - 34 19
* 34 TTY 2400/2400 L modem 100448 1100 - - - 25 34
35 TTY 9600/9600 L modem 100020 1100 - 1 - 0 0
36 TTY 2400/2400 L modem 100020 1100 - 1 - 49 2258
37 TTY 2400/2400 L modem 100020 1100 - 1 - 9 0
40 TTY 9600/9600 L modem 100020 1100 - 1 - 0 0
* 41 VTY - virtual 120440 1 - 2 - 45 0
* 42 VTY - virtual 120440 0 - 2 - 3 0
43 VTY - virtual 120240 0 - 2 - 8 0
44 VTY - virtual 120020 0 - 2 - 0 0
45 VTY - virtual 120020 0 - 2 - 0 0
hilltop#quit
Connection closed by foreign host.
The primary issue with cisco is that they are a new and small company.
We have had problems show up, as you would expect with any new
product. They have all been fixed reasonably quickly. Simple coding
blunders are normally fixed within a couple of days. (I trust no one
expects a bug free product. We are not so concerned about finding
bugs in new products. We had at least as many bugs in the early days
of the CS-100's. We are more concerned with how hard it is to get
them fixed. As far as I know, there are no unfixed bugs at the moment
in the Cisco software.) Rutgers has, as usual, run into a few really
difficult problems. The most difficult ones turned out to be design
problems with two different boards used in the Arpanet gateway (both
from established vendors). The question that a few of us have is
whether they will be able to continue their good support when they
have hundreds of customers. The biggest problem we have with small
companies is that when they succeed, it is no longer practical for
everyone to talk to their wizards. Either they supply no support, or
they build up a large staff of turkeys to deal with the users. (We
have seen both strategies.) However we don't know of any more
established vendors that have really brilliant solutions to this
problem either. One of the stengths of the company is Len Bosack's
expertise in the area of TCP/IP and routing technology. The folks
at Bridge are certainly competent, but our evidence does not suggest
that they have anyone of his caliber. This shows up in how the
nooks and cranies of TCP are handled. (It is inconceivable that
cisco would put out a TCP that failed to implement RST.) It also
matters if you are thinking of building a large network, where
routing technology matters. (However Bridge has in the past tended
to license DEC's technology for routing. That is a perfectly
acceptable solution.)
In short, both Bridge and cisco make useful products. We think
cisco's software design is somewhat superior. But you have to balance
this against the dangers of dealing with a startup company, with
the details of what the particular products do and don't support,
and with the cost of the equipment needed for your particular
configuration. (E.g. the Bridge CS-200 would tend to be more
cost-effective for locations with very small numbers of terminals,
but in most situations, cisco would probably come out ahead.)
∂09-Dec-86 1356 KHOWARD@Score.Stanford.EDU message
Received: from SCORE.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 9 Dec 86 13:55:05 PST
Date: Tue 9 Dec 86 13:52:32-PST
From: Katherine Howard <KHOWARD@Score.Stanford.EDU>
Subject: message
To: les@Sail.Stanford.EDU
Message-ID: <12261500251.37.KHOWARD@Score.Stanford.EDU>
Evelyn Broach returned your call. Her number is 3-0630.
--Kathy
-------
She says that cost plus indirect cost does not generate unrelated business
income.
∂10-Dec-86 0327 CS.EJK@forsythe.stanford.edu LEN BOSACK MEETING
Received: from FORSYTHE.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 10 Dec 86 03:27:17 PST
Received: by lindy.STANFORD.EDU; Wed, 10 Dec 86 03:18:30 PST
Date: Wed, 10 Dec 86 03:25:04 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: LEN BOSACK MEETING
Les:
I have talked with Len Bosack and set up a meeting with him
for Dec 23 to discuss the license. This is the earliest we
could get together as he is going out of town until then.
Will keep you posted!
Lisa
To: LES@SAIL
cc: GD.WHY, HK.NJR, HK.PLD, HK.JXW
∂16-Dec-86 1928 LES Licensable Materials
To: CS.EJK@FORSYTHE.STANFORD.EDU
CC: Nilsson@SCORE.STANFORD.EDU, HK.JXW@FORSYTHE.STANFORD.EDU,
HK.PLD@FORSYTHE.STANFORD.EDU, GD.WHY@FORSYTHE.STANFORD.EDU,
Hansen@SIERRA.STANFORD.EDU
In response to your verbal request today, the things that I believe should
explicitly claimed as Stanford property as part of the prospective cisco
licensing agreement are:
MEIS board design (artwork copyright)
MEIS software (source file copyright)
SUN board with 256k memory chips (artwork copyright)
SUN gateway/TIP programs (source file copyright)
I trust that you have a copy of the gateway/TIP software in the form of a
magnetic tape from Steve Hansen. Jasper Williams is looking into the
question of when these various copyrights should be registered.
I invite dissenting opinions, if any.
Les
∂17-Dec-86 1425 GD.WHY@forsythe.stanford.edu Licensable Materials
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 17 Dec 86 14:25:05 PST
Received: by lindy.STANFORD.EDU; Wed, 17 Dec 86 13:30:18 PST
Date: Wed, 17 Dec 86 13:29:44 PST
From: Bill Yundt <GD.WHY@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: Licensable Materials
REPLY TO 12/16/86 19:34 FROM LES@SAIL.STANFORD.EDU "niz": Licensable Materials
Lisa,
...not a dissent, but some footnotes...
Adrian Arima told me he as hard copy listings of the
Tip/Gateway code...in his office I think...a few feet of it.
You might want to have those for Len's inspection when you
talk. I would assume they are all dated and have version
identification in the code. He may wish to claim some parts
are not Stanford owned, in which case we should discuss
same.
I trust you are in possession of all previous correspondence
including that between Street and Bosack, etc. Should think
your position should be consistent with views therein and
that you can point to the previous statements as the
position of academic administration (for whatever worth that
is).
I do not understand Les' statement about Jasper and dates
for filing copyright....unless Jasper means the date
the works were authored is in question...which I would
not doubt. I thought there was no disagreement on the
Gateway/TIP code registration and that it was proceeding
or had been accomplished. If not, it sure should be!!!
...Bill
To: CS.EJK
cc: LES@SAIL, NILSSON@SCORE.STANFORD.EDU, HK.JXW, HK.PLD
∂19-Dec-86 1851 A.ERIC@GSB-HOW.STANFORD.EDU Ampex memory situation
Received: from GSB-HOW.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 19 Dec 86 18:51:38 PST
Date: Fri 19 Dec 86 18:47:44-PST
From: Eric M. Berg <A.ERIC@GSB-HOW.STANFORD.EDU>
Subject: Ampex memory situation
To: LES@SU-AI.ARPA
cc: a.bill@GSB-HOW.STANFORD.EDU
In-Reply-To: Message from "Les Earnest <LES@SAIL.STANFORD.EDU>" of Fri 19 Dec 86 18:38:00-PST
Phone-#s: 723-1576 (GSB-CF), 329-9940 (home), 725-6900 x.31576 (Voice mail)
At our request, Cisco Systems assisted with the original effort to install
the new Ampex cabinet on GSB-Why, since no one here was knowledgable about
how to do so. I don't have the details of how much they were paid yet.
As I recall, when we went to install it, we had (several) problems with power
supplies, and the only power supply available didn't have enough output to
power the entire 2 meg. of memory. Thus, the cabinet was installed on the
system, but with only 1 meg. operational. Later, when we got the original
power supply repaired, we tried unsuccessfully to bring up the rest of the
memory. (I'll have to double-check the details here with other people on
our staff.)
I've talked to Tom D., and he wants us to bring the cabinet back to CSD
for work there, rather than his coming over here to look at it. That's
okay with us, but, I'm concerned about the possibility of a situation where
(i) CSD says the memory works fine on Sushi or Truffle but (ii) we can't get
it to work at GSB. I think we'd like some commitment from CSD that you will
help us get the memory running \over here/ once it has been determined to be
in working order, rather than just saying "it works, take it, you're on your
own". (We'd be willing to pay for that help, if necessary.)
Bill Elbring mentioned that he'd like to get a chance to meet you, and
suggested that I set up a time for the three of us to get together to
discuss topics of mutual interest, including the Ampex situation. I'll
be in touch to try & schedule such a meeting.
Thanks.
-------
∂21-Dec-86 1715 LES re: Licensable Materials
To: GD.WHY@FORSYTHE.STANFORD.EDU
[In reply to message sent Wed, 17 Dec 86 13:29:44 PST.]
The licensable materials have not been registered with the copyright
office yet. In my last conversation with Jasper Williams, he did not
have a clear idea about whether it was important to formalize the
copyright registration in advance of the licensing or not, given that
Stanford owns the copyright even before it is registered.
I recommended that Jasper get some outside help with the copyright
issues and suggested local lawyer Aldo Test as one possible source.
He indicated that he would do that. I gather that you didn't hear
anything from Iris Brest.
Incidentally, a letter to cisco asking for full payment on the ethernet
transaction supposedly went out on Friday. Jasper chose to use the 3COM
price that I got from you of $840, even though I pointed out that the 3COM
units were produced in large volume and were therefore made at much lower
cost than were our boards. In fact, according to my friends in industry
(i.e. using their pricing formula based on fabrication cost), the 3 Mbit
ethernet boards should have been priced at about $1600 each.
Les
∂24-Dec-86 0900 CS.EJK@forsythe.stanford.edu LEN BOSACK MEETING
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 24 Dec 86 08:59:56 PST
Received: by lindy.STANFORD.EDU; Wed, 24 Dec 86 08:57:31 PST
Date: Wed, 24 Dec 86 08:59:28 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: LEN BOSACK MEETING
Dear Folks:
I had a good meeting with Len Bosack yesterday. Will summarize
briefly.
We discussed the items to be included in the license: MEIS
hardware and software, SUN board(256K version), and Gateway/TIP
software.
Len said that they had sold 6-7 of the MEIS boards, are not
actively marketing it anymore.
Len described their business and product strategy as well as
pricing information. He also promised to send product
literature to me.
Len agreed to provide enhancements to the Gateway/TIP software
"almost indefinitely" to Stanford as a part of the agreement.
cisco is not interested in any exclusivity in the license.
cisco has the equivalent of 12 full time employees and Len expects
their sales this year to be $1-2 M. They have not gotten
outside funding as of yet but will probably be doing so sometime
during the next year in order to expnad the business.
Their typical products sell for $9-12K.
We discussed Stanford's wants in a license, namely royalties and
access to the software.
I have more detailed information. If anyone is interested, please
let me know.
As soon as convenient, I would like to
meet with you Bill and Les to discuss a licensing proposal to
go back to cisco with.
Happy Holidays!
Lisa
To: LES@SAIL
cc: GD.WHY, HK.NJR, HK.JXW, HK.PLD, HANSEN@SIERRA, HK.AXA
∂26-Dec-86 1242 HK.JXW@forsythe.stanford.edu LEN BOSACK MEETING
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 26 Dec 86 12:42:49 PST
Received: by lindy; Fri, 26 Dec 86 12:40:18 PST
Date: Fri, 26 Dec 86 12:29:19 PST
From: Jasper Williams <HK.JXW@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: LEN BOSACK MEETING
REPLY TO 12/24/86 08:58 FROM CS.EJK "Lisa Kuuttila": LEN BOSACK MEETING
Lisa, This sounds good and I would like to go over with you on
the phone or the next time that that I am down there. It probably
would be good if Bill, Les, Steve Hansen and I met with you
to go over the details. Jasper
To: CS.EJK
cc: LES@SAIL, GD.WHY, HK.NJR, HK.PLD, HANSEN@SIERRA, HK.AXA
∂25-Dec-86 1114 YEAGER@SUMEX-AIM.STANFORD.EDU cisco
Received: from SUMEX-AIM.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 25 Dec 86 11:14:32 PST
Date: Thu 25 Dec 86 11:14:22-PST
From: Bill <Yeager@SUMEX-AIM.STANFORD.EDU>
Subject: cisco
To: les@SAIL.STANFORD.EDU
Message-ID: <12265665762.13.YEAGER@SUMEX-AIM.STANFORD.EDU>
Les, so what's up with cisco these days? Is this thing ever going to get
settled?
I would like to know for many reasons ...
Thanks, and happy holidays!
Bill
-------
∂26-Dec-86 1311 LES re: cisco
To: Yeager@SUMEX-AIM.STANFORD.EDU
[In reply to message sent Thu 25 Dec 86 11:14:22-PST.]
Len is talking with OTL about license and things seem to be moving in the
right direction. It may get weird again now that we have asked for
reimbursement for the ethernet boards at half their market value (i.e.
double what Len was offering).
Merry New Year!
Les
∂20-Jan-87 1507 CS.EJK@forsythe.stanford.edu STATUS OF CISCO LICENSE/ SUN PATENT
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 20 Jan 87 15:07:14 PST
Received: by lindy.STANFORD.EDU; Tue, 20 Jan 87 15:05:05 PST
Date: Tue, 20 Jan 87 15:08:08 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: STATUS OF CISCO LICENSE/ SUN PATENT
Les:
This is to confirm our conversation re the cisco license
and Sun patent.
I am revising the draft License Agreement which I gave
everyone at our Jan 9 meeting to incorporate the items we
discussed:
o The royalty will cover both hardware and software
sales. I proposed a declining royalty over time
starting at 7% and coming down to 5%. I discussed
the issue of copyright protection with David Lovejoy-
he says that the question of copyrighting PC artwork has
not been challenged in court, his opinion was to assert
copyright protection until we know otherwise.
o I proposed a $40,000 license fee.
o I included a discount for Stanford on purchases from
cisco of 40% off list.
I plan to send the revised agreement to cisco by this Thursday.
If youhave any other comments, please let me know before then.
I also got a call from Bill Graves, their new CEO. He has been
very helpful in bringing by the information I requested
about their products and pricing. It appears tha Len is out of
the negotiations. Bill wants to keep the negotiations on a very
business-like basis and doesn't want to know about past disputes
except as they may impact the license itself.
Jasper, he did mention that he had brought in new corporate
counsel who will be responding to your letter; Brobeck, Phleger,
& Harrison.
Bill made it clear that they would like to resolve the issue of
what Stanford's rights vs. SUN rights are in the SUN board.
He said that they may decide to procure a comparable board
commercially (cost $200) if it is too expensive or the issues
can't be sorted out. I said that we believed that Stanford had
rights and would list the components which we were licensing
to them in the license agreement.
I propose to send the agreement to Bill with Attachment 2
(definition of SUN Components) left blank at this time. As we
work out the issues with SUN we will define what we are
licensing to cisco.
Steve, I also need the description of the Gateway/TIP software
to include in one of the attachments.
On the issue of the SUN patents, Les has determined that the
SUN patent is based on work done at Stanford. We have set up
a meeting with Les, Jasper, Niels and myself next Tuesday
Jan 27 at 2:00PM to discuss how to proceed with SUN. If anyone
else would like to attend, please feel free to come.
I think this lengthy message covers the status of this complex
case! Please let me have any comments you have on the
license by Thursday.
Thanks,
Lisa
To: LES@SAIL
cc: GD.WHY, HANSEN@SIERRA, HK.JXW, HK.AXA, HK.NJR
∂27-Jan-87 0848 CS.EJK@forsythe.stanford.edu CISCO NEGOTIATIONS STATUS
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 27 Jan 87 08:48:14 PST
Received: by lindy.STANFORD.EDU; Tue, 27 Jan 87 08:46:19 PST
Date: Tue, 27 Jan 87 08:48:31 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: CISCO NEGOTIATIONS STATUS
Dear cisco fans:
To bring everyone up to date on the cisco negotiations, I had a
couple of conversations with Bill Graves of cisco yesterday.
He had received the draft license agreement. At first he was upset
with the numbers; he had done some calculations and figured
that there should be a cap on the overall royalties of about
$60K since this is what it would cost him to replace the
Stanford technology at this point.
He calculated this by saying that they have 40 boards in the
field which would have to be replaced at a cost of $500 each
($20K) and the cost to rewrite the Stanford portions of the
software is 6 person-months ($40K).
We discussed his numbers and the fact that cisco had the use of t
the technology all this time, etc.
I think he will accept an up-front fee close to $40K and a lower
royalty (he proposes a fixed amount per board of $200 ). Bill
will get back to us with a firm counter-proposal within a week.
After our discussions, I think that Bill is fairly reasonable to
deal with but he gets input from Len who advocates "letting
Stanford try to take them to court" as a result of our proposal.
I am hopeful that Bill can be kept objective about this and that
Len stays out of the discussions.
Bill also said that cisco would like a license to the 3MB Ethernet
boards. I need to get some information from you all about what
these boards are. Were they also developed while Len was at
Stanford? cisoc is willing to pay a fixed royalty per board.
I propose that we simply add this item to the license and charge
a fixed fee per board with no additional up-front payment. How
does this sound?
Please give me your comments either by em or at the meeting this
afternoon scheduled to discuss the SUN situation.
Lisa
To: LES@SAIL, GD.WHY, HANSEN@SIERRA
cc: HK.JXW, HK.NJR
∂28-Jan-87 0828 HK.JXW@forsythe.stanford.edu CISCO NEGOTIATIONS STATUS
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 28 Jan 87 08:28:15 PST
Received: by lindy.STANFORD.EDU; Wed, 28 Jan 87 08:26:11 PST
Date: Wed, 28 Jan 87 08:28:32 PST
From: Jasper Williams <HK.JXW@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: CISCO NEGOTIATIONS STATUS
REPLY TO 01/27/87 08:45 FROM CS.EJK "Lisa Kuuttila": CISCO NEGOTIATIONS STATUS
Lisa, I have a call from their attorney withrespect to
the boards. Let me see what he has to say before we discuss
the boards further. Licensing the boards makes less sense
to me because they have treated it as a straight sale all along.
They have paid us $14,875 and we want an additional $20,396--
the profit they made on a sell to Xerox. Licensing the
boards would merely be a way of spreading this payment out and
perhaps reducing it. After I connect with their new attorney
I will talk to you. Jasper
To: CS.EJK
cc: HK.NJR, LES@SAIL, GD.WHY, HANSEN@SIERRA
∂28-Jan-87 1147 hansen@talbots.STANFORD.EDU Stanford Tip/Gateway Software Description.
Received: from TALBOTS by SAIL.STANFORD.EDU with TCP; 28 Jan 87 11:47:08 PST
Received: by talbots.STANFORD.EDU (1.1/4.7); Wed, 28 Jan 87 11:46:10 PST
Date: Wed, 28 Jan 87 11:46:10 PST
From: hansen@talbots.STANFORD.EDU (Stephen Hansen)
To: CS.EJK@forsythe
Subject: Stanford Tip/Gateway Software Description.
Cc: HK.JXW@forsythe, LES@sail
Lisa,
Following is a description of the components that make up
Stanford's Tip/Gateway Software.
Stephen Hansen
----------------------------------------------
o A Multiple Protocol Networking Operating System
o A TIP Application using this operating system,
o A Gateway Application using this operating system.
The Stanford Gateway/Ethertip software comprises a multiple
protocol networking operating system(MPNOS) and two applications
built on top of the MPNOS. The MPNOS includes a triple round-robin
process scheduler, dynamic memory management, device drivers for
the SUN 3megabit, 3COM and Interlan ethernet controllers, and
level-0 through level-2 support for PUP, IP, Chaosnet and XNS protocols,
The applications are a PUP and IP based
ethertip, and a gateway. The ethertips augment the MPNOS with
i/o drivers for both RS232 uart and parallel lineprinter
interfaces, a command processor, PUP and IP telnet interfaces, name/address
resolution using both PUP miscellaneous services and IP domains,
and support for multiple connections and virtual terminals.
Additionally, the EECF gateway provides class A and class B subnet
routing for the IP protocol, subnet routing for the PUP protocol,
address and reverse address resolution, routing table maintenance and
broadcast algorithms, and IP/ICMP, and IP/UDP internals including bootp
support. Along with these, the SUMEX gateway has routers for Chaosnet and XNS
protocols and provides a suite of internal services for these protocols,
such as Chaosnet RUT table broadcast, hoststat, and address resolution;
XNS routing table maintenance and broadcast, address and reverse address
resolution, and time servers.
∂28-Jan-87 1147 hansen@talbots.STANFORD.EDU Stanford Tip/Gateway Software Description.
Received: from TALBOTS by SAIL.STANFORD.EDU with TCP; 28 Jan 87 11:47:08 PST
Received: by talbots.STANFORD.EDU (1.1/4.7); Wed, 28 Jan 87 11:46:10 PST
Date: Wed, 28 Jan 87 11:46:10 PST
From: hansen@talbots.STANFORD.EDU (Stephen Hansen)
To: CS.EJK@forsythe
Subject: Stanford Tip/Gateway Software Description.
Cc: HK.JXW@forsythe, LES@sail
Lisa,
Following is a description of the components that make up
Stanford's Tip/Gateway Software.
Stephen Hansen
----------------------------------------------
o A Multiple Protocol Networking Operating System
o A TIP Application using this operating system,
o A Gateway Application using this operating system.
The Stanford Gateway/Ethertip software comprises a multiple
protocol networking operating system(MPNOS) and two applications
built on top of the MPNOS. The MPNOS includes a triple round-robin
process scheduler, dynamic memory management, device drivers for
the SUN 3megabit, 3COM and Interlan ethernet controllers, and
level-0 through level-2 support for PUP, IP, Chaosnet and XNS protocols,
The applications are a PUP and IP based
ethertip, and a gateway. The ethertips augment the MPNOS with
i/o drivers for both RS232 uart and parallel lineprinter
interfaces, a command processor, PUP and IP telnet interfaces, name/address
resolution using both PUP miscellaneous services and IP domains,
and support for multiple connections and virtual terminals.
Additionally, the EECF gateway provides class A and class B subnet
routing for the IP protocol, subnet routing for the PUP protocol,
address and reverse address resolution, routing table maintenance and
broadcast algorithms, and IP/ICMP, and IP/UDP internals including bootp
support. Along with these, the SUMEX gateway has routers for Chaosnet and XNS
protocols and provides a suite of internal services for these protocols,
such as Chaosnet RUT table broadcast, hoststat, and address resolution;
XNS routing table maintenance and broadcast, address and reverse address
resolution, and time servers.
∂28-Jan-87 1427 LES re: CISCO NEGOTIATIONS STATUS
To: HK.JXW@FORSYTHE.STANFORD.EDU
[In reply to message sent Wed, 28 Jan 87 08:28:32 PST.]
Jasper,
I thought that instead of requesting the full proceeds of the Xerox sale
(recognizing that cisco would claim that they performed some service in
support of the sale and should therefore share in the proceeds), we
decided to collect the nominal market value of $29,400. (As I remarked in
our earlier conversations, I believe that the true market value of those
boards was about $56,000). -Les
∂29-Jan-87 0843 HK.JXW@forsythe.stanford.edu re: CISCO NEGOTIATIONS STATUS
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 29 Jan 87 08:42:54 PST
Received: by lindy.STANFORD.EDU; Thu, 29 Jan 87 08:41:00 PST
Date: Thu, 29 Jan 87 08:43:20 PST
From: Jasper Williams <HK.JXW@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: re: CISCO NEGOTIATIONS STATUS
REPLY TO 01/28/87 14:42 FROM LES@SAIL.STANFORD.EDU "Les Earnest": re: CISCO
NEGOTIATIONS STATUS
Les, you are correct. This is the amount that we requested.
Sorry if my message conveyed a different impression. Jasper
To: LES@SAIL.STANFORD.EDU
∂29-Jan-87 1531 CS.EJK@forsythe.stanford.edu CISCO STATUS REPORT
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 29 Jan 87 15:28:26 PST
Received: by lindy.STANFORD.EDU; Thu, 29 Jan 87 15:26:31 PST
Date: Thu, 29 Jan 87 15:28:44 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: CISCO STATUS REPORT
cisco fans:
I had another meeting with Bill Graves today. We are working
toward determining a value for the individual components. I
think we have reached agreement on a fully paid license to
MEIS for $10K..... more to come on the SUN boards and the
software. He is looking into their costs to duplicate the
work to see what their bottom line is if they don't license.
He will also get me information on what their sales to date
have been. He estimates 40-60 systems. That should be good for
a portion of the up-front license fee.
Bill, the only objection they have on the equipment discount
is that it not be a flat 40%. They would like the agreement
to specify that we would get at least as good a deal as the
best deal an OEM customer gets. What do you think about this?
They claim that their margins will change over time; 40%
may not be reasonable a year from now.
Jasper, I don't see how settlement of the sale of the Ethernet
boards affects the licensing. They still have to repay
Stanford for the materials which Len took. In the license,
we are not licensing the physical board, only the copyright
protection and knowhow behind the design. I think we should
be able to license them the Ethernet board as well as settle
the sale to Xerox.
Bill has given the agreement to his attorney (groan....) WE
still may have some hurdles to overcome before this is all over.
Bill agreed to give me a firm proposal next week.
Will keep you posted in the continuing cisco saga.
lisa
To: LES@SAIL, GD.WHY, HANSEN@SIERRA, HK.JXW
cc: HK.NJR
∂29-Jan-87 2015 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU [Bill <Yeager@SUMEX-AIM.STANFORD.EDU>: OTL licensing]
Received: from SUMEX-AIM.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 29 Jan 87 20:14:15 PST
Date: Thu 29 Jan 87 17:56:25-PST
From: T. C. Rindfleisch <Rindfleisch@SUMEX-AIM.STANFORD.EDU>
Subject: [Bill <Yeager@SUMEX-AIM.STANFORD.EDU>: OTL licensing]
To: Les@SAIL.STANFORD.EDU
cc: Rindfleisch@SUMEX-AIM.STANFORD.EDU
Message-ID: <12274913992.21.RINDFLEISCH@SUMEX-AIM.STANFORD.EDU>
Les, do you know what is going on as far as CISCO royalty distribution? Is
there any hope of royalties?
Tom R.
---------------
Mail-From: YEAGER created at 29-Jan-87 17:19:07
Date: Thu 29 Jan 87 17:19:06-PST
From: Bill <Yeager@SUMEX-AIM.STANFORD.EDU>
Subject: OTL licensing
To: hansen@TALBOTS
cc: rindFLEISCH@SUMEX-AIM.STANFORD.EDU
Message-ID: <12274907200.43.YEAGER@SUMEX-AIM.STANFORD.EDU>
Steve,
I want to know what kind of licensing agreement is being written for
the software that CISCO negotiating. THe point is that the EECF code
comes directly from the SUMEX code and in fact I had worked 5
years on the code and Kirk about 1 good year. SO, any fees/etc
for the software should be shared and I believe 80/20 SUMEX/EECF
if there are any involved. I WILL push this very hard.
I personally would like to see all of it in the public domain, but
if that can't happen, then fair is fair. Please let me know who
you are dealing with at OTL. I would like to talk with them.
Thanks,
Bill
-------
-------
∂30-Jan-87 1250 LES re: [Bill <Yeager@SUMEX-AIM.STANFORD.EDU>: OTL licensing]
To: Rindfleisch@SUMEX-AIM.STANFORD.EDU
[In reply to message sent Thu 29 Jan 87 17:56:25-PST.]
Yes, negotiations are proceeding better now that cisco has a new CEO, Bill
Graves, who has a business background and is reportedly fairly rational.
I have not met him myself, but Lisa Kuutila of OTL reports steady
progress, interrupted only when Graves talks to Len and gets another
earful of nonsense.
I certainly concur with Bill Yeager's view that he played a major role in
the devlopment of the gateway/TIP software and I believe that other
interested parties are also aware of this. As I recall, Bill's code built
upon still earlier versions done by Jeff Mogul and others.
There is another complication that we are trying to resolve in parallel
with cisco negotiations: Sun patented the memory management scheme used
in the Sun CPU board even though it originated at Stanford. (I think that
I mentioned this discovery at a recent Senior Staff meeting and that it
should be treated as confidential for now). This complicates our plan to
immediately license that board to cisco. We are looking into the
possibility of rearranging the patent and getting royalties also from Sun,
a much bigger fish.
Les
∂01-Feb-87 0932 CSL.ALLISON@Sierra.Stanford.EDU Re: Memory Management
Received: from SIERRA.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 1 Feb 87 09:32:26 PST
Date: Sun 1 Feb 87 09:33:03-PST
From: Dennis Allison <CSL.ALLISON@Sierra.Stanford.EDU>
Subject: Re: Memory Management
To: LES@Sail.Stanford.EDU
In-Reply-To: Message from "Les Earnest <LES@SAIL.STANFORD.EDU>" of Sat 31 Jan 87 17:11:00-PST
Message-ID: <12275608791.9.CSL.ALLISON@Sierra.Stanford.EDU>
I'm not sure exactly whgat you're getting at. I've not looked at the SUN
patent though I'd like to. Andy told me that SUN was using the refresh cycle
on their dynamic RAMs to dothe screen update, a ct
-------
∂02-Feb-87 1352 CSL.ALLISON@Sierra.Stanford.EDU re: Memory Management
Received: from SIERRA.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 2 Feb 87 13:52:27 PST
Date: Mon 2 Feb 87 13:53:02-PST
From: Dennis Allison <CSL.ALLISON@Sierra.Stanford.EDU>
Subject: re: Memory Management
To: LES@Sail.Stanford.EDU
In-Reply-To: Message from "Les Earnest <LES@SAIL.STANFORD.EDU>" of Mon 2 Feb 87 12:43:00-PST
Message-ID: <12275918262.43.CSL.ALLISON@Sierra.Stanford.EDU>
I'd like to see a copy. Are you interested in killing the patent. I don't
think the Four Phase scheme applies, but there are many other systems which
use that trick to do single cycle reads.
Could you copy it and mail it to me at ERL 444. Thanks.
-dra
-------
∂30-Jan-87 2152 YEAGER@SUMEX-AIM.STANFORD.EDU gateway code
Received: from SUMEX-AIM.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 30 Jan 87 21:52:24 PST
Date: Fri 30 Jan 87 22:04:21-PST
From: Bill <Yeager@SUMEX-AIM.STANFORD.EDU>
Subject: gateway code
To: les@SAIL.STANFORD.EDU
cc: RINDFLEISCH@SUMEX-AIM.STANFORD.EDU
Message-ID: <12275221272.19.YEAGER@SUMEX-AIM.STANFORD.EDU>
Les, a slight correction. My code was NOT based on earlier versions by
Jeff Mogul and others. Rather, certain header files that defined uart
interfaces to the SUN board, and SUN PROM vector locations, etc ...
were freely used by all of us working on the first SUN board applications.
Those individuals that contributed to the SUN stuff were Jeff, Bill Nowicki,
and Vaughan Pratt. Others involved were Louis Prado(?sp) and John Seamons,
but I didn't directly use any of their PROM definitions.
Jeff Mogul built an earlier gateway that was based on code written by
the fellow who originally maintained Altos for CSD, and his gateway
in no way resembled what I did. They were two separate paths. IN fact,
there was a meeting around 1982 led by Keith Lantz in which we tried
to settle the gateway question at Stanford, ie, who would be responsible
for Stanford gateways and since I was a staff member and more permanent,
we decided to us my code instead of Jeff's. Jeff still contributed
intellectually to algorithms and we had many discussions as to the
correct way to do things.
In the TIP world, there was originally a TIP by Bill Nowicki running
at CSD. I also used the same operating system that my gateway used to
build a TIP. After discussions with Benjy Levy he decided to work
on extending my TIP rather than Bill's because it was more general, ie,
a real operating system with a process scheduler, etc. Yes, we cannot
forget Benjy. He did lots of stuff to extend the SUMEX TIP to fit
into the more general CSD/LOTS environment, and wrote the device
driver for the Schnurel duart boards that we all now use.
I know this history very well as you can imagine... Then in Feb.
of 1985 KirkL. was given my sources by me and had Benjy's
tip additions. The rest is history.
In any case, Tom and I want to be directly envolved in the OTL
negotiations.
Cheers,
Bill
-------
∂03-Feb-87 1357 CS.EJK@forsythe.stanford.edu TIP/GATEWAY ETC. ROYALTY-SHARING AGREEMENT
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 3 Feb 87 13:52:41 PST
Received: by lindy.STANFORD.EDU; Tue, 3 Feb 87 13:50:45 PST
Date: Tue, 3 Feb 87 13:51:53 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: TIP/GATEWAY ETC. ROYALTY-SHARING AGREEMENT
Dear cisco fans:
I have been talking with Bill Yeager about the royalty-sharing
agreement for the Tip/Gateway software etc.
It probably is a good idea to get the agreement worked out
in parallel with the negotiation of the cisco license.
I think it makes sense to have one agreement for all the components
included in the cisco license, namely the MEIS components, the
SUN components, and the Tip/Gateway software.
As you are probably aware, 15% of the income goes to our office
plus any out of pocket expenses are deducted (probably none
in this case). The net income is divided three ways: 1/3 to the
School, 1/3 to the department, and 1/3 to the developers.
Bill's proposal for the developers share is as follows:
Bill Yeager (SUMEX) 80%
Benjy Levy (CSD) 5%
Kirk Lougheed (EECF) 15%
The department and school shares will depend on the percentage
attributed to each developer. I invite your comments and
proposals for the royalty-sharing agreement.
Thanks,
Lisa
P.S. Elizabeth Batson is now back from her leave. She and I
will be working on this case together.
To: YEAGER@SUMEX-AIM, GD.WHY, HANSEN@TALBOTS, LES@SAIL, RINDFLEISCH@SUMEX
cc: CS.EAB, CS.EJK
∂03-Feb-87 1924 LES TIP/Gateway Software Royalties
To: YEAGER@SUMEX-AIM.STANFORD.EDU, GD.WHY@FORSYTHE.STANFORD.EDU,
HANSEN@TALBOTS, RINDFLEISCH@SUMEX-AIM.STANFORD.EDU,
CS.EAB@FORSYTHE.STANFORD.EDU, CS.EJK@FORSYTHE.STANFORD.EDU
I disgree with the idea that individuals should receive royalties from
licensing the TIP/gateway software. The proceeds, if any, should go to
various Stanford entities. I do agree that it is time to work out a
formula for distributing the proceeds among the interested
sub-organizations.
The creation of software that is substantially outside of (and in addition
to) regular staff responsibilities might be considered to be eligible for
revenue sharing, but any such cases should be handled very carefully -- I
do NOT wish to get into a situation where it becomes necessary to argue
with staff members about which projects they should work on because they
see the potential for royalties on certain projects and not on others.
In the case of the TIP/gateway software, it seems clear that the various
individuals who undertook development work did so as part of their regular
responsibilities and that no individual remuneration is appropriate, other
than the salaries that they have already received for doing this work. I
have no personal knowledge of the relative contributions of the various
parties but suggest that a meeting among those who do have such knowledge
should be able to resolve (or at least frame) this question.
Les Earnest
∂03-Feb-87 2155 YEAGER@SUMEX-AIM.STANFORD.EDU Re: TIP/Gateway Software Royalties
Received: from SUMEX-AIM.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 3 Feb 87 21:55:48 PST
Date: Tue 3 Feb 87 21:55:59-PST
From: Bill <Yeager@SUMEX-AIM.STANFORD.EDU>
Subject: Re: TIP/Gateway Software Royalties
To: LES@SAIL.STANFORD.EDU
cc: GD.WHY@FORSYTHE.STANFORD.EDU, HANSEN%TALBOTS@SUMEX-AIM.STANFORD.EDU,
RINDFLEISCH@SUMEX-AIM.STANFORD.EDU, CS.EAB@FORSYTHE.STANFORD.EDU,
CS.EJK@FORSYTHE.STANFORD.EDU
In-Reply-To: Message from "Les Earnest <LES@SAIL.STANFORD.EDU>" of Tue 3 Feb 87 19:24:00-PST
Message-ID: <12276268326.11.YEAGER@SUMEX-AIM.STANFORD.EDU>
Les,
The idea that individuals should be payed royalties is OTL's, and I understand
this to be Stanford policy. Whether we as individuals chose to keep the
royalties or place it in an unrestricted account for use by our respective
"entities" is another issue of interest here. So, I suggest you grapple
with Stanford/OTL about the "royalties," that "may" be paid. To be
quite honest I wanted to put the TIP/Gateway software in the public
domain(this is at the descretion of the PI under which is was
developed) but lost.
But, where the software came from is very clear, and cannot be argued.
It was essentially the work of one person, me, from late 1979 until
1985 or so with one exception. Benjy Levy decided to use the SUMEX
tip as a basis for CSD TIP development in 1983 and added lots of
neat features. I put my heart and sole into that work and spent
many a weekend pouring over memory dumps ferreting out the bugs
in the basic design. The results of the effort are what
Chuck Hedrick says "is the only piece of software he has ever
seen that NEVER crashes."
Kirk was given the SUMEX sources in Feb of 1985 and the rest is
history. As to whether or not a staff member will refuse to work
on a project as a function of it producing or not producing
potential royalites from commercial exploitation is a "red
herring" as far as I am concerned. As I see it we work on
what we do because of its research potential and that is
why we are at Stanford and not IBM.
I don't know why Stanford has decided to give a portion of
royalties from work such as this to the developers, and would
be interested in understanding the background of this policy,
if indeed Lisa Guuttila(sp?) is correct, and to be quite honest
and don't totally agree with your criticism. Perhaps we could
discuss this at lunch sometime...
Bill
PS>What do you think I would do with such royalties from cisco? (Certainly
NOT get rich!) I have been thinking about this since yesterday, and
believe I would want the money to go into an unrestricted account to
be used at my discretion to further research in networking/(?)
at SUMEX. You see "cisco wars" have set the networking effort
here at Stanford back on its heels, and I'd like to see it regain
the prominence it rightly deserves here - right Bill Yundt!
-------
∂04-Feb-87 1015 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU Network Software Royalties
Received: from SUMEX-AIM.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 4 Feb 87 10:12:57 PST
Date: Wed 4 Feb 87 10:13:16-PST
From: T. C. Rindfleisch <Rindfleisch@SUMEX-AIM.STANFORD.EDU>
Subject: Network Software Royalties
To: Les@SAIL.STANFORD.EDU
cc: Rindfleisch@SUMEX-AIM.STANFORD.EDU
Message-ID: <12276402543.36.RINDFLEISCH@SUMEX-AIM.STANFORD.EDU>
FYI, Les. I hope we can keep the lid on a possibly polarizing public argument
over this royalty business. Yeager seems to be settling on a more reasonable
tack, at least in private.
Tom R.
---------------
1) 4-Feb Bill Re: LOTS TIP Coordination Meeting
2) 4-Feb To: Yeager@SUMEX-AIM Re: LOTS TIP Coordination Meeting
Message 1 -- ************************
Mail-From: YEAGER created at 4-Feb-87 10:03:03
Date: Wed 4 Feb 87 10:03:02-PST
From: Bill <Yeager@SUMEX-AIM.STANFORD.EDU>
Subject: Re: LOTS TIP Coordination Meeting
To: Rindfleisch@SUMEX-AIM.STANFORD.EDU
In-Reply-To: <12276398047.36.RINDFLEISCH@SUMEX-AIM.STANFORD.EDU>
Message-ID: <12276400681.72.YEAGER@SUMEX-AIM.STANFORD.EDU>
Great! By the way, I have thought over the "royalties" issue and have decided
that an unrestricted account is the ethical path for such monies.
I would like to have some say over how the money is spent though. EG. Maybe
something special for virtual graphics work will be needed (Some neat
workstation that can run X and be used for testing rs232 stuff...?).
DOes this sound reasonable?
Cheers,
Bill
-------
Message 2 -- ************************
Mail-From: RINDFLEISCH created at 4-Feb-87 10:08:14
Date: Wed 4 Feb 87 10:08:13-PST
From: T. C. Rindfleisch <Rindfleisch@SUMEX-AIM.STANFORD.EDU>
Subject: Re: LOTS TIP Coordination Meeting
To: Yeager@SUMEX-AIM.STANFORD.EDU
cc: Rindfleisch@SUMEX-AIM.STANFORD.EDU
In-Reply-To: <12276400681.72.YEAGER@SUMEX-AIM.STANFORD.EDU>
Message-ID: <12276401624.36.RINDFLEISCH@SUMEX-AIM.STANFORD.EDU>
Bill, I think it is a good idea to allocate any CISCO royalty money for network
research and would expect to involve you heavily in decisions about how to use
it.
Tom R.
-------
-------
∂04-Feb-87 1229 LES re: TIP/Gateway Software Royalties
To: Yeager@SUMEX-AIM.STANFORD.EDU
CC: GD.WHY@FORSYTHE.STANFORD.EDU, HANSEN%TALBOTS@SUMEX-AIM.STANFORD.EDU,
RINDFLEISCH@SUMEX-AIM.STANFORD.EDU, CS.EAB@FORSYTHE.STANFORD.EDU,
CS.EJK@FORSYTHE.STANFORD.EDU
[In reply to message from Yeager@SUMEX-AIM.STANFORD.EDU sent Tue 3 Feb 87 21:55:59-PST.]
The policy that royalties on PATENTS are paid to the inventor(s) has been
set by the University. With respect to the marketing of copyrighted
materials and technological knowhow, the policies leave some leeway. My
previous message described what I believe is the appropriate handling in
this case and is consistent with some earlier cases.
Your impression that a PI may place anything that is developed by his
group in the public domain is a widespread misconception -- the written
policies state otherwise. In fact, some PIs have done this and gotten
away with it because the bureaucrats either didn't hear about it or were
too apathetic or fearful to challenge them. I have been grumbling for
years about the inconsistency between "official" policies and reality.
This inconsistency is a reflection of the fact that Stanford is still a
feudal society. For better or for worse, it is unlikely to change in the
near future.
In any case, your substantial contribution to networking at Stanford is
widely recognized and greatly appreciated. You will be rewarded in
heaven, if not sooner.
Les
P.S. I would enjoy talking this over with you at lunch sometime.
In fact, I'll buy.
∂06-Feb-87 1525 HK.JXW@forsythe.stanford.edu
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 6 Feb 87 15:23:37 PST
Received: by lindy.STANFORD.EDU; Fri, 6 Feb 87 15:21:47 PST
Date: Fri, 6 Feb 87 15:20:31 PST
From: Jasper Williams <HK.JXW@forsythe.stanford.edu>
To: LES@sail.stanford.edu
CiscoTo: Cisco Group,
I have discussed the Cisco issues with Iris and Adrian
and have some answers.
In the first place, Les is concerned about how one protects
printed circuit boaqrds. This is not the most clearly defined
issue but we have determined three possible approaches and
suggest that you do all three because of the unsettled nature
of the issue. The three steps are:
-Copyright the mask (this is the approach Les initially
proposed.
-Copyright the program that generates the mask.
-Copyright the software that is used with the board.
By the way what is the status of the conversations with
Andy B? The issue has arisen as to how to treat any derivative
works from the Stanford owned<?> board. If Stanford owns the
original board then it also owns any derivative works unless
it gave a third party the right to produce and own derivative
works. Since we did not give Sun this right we can take the
going in position that we own the Stanford work as well as
derivative work done by Sun.
I have not been able to get in touch with the Brobeck
attorneys that now represent cisco. If Lisa has any conversations
with cisco I would appreciate it if she would lmention that we
have this communication problem.
Let me know if you need any help with the copyright deposits
or have any questions about any of this.
Jasper
To: LES@SAIL
∂06-Feb-87 1825 CS.EJK@forsythe.stanford.edu ROYALTY SHARING POLICY
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 6 Feb 87 18:22:05 PST
Received: by lindy.STANFORD.EDU; Fri, 6 Feb 87 18:20:14 PST
Date: Fri, 6 Feb 87 18:19:23 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: ROYALTY SHARING POLICY
We wanted to respond to the comments which most everyone
has made about the royalty-sharing issue.
Stanford's policy with respect to software is slightly
different than the patent policy.
The policy for software says that if it is developed by staff
"during
the course of employment" and under direction, then generally
we would not share royalties with the developers. However,
individual situations can be different and dictate that
exceptions to this general policy be made.
The director of Technology Licensing, Niels Reimers, has the
final say in whether or not it is appropriate to share royalties
with developers. Of course, we look for guidance from the
individuals involved, particularly the PI, Department Chair,
and the Dean. Some of the guidelines certain departments
use when assessing a particular situation are the number of
developers involved, the past work on the software problem
which may have been done, whether the software was part of
an assigned project, the amount of work done outside of
normal expectations, etc.
The patent policy is different because patents are considered
"fortuitous" and unpredictable. There is no concept of a
"work-for-hire" in the patent case.
We would like to have general agreement on the issue of sharing
royalties with the developers. However, Niels Reimers would have
the final authority on this issue. If we proceed with the concept
of not sharing royalties with the developers then the following
distribution applies.
The 2/3 distrubution outside of the school share should still
reflect the contributions of the developers. If the percentage
contributions outlined in the previous em are correct, then
the 2/3 share after the school share would be distributed
as follows:
EE 15%
CS 5%
SUMEX 80%
We would like to have your comments on this proposal.
Steve, we would like to have your specific comments on the
15% contribution by Kirk Lougheed. Do you agree with this
assessment? It will affect the EE share.
Lisa and Elizabeth
To: GD.WHY, HANSEN@TALBOTS, LES@SAIL, RINDFLEISCH@SUMEX
cc: CS.EAB, CS.EJK
∂07-Feb-87 0045 CS.EJK@forsythe.stanford.edu CISCO NEGOTIATIONS
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 7 Feb 87 00:45:31 PST
Received: by lindy.STANFORD.EDU; Sat, 7 Feb 87 00:43:42 PST
Date: Sat, 7 Feb 87 00:42:36 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: CISCO NEGOTIATIONS
Dear cisco fans:
We had a meeting with Bill Graves from cisco this morning and
the negotiations continue. We have decided to begin keeping
much of the information which is discussed with cisco only
within OTL. This is for a couple of reasons:
We are receiving information from cisco which is confidential
and is needed for us to conclude the license but not necessary
for Stanford otherwise. In addition, Adrian's earlier comments
about exerting care when discussing the cisco situation make
this necessary. We also want to protect certain Stanford
information from getting back to cisco; it appears that some
problems have occurred in this area.
We certainly will continue to consult individuals regarding
specific issues or if we have questions.
Hopefully this will all be wrapped up soon.
Lisa and Elizabeth
To: CS.EJK
To: GD.WHY, HANSEN@TALBOTS, LES@SAIL, RINDFLEISCH@SUMEX
cc: CS.EAB, CS.EJK, HK.NJR
∂09-Feb-87 0824 YEAGER@SUMEX-AIM.STANFORD.EDU Schnurel duart
Received: from SUMEX-AIM.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 9 Feb 87 08:24:21 PST
Date: Mon 9 Feb 87 08:25:43-PST
From: Bill <Yeager@SUMEX-AIM.STANFORD.EDU>
Subject: Schnurel duart
To: les@SAIL.STANFORD.EDU
Message-ID: <12277693684.23.YEAGER@SUMEX-AIM.STANFORD.EDU>
Les, I don't know if you will find it difficult to wedge the information
with respect to cisco using the CSD designed duart(dual-uart) boards from
Bill Graves.
If you do run into problems, I believe we can simply look at one of the
cisco TIPs to see what the board looks like. These TIPs are at tecknowledge
and HP and I know people at both places.
My feeling is the chances are about 99% that Len again walked off with
some Stanford property, and I don't believe he should get away with
it.
Thanks again for lunch and the interesting discussions,
Bill
-------
∂10-Feb-87 1633 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU Re: ROYALTY SHARING POLICY
Received: from SUMEX-AIM.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 10 Feb 87 16:32:56 PST
Date: Tue 10 Feb 87 16:33:53-PST
From: T. C. Rindfleisch <Rindfleisch@SUMEX-AIM.STANFORD.EDU>
Subject: Re: ROYALTY SHARING POLICY
To: CS.EJK@FORSYTHE.STANFORD.EDU
cc: Rindfleisch@SUMEX-AIM.STANFORD.EDU, GD.WHY@FORSYTHE.STANFORD.EDU,
HANSEN%TALBOTS@SUMEX-AIM.STANFORD.EDU, LES@SAIL.STANFORD.EDU,
CS.EAB@FORSYTHE.STANFORD.EDU, CS.EJK@FORSYTHE.STANFORD.EDU
In-Reply-To: Message from "Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>" of Fri 6 Feb 87 18:50:02-PST
Message-ID: <12278044697.24.RINDFLEISCH@SUMEX-AIM.STANFORD.EDU>
Lisa, re your questions on royalty sharing policy:
1) our KSL policy has been that software royalties (e.g., from MAINSAIL,
EMYCIN, AGE, etc.) should be used to support further research in our lab rather
than go to any individuals. We believe that individual remuneration should
come through outside consulting or other authorized involvement in business
ventures relating to commercializing the software. On a similar front, I have
made arguments that the research labs should get both the school and department
2/3 payments in most cases. My argument here is that we get no support from
the university for our work, nor any cut of the 73% overhead our grants and
contracts bring into the university -- this year that amounts to about $4.4M!
Therefore I believe that the schools and departments have taken their cuts
up-front from our work and that the small amount of royalties we get
occasionally should all go toward promoting new work.
2) re the split of royalties between SUMEX, CS, and EE for the network software
CISCO is trying to commercialize, the 80:5:15 division you mention may be OK
but frankly, I have no real basis for making that judgment independently. I
think we should get the players together in a meeting and try to assess the
relative contributions to see it that split makes sense.
Tom R.
-------
∂11-Feb-87 1025 YEAGER@SUMEX-AIM.STANFORD.EDU Duart
Received: from SUMEX-AIM.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 11 Feb 87 10:25:30 PST
Date: Wed 11 Feb 87 10:26:40-PST
From: Bill <Yeager@SUMEX-AIM.STANFORD.EDU>
Subject: Duart
To: les@SAIL.STANFORD.EDU
Message-ID: <12278239991.31.YEAGER@SUMEX-AIM.STANFORD.EDU>
Les, I was thinking that my reference to going elsewhere to uncover what
cisco has in its hardware is really inappropriate given the nature of
the "real world." So, I've flushed that idea.
Off the record though, I really yearn for the kind of things that happened
in the late 60's and early 70's that we were talking about last week. I
have trouble fitting into a particular mold of the 80's because I am
simply too spontaneous. There was once a special closeness and trust
between people and that seems to be going by the wayside. Tom thinks
I am too naive, and he doesn't realize that is a complement! Take
care ... perhaps we can all find a nice hot spring this Summer in
some high-country place where we can feel a tad closer to what we
are really made of ...
Bill
-------
∂13-Feb-87 1217 CS.EJK@forsythe.stanford.edu STATUS OF CISCO NEGOTIATIONS
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 13 Feb 87 12:15:19 PST
Received: by lindy.STANFORD.EDU; Fri, 13 Feb 87 12:13:36 PST
Date: Fri, 13 Feb 87 12:15:41 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: STATUS OF CISCO NEGOTIATIONS
We met with Bill Graves this morning and made considerable
progress toward reaching an agreement. We are very close
to final financial terms and plan to meet on Tuesday to
finalize the contract T's and C's.
We have decided to split the agreement into separate license
agreements covering the various components, such as the MEIS,
Gateway/TIP etc.
We hope to have final agreement on everything next week and
prepare final agreements for signature within the following week.
Lisa and Elizabeth
To: LES@SAIL, GD.WHY, HANSEN@TALBOTS, YEAGER@SUMEX, RINDFLEISCH@SUMEX,
HK.JXW, HK.AXA
cc: CS.EJK, CS.EAB, HK.NJR
∂13-Feb-87 1321 LES re: STATUS OF CISCO NEGOTIATIONS
To: CS.EJK@FORSYTHE.STANFORD.EDU
[In reply to message sent Fri, 13 Feb 87 12:15:41 PST.]
Remember that we probably must reach some kind of agreement with Sun on
the CPU board before a Cisco licensing agreement covering it can be signed.
∂13-Feb-87 1250 CS.EJK@forsythe.stanford.edu Ethernet boards, UART boards
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 13 Feb 87 12:49:59 PST
Received: by lindy.STANFORD.EDU; Fri, 13 Feb 87 12:48:15 PST
Date: Fri, 13 Feb 87 12:49:36 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: Ethernet boards, UART boards
Jasper:
At the cisco meeting this morning, a question of the Ethernet
board payment came up. I mentioned that we wanted to treat the
license of the board separate from the past payment for the
sale to Xerox. (Based on Les Earnest's comments). This is
fine with them.
Bill mentioned that they would probably like to compromise on
the amount owed somewhat; they appear willing to pay some additional
fee (he mentioned splitting the difference) but do not want to pay
the entire $20K profit since they still have warranty costs.
In light of his quite reasonable and businesslike approach on the
other items, I am supportive of a compromise. I told him
that you were handling this matter and his attorney should talk
to you directly.
Please don't mention to their attorney the comments about
compromising which we heard from Bill.
Les, cisco has also agreed to license the UART board from
Stanford! We have worked out terms on this as well as the
other items.
Lisa
To: AS.JXW, LES@SAIL
cc: CS.EJK, CS.EAB
∂13-Feb-87 1556 LES re: Ethernet boards, UART boards
To: CS.EJK@FORSYTHE.STANFORD.EDU
CC: HK.JXW@FORSYTHE.STANFORD.EDU, CS.EAB@FORSYTHE.STANFORD.EDU
[In reply to message sent Fri, 13 Feb 87 12:49:36 PST.]
I am pleased to hear that Cisco has agreed to license the UART board.
Now I wonder why their use of it was not mentioned before we figured it
out and asked. I continue to worry about their integrity. I wonder how
many other Stanford-developed devices and software will surface in future
Cisco products.
As for the idea of "compromise," that has already been done in arriving
at the price of $840 per board. The price should be based on the market
value of the board at the time of sale and has nothing to do with Cisco's
(nonexistent) warranty or any other theoretical services that they might
have performed.
My best estimate of the true market value of the ethernet boards in early
1985 is $1600 each, based on representative industrial cost-based pricing
for comparable volumes. A high-volume 10 megabit ethernet board was sold
by 3COM in that period for $840 in quantity 25 and over (and at higher
prices by others). Inasmuch as high-volume products are cheaper to make
and buy than low-volume products such as the 3 megabit ethernet boards,
the choice of this lower price represents a good faith effort on our part
to settle the issue. $840 is substantially less than the price they would
have had to pay for an equivalent board anywhere else -- in fact there were
no such boards on the market.
When we were discussing the pricing issue, it was Jasper's judgement that
we had a much better chance of settling this matter by accepting a price
close to what Cisco had received rather than demanding the true market
value. [Jasper: I hope that this is an accurate restatement of your
position.] He proposed that we use the 3COM price and I accepted his
view.
If we now get into a negotiation or a shootout, I believe that we should
seek at least $1600 per board. I can arrange for testimony from
disinterested industrial manufacturing experts supporting values in that
range.
To put it bluntly, I believe that offering Cisco a 47% discount on the
settlement of a successful equipment theft constitutes rather liberal
terms.
Les
------- End undelivered message -------
∂17-Feb-87 0908 CS.EJK@forsythe.stanford.edu SUN Boards
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 17 Feb 87 09:08:00 PST
Received: by lindy.STANFORD.EDU; Tue, 17 Feb 87 08:22:26 PST
Date: Tue, 17 Feb 87 08:24:27 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: SUN Boards
Les and Jasper:
Regarding the SUN boards, we do not plan to sign the license (each
of them will be separate, the SUN boards, Gateway/Tip Software,
Ethernet boards, MEIS, and UART) until we settle the issues
with SUN. I don't see why we can't sign the other license
agreements however.
Lisa
To: HK.JXW, LES@SAIL
cc: CS.EAB, CS.EJK
∂18-Feb-87 1610 CS.EJK@forsythe.stanford.edu cisco questions and comments
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 18 Feb 87 16:10:54 PST
Received: by lindy.STANFORD.EDU; Wed, 18 Feb 87 16:09:16 PST
Date: Wed, 18 Feb 87 16:07:14 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: cisco questions and comments
We are in the process of finalizing the agreements to
send to cisco for final review. I would like to receive
your input on a couple of items:
1. Was there any outside funding involved (government or
otherwise) on any of the projects?
2. Since we are splitting the agreements in five separate
license agreements, SUN Boards, Gateway/Tip Software, MEIS,
Ethernet boards and UART boards, the royalty sharing
agreements for each item may be different. I would like
to receive your comments on what groups were involved
with the development of each component. We may need a
meeting to sit down and talk about this. Please let
me have your thoughts.
If anyone has strong feelings about wanting to see the
agreements at this point, please let me know.
Lisa
To: GD.WHY, LES@SAIL, HK.JXW, HANSEN@TALBOTS, YEAGER@SUMEX,
RINDFLEISCH@SUMEX, HK.AXA
cc: CS.EAB, CS.EJK
∂18-Feb-87 1927 LES re: cisco questions and comments
To: CS.EJK@FORSYTHE.STANFORD.EDU
CC: CS.EAB@FORSYTHE.STANFORD.EDU, GD.WHY@FORSYTHE.STANFORD.EDU,
HK.JXW@FORSYTHE.STANFORD.EDU, HK.AXA@FORSYTHE.STANFORD.EDU,
HANSEN@TALBOTS, RINDFLEISCH@SUMEX-AIM.STANFORD.EDU
[In reply to message sent Wed, 18 Feb 87 16:07:14 PST.]
> 1. Was there any outside funding involved (government or
> otherwise) on any of the projects?
Yes. I hope that you don't need to identify all of them.
2. Who did what, as near as I can tell:
SUN board - 100% Computer Science -- It was done principally by
Andy Bechtolsheim, who was an EE PhD student
working in the Artificial Intelligence Lab under
CS contracts.
Gateway/Tip Software - I have little information; you have Bill
Yeager's estimate; I have heard no other views.
Ethernet board - 100% CS, by Bechtolsheim, same as SUN board.
UART Board - 100% CS -- designed by George Schnurle.
MEIS Board - 100% CS -- designed by George Schnurle.
MEIS software - predominently CS, but some other people touched it;
e.g. I think Lougheed did some work.
I will look into this some more.
Cheers,
Les
∂18-Feb-87 1618 HK.JXW@forsythe.stanford.edu SUN Boards
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 18 Feb 87 16:18:21 PST
Received: by lindy.STANFORD.EDU; Wed, 18 Feb 87 16:16:43 PST
Date: Wed, 18 Feb 87 16:18:39 PST
From: Jasper Williams <HK.JXW@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: SUN Boards
REPLY TO 02/17/87 08:21 FROM CS.EJK "Lisa Kuuttila": SUN Boards
Lisa, I agree that it is ok to sign the other licenses--
but not the Sun Board license--as soon as they are ne-
gotiated. Jsper
To: CS.EJK
cc: LES@SAIL, CS.EAB, CS.EJK
∂19-Feb-87 0837 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU re: cisco questions and comments
Received: from SUMEX-AIM.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 19 Feb 87 08:37:27 PST
Date: Thu 19 Feb 87 08:38:31-PST
From: T. C. Rindfleisch <Rindfleisch@SUMEX-AIM.STANFORD.EDU>
Subject: re: cisco questions and comments
To: CS.EJK@FORSYTHE.STANFORD.EDU
cc: CS.EAB@FORSYTHE.STANFORD.EDU, GD.WHY@FORSYTHE.STANFORD.EDU,
HK.JXW@FORSYTHE.STANFORD.EDU, HK.AXA@FORSYTHE.STANFORD.EDU,
HANSEN%TALBOTS@SUMEX-AIM.STANFORD.EDU, Rindfleisch@SUMEX-AIM.STANFORD.EDU,
Les@SAIL.STANFORD.EDU, Yeager@SUMEX-AIM.STANFORD.EDU
In-Reply-To: Message from "Les Earnest <LES@SAIL.STANFORD.EDU>" of Wed 18 Feb 87 19:27:00-PST
Message-ID: <12280317454.29.RINDFLEISCH@SUMEX-AIM.STANFORD.EDU>
Lisa, I agree with the information Les sent you about the various elements in
the CISCO agreement. Funding of Yeager's gateway/TIP software development was
primarily from NIH as part of SUMEX resource research. Also, as I recall, John
Seamons (CSD) was involved in the prom coding for at least early versions of
the SUN board and Ethernet board.
Good luck in wrapping up a deal with CISCO. Tom R.
-------
∂19-Feb-87 0834 CS.EJK@forsythe.stanford.edu
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 19 Feb 87 08:34:22 PST
Received: by lindy.STANFORD.EDU; Thu, 19 Feb 87 08:32:42 PST
Date: Thu, 19 Feb 87 08:34:29 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: les@sail.stanford.edu
Les:
We don't need to specifically identify all of the sponsors
except to the extent that if it is government funded,
we need to have a government rights clause in those contracts.
It sounds like the SUN and Gateway/TIP agreements need these
clauses; how about any of the others? We aslo normally
identify the sponsoring agency in the contract.
If there was funding by a sponsor other that the govnt, we
only need identify them if there were strings attached to
the property rights.
Lisa
∂19-Feb-87 1625 CS.EJK@forsythe.stanford.edu cisco agreements
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 19 Feb 87 16:25:28 PST
Received: by lindy.STANFORD.EDU; Thu, 19 Feb 87 16:23:50 PST
Date: Thu, 19 Feb 87 16:25:45 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: cisco agreements
Les:
Before signing the agreements, I still need descriptions
(not in a lot of detail) for the MEIS, Ethernet, UART
and SUN boards. Steve Hansen sent a description of the
Gateway/TIP software.
Can you help out with this?
Thanks, Lisa
To: LES@SAIL
cc: CS.EAB, CS.EJK
∂20-Feb-87 0803 TOM@Score.Stanford.EDU meis/030
Received: from SCORE.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 20 Feb 87 08:03:33 PST
Date: Fri 20 Feb 87 08:02:11-PST
From: Thomas Dienstbier <TOM@Score.Stanford.EDU>
Subject: meis/030
To: les@Sail.Stanford.EDU
Message-ID: <12280572984.10.TOM@Score.Stanford.EDU>
Tom Harvey left a little after 5 friday and was the last person out and
made sure the door was closed. Marty came bace in around 7:00 and then
left around 10:00. He also made sure the door closed when he left. At both
of these times he was the only one around. When Marty left he was not sure
if the RT was still there. Tom H. was sure it was there because it makes
a larges unpleasant noise when it is running which he did hear at 5:00.
So I think that we can say that the door was indeed locked at the 10:00
time but we don't know whether the RT was there.
Another problem is that we did some tests on the door and that if you let the
door close from a partially open position it will not latch. If the door
is opened all the way it works fine. When Jack Alpert came in sunday morning
the door was not closed completely, and the RT was gone.
So we still do not have a good idea if it was an inside job but I tend
to lean towards that direction. Who new that it was there? There are a lot
of keys out for that room that we know about and god knows how many we don't
know about. I think that we had need to control our keys a little better
so we at least we know for sure who has them and better yet who doesn't.
MEIS software
Diagnostics
David Todd(visiting professor)75%
David Eppstein(cs student)25% with direction from Len Bosack
and Ralph Gorin.
MEIS-Interface to tops20
Kirk Lougheed wrote most of the code with direction
from Len Bosack,and Ralph Gorin.
tom
-------
∂20-Feb-87 1336 TOM@Score.Stanford.EDU re: meis/030
Received: from SCORE.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 20 Feb 87 13:36:04 PST
Date: Fri 20 Feb 87 13:34:38-PST
From: Thomas Dienstbier <TOM@Score.Stanford.EDU>
Subject: re: meis/030
To: LES@Sail.Stanford.EDU
cc: TOM@Score.Stanford.EDU
In-Reply-To: Message from "Les Earnest <LES@SAIL.STANFORD.EDU>" of Fri 20 Feb 87 10:56:00-PST
Message-ID: <12280633505.10.TOM@Score.Stanford.EDU>
After the wave of the hands,,,I would go along with the 50-50 split.
Kirk did alot of the manual inputting of the code and the code itself along
with a few students, but all the ideas came from CSD(Len,Ralph,David Todd)
which I value at a higher rate. I don't even think that EE new that Kirk
was working on this project, but thats another subject.
tom
-------
∂20-Feb-87 1403 LES re: cisco agreements
To: CS.EJK@FORSYTHE.STANFORD.EDU
CC: CS.EAB@FORSYTHE.STANFORD.EDU
[In reply to message sent Thu, 19 Feb 87 08:34:29 PST.]
I believe that there were no non-government contracts involved in any of
the development. I am looking into the question of which source(s)
supported Bechtolsheim's work.
The MEIS and UART boards were done by CSD-CF, which was and is a cost
center that gets most of its income from various government contracts.
I assume that you do not care which ones, given that the money has been
"laundered."
There is quite a bit of handwaving involved in assessing the contributions
to the MEIS software. Overall development responsibility for it rested in
CSD-CF, but Lougheed apparently did quite a bit of work, probably without
the knowledge of his management. I propose a 50-50 split between CS and
EE.
If you are licensing MEIS as an entity (board and software), then we will
need to assess the relative amount of effort in hardware and software in
order to come up with a final split.
Les
∂23-Feb-87 1028 CS.EJK@forsythe.stanford.edu
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 23 Feb 87 10:28:20 PST
Received: by lindy.STANFORD.EDU; Mon, 23 Feb 87 10:26:40 PST
Date: Mon, 23 Feb 87 10:28:34 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: les@sail.stanford.edu
Les:
Bill Graves came in on Friday to discuss a couple of issues.
They apparently have two designs of the UART board. The
first one they believe is the Stanford version; the second
is not. They only want to pay for Version 1. Bill dropped
off a Version 2 board so that we can verify that it is not
a derivative of the Stanford work. Can you look at it and
make that determination?
On the support issues, we only need to know which components
had government support, as we have a general clause
indicating government support. The MEIS hardware and
software are included in one license agreement.
Thanks for all your help with these items!
Lisa
∂23-Feb-87 1507 LES reply to message
To: CS.EJK@FORSYTHE.STANFORD.EDU
[In reply to message sent Mon, 23 Feb 87 10:28:34 PST.]
Yes, we can look over the two UART boards and confirm the claim.
With respect to the division of labor on the MEIS boards and software,
our current best estimates are:
software: 50% EE, 50% CS
hardware: 100% CS
ratio of hardware effort to software effort: 4:1
On this basis, the overall contributions to the MEIS project would be
CS 90%, EE 10%.
Les
∂23-Feb-87 1747 LES "Cisco" UART board
To: CS.EJK@FORSYTHE.STANFORD.EDU
CC: Nilsson@SCORE.STANFORD.EDU
[In reply to message sent Mon, 23 Feb 87 10:28:34 PST.]
A cursory examination of the UART board from Cisco that you sent over this
afternoon reveals the following.
1. The component layout is essentially identical to that of the Stanford
CSD-designed UART board. There appear to be some minor modifications, but
from a copyright standpoint, I believe that it is a derivative work.
2. There is no copyright notice visible, but there is a mark put on by
the board shop that fabricated it. They are the same organization that
was used to make the Stanford UART boards.
3. This "new" board was clearly designed with the SUDS design system,
which runs locally only on SAIL. The work most likely was done by Bosack.
Cisco never had an account on SAIL. Therefore, it appears to have been
designed by a Stanford employee using a Stanford computer. My conclusion
is that the design belongs entirely to Stanford.
4. An addition conclusion is that this is another example of Bosack's use
of Stanford facilities for personal gain, which he knew was illegal.
This brings the known number of stolen designs to 8 distinct boards and
two major software packages. I wonder how many more he still has in his
pocket and how long it will take us to discover them?
For example, I would be willing to bet a large sum that Bosack has a copy
of the SUDS drawing system, which was used to design all of the boards
under discussion. If so, you can bet that he either has, or soon will
get, a DECSystem-20 machine to run it for Cisco use and may later try to
market SUDS, or a derivative version of it, as another Cisco product.
Cheers,
Les
∂26-Feb-87 1927 LES Sun & Ethernet Board Support
To: CS.EJK@FORSYTHE.STANFORD.EDU
[This may be a dupicate message -- I found the template, but no confirmation
that it was sent.]
Andy Bechtolsheim confirms that his work on the SUN workstation, including
the SUN CPU and Ethernet boards, was supported by some government contract
that Forest Baskett had. It is possible that we could figure out which
one, but I guess it doesn't matter.
The only other item that had direct government support, I believe, was
the TIP/gateway software -- Yeager was supported by one or more
government contracts as I recall.
Les
∂26-Feb-87 0051 CS.EJK@forsythe.stanford.edu cisco
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 26 Feb 87 00:50:55 PST
Received: by lindy.STANFORD.EDU; Thu, 26 Feb 87 00:49:20 PST
Date: Wed, 25 Feb 87 21:29:08 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: cisco
We had a visit from Bill Graves this morning. cisco has
determined that someone from Stanford has ordered what
they believe to be a cisco-designed board from their vendor,
Twin Industries. This is a P-20 board which is the processor
board which they say has proprietary cisco changes and
Stanford is not authorized to use. Bill is looking into
the details more closely to find out who did the ordering etc.
I will give you the account of what he knows so far:
The manufacturer of the board itself is Proto. Apparently,
the tooling for this P-20 board was erroneously given to
Stanford by Proto. When Proto realized the error, they asked
Stanford for the toolings back. Stanford refused. Now,
Stanford has approached their board stuffer, Twin, to have
boards made using this design.
cisco is obviously quite concerned about this situation and
has discussed the matter with their lawyer. If these facts are
true, they do not want the individuals involved (after a
settlement is reached) to have access to the cisco code
which we will receive under the license.
Do you know anything about this situation? Bill will give us
some more information on Friday when we meet again. I would
like to have the Stanford perspective at that time.
A second issue which we discussed was the SUN Mic/Andy Bechtolsheim
issue. cisco has met with SUN and worked out a license to the
SUN patent. SUN apparently claimed that they are the sole
owner of the patent and the board which Andy designed. We
should hear back from SUN this week and we will continue to
follow up. However, Bill raised another point with respect to
the SUN board. He claims that Andy gave Len access to his design
and that he believes that Len made changes to the design on
another non-Stanford computer; thus why should they license
the board form us at all? Bill is double checking on whether
Len in fact did make any changes to the board here at Stanford.
Do we have evidence to support this?
I am happy to report that we did reach a resolution of the UART
board issue, Les. I told Bill that we believed it to be a
derivative work and that the fully-paid license we were discussing
did not seem to be worth disputing. He agreed and the matter
is closed.
Bill, cisco also will sell boards to Stanford and Bill Graves
is sending me a copy of their new price list which includes
pricing for the boards. I will explicitly state boards in the
license, also.
We have also asked cisco whether they have a copy of the SUDS
design system - they will have to license if they do.....
I think this is enough of a long rambling message. Please
give us your assistance as soon as possible on the Stanford
purchase of cisco boards issue and any comments on the Len
Bosack changes to the SUN board here at Stanford.
Lisa
To: LES@SAIL, GD.WHY
cc: HK.JXW, HK.NJR, CS.EAB, CS.EJK
∂26-Feb-87 2134 LES re: cisco
To: CS.EJK@FORSYTHE.STANFORD.EDU
CC: GD.WHY@FORSYTHE.STANFORD.EDU, HK.JXW@FORSYTHE.STANFORD.EDU,
HK.NJR@FORSYTHE.STANFORD.EDU, CS.EAB@FORSYTHE.STANFORD.EDU
[In reply to message sent Wed, 25 Feb 87 21:29:08 PST.]
Lisa,
Sad to say, it looks like we are headed for court, rather than the
licensing agreement that we had hoped for. It is becoming increasingly
clear that cisco is negotiating in bad faith. Your report of Wednesday's
discussion with Bill Graves is straight out of Bosack fantasy land.
I learned today that there is still more ugliness in the story.
As I recently confirmed to you, cisco was attempting to conceal their use
of a Stanford-designed UART board. After admitted that, they claimed that
they had recently designed their own version and should not have to pay
royalties on that. Upon closer examination, it was clear that their "new"
board was a trivial modification of the Stanford design that had been done
using our computer-aided design system, called SUDS.
Today I learned from a reliable source that cisco has a DEC-20, confirming
a conjecture in my Feb. 23 message to you. This makes it fairly certain
that Bosack has also taken SUDS and is trying to conceal this fact.
Regarding your discussion with Graves, we have been over the SUN CPU story
before, but I will review the lowlights, in case you have forgotten.
1. The "cisco-designed board" that you refer to is the one that was stolen
from Stanford and given a cisco copyright notice. It was designed at
Stanford on the SAIL computer by a Stanford employee (Bosack) based on an
earlier Sun design that had been made available to Stanford by
Bechtolsheim.
2. The Sun Microsystems design was, in turn, derived from the original SUN
CPU board that was developed at Stanford by Bechtolsheim. It appears that
its use by Sun Microsystems did not conform to the original terms set
forth by Stanford, but that is a side issue.
3. At the time that Bosack walked off with the modified Sun CPU design, he
neglected to leave a copy of the artwork here. I suppose that he expected
us to buy all future copies from cisco.
4. Last Summer, when it became clear what had happened, we went to Proto
Engineering, the board house that had made the earlier batch of boards for
us, and asked for our artwork. They had none filed under "Stanford," but
did have the artwork and drill board filed under "cisco." We asked for
them and got them. We subsequently erased the "cisco" copyright and
applied a Stanford copyright. It was noticed that the artwork for this
board was trivially different from the one that was use in the first batch
(which also erroneously carried a "cisco" copyright).
5. A couple of weeks after we retrieved our artwork (still last Summer),
cisco asked Proto to make some more of these boards. Proto called us. We
talked to Adrian about it, then said "tough luck," knowing that this
actually would not stop cisco, since they still had the original artwork.
Proto did not "realize their error" because there was no error, other than
cisco's.
6. Recently, the Sumex group found that they needed some more of the
new SUN CPU boards, so we loaned them the artwork. They are now getting
them manufactured.
You say that Graves "claims that Andy gave Len access to his design and
that he believes that Len made changes to the design on another
non-Stanford computer; thus why should they license the board form us at
all?" As you should already know, this is nonsense for the following
reasons.
(a) Andy made the design available to Stanford, not to Len and certainly
not to cisco -- indeed, he and the rest of us did not even know that there
*was* a cisco at that time.
(b) The redesign was done on SAIL -- we have the design files to prove it.
Even though Len was Director of our computer facility, he foolishly failed
to set up a cover story by creating a cisco account on SAIL. It is a
little too late for him to now claim that the design modification was done
by cisco.
Cisco had no DEC-20 computer of its own during the period in question and
had no access to a legitimate copy of SUDS anywhere. This makes the
remark about a "non-Stanford computer" very intriguing. As I mentioned
above, there *was* a trivial change in the artwork between the first
version of the new CPU that was built here and the artwork that we
retrieved from Proto. Indeed, following up on that lead has yielded more
sordid details.
From conversations with CSD-CF staff members today, I learned that Len did
quite a bit of consulting at Hewlett-Packard while he was still a Stanford
employee and that H-P has a copy of SUDS. I also learned that Len
apparently facilitated the flow of SUDS updates to and from H-P. However,
the initial acquisition of SUDS by H-P probably predates Len's arrival on
the scene.
Back around 1975, about three years after SUDS was completed, one of my
A.I. Lab. staff members (Ted Panofsky) informed me that he had been
offered consulting fees by a group at H-P if he would make an up-to-date
copy of SUDS and bring it to them. He also remarked that he thought they
had already obtained an earlier version somehow. I confirmed what he
already knew -- that he should not cooperate with them.
Ted also gave me the name of the person who had contacted him, a name that
I have now forgotten. I called this fellow up, threatened him with bodily
harm, and made it clear that if he wanted to use SUDS, he had better take
out a license from Stanford. He made it clear that he would try to avoid
me in the future, but I heard no sense of remorse. Rumors persisted that
H-P continued to use SUDS, but I was unable to confirm it.
It now appears likely that Len somehow linked up with this group, or
perhaps brought a copy of SUDS to a new group at H-P. Anyway, this whole
thing is getting ugly.
Remember that the only thing that cisco voluntarily sought a license for
was the MEIS boards and that happened only after Len had already sold a
bunch of them. Len also crafted a sale of Stanford property to Xerox
that, depending on ones viewpoint, was either an elaborate embezzlement or
a concealed personal loan of $28k.
Every other item seems to have gotten on the agenda by our saying "We
notice that you are selling X, which you obviously swiped from Stanford.
Why don't you let us sell you a license." In this way, we incrementally
took up the MEIS software, the SUN CPU board, the TIP/gateway software,
the ethernet board, the original UART board and the modified one. Will we
now add SUDS to the list and go on negotiating, or should we call a halt
to this fiasco?
When we started this negotiation, a faculty member who was quite familiar
with Len's style argued with me that the ethically correct thing to do and
the cheapest solution in the long run would be to bring legal action aimed
at putting cisco out of business. I said, "Why spend money on punishment
when we can make a little from licensing?" I am afraid that he was right
-- it is time to cut our losses.
Sadly,
Les Earnest
∂27-Feb-87 1240 LES Modified Sun boards
To: CS.EJK@FORSYTHE.STANFORD.EDU, CS.EAB@FORSYTHE.STANFORD.EDU
Tom Dienstbier has informed me that the only difference between the
modified Sun boards that were fabricated here in early 1985 and the ones
that we are now fabricating is that the former had jumper wires to
accomodate various sizes of EPROM memories whereas the current version is
hard-wired for 512k EPROMs. This was the "trivial modification" mentioned
in my last message.
Les
∂03-Mar-87 1039 LES Design Date
To: CS.EJK@FORSYTHE.STANFORD.EDU, HK.AXA@FORSYTHE.STANFORD.EDU,
HK.JXW@FORSYTHE.STANFORD.EDU, HK.NJR@FORSYTHE.STANFORD.EDU,
CS.EAB@FORSYTHE.STANFORD.EDU
Responding to one of Adrian's questions in Monday's meeting, the last
version of the modified SUN CPU board that was developed at Stanford
was completed on June 6, 1986. This was about six months after the
version that was first fabricated here and one month beford Len left
Stanford.
Les
∂27-Feb-87 1429 HK.NJR@forsythe.stanford.edu
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 27 Feb 87 14:28:57 PST
Received: by lindy.STANFORD.EDU; Fri, 27 Feb 87 14:27:01 PST
Date: Fri, 27 Feb 87 14:28:09 PST
From: Sally Hines <HK.NJR@forsythe.stanford.edu>
To: les@sail.stanford.edu
To: CS.EJK
HK.JXW
HK.AXA
LES@SAIL
Re: SUN
I talked to Andy B. They met last friday to discuss the Stanford
situation. the SUN lawyer, Jeff Blatt (213-550-8282), was to call
me. If maybe best for Adrian or Jasper to call him. Let's
discuss at the monday meeting.
Niels
∂03-Mar-87 1055 CS.EJK@forsythe.stanford.edu cisco meeting
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 3 Mar 87 10:55:37 PST
Received: by lindy.STANFORD.EDU; Tue, 3 Mar 87 10:52:35 PST
Date: Tue, 3 Mar 87 10:53:59 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: cisco meeting
We had a good meeting with Bill Graves this morning.
I won't go into all the details about the discussion regarding
the boards but will try to summarize:
cisco is in agreement that the P-19 version of the board (the
version which Len originally gave to the board manufacturer
on behalf of Stanford and cisco) belongs to Stanford. They
agree that our copyright notice should be put on this version
of the board if it has not already been done.
cisco's modifications (which include the change that Les outlined
in his previous em about hardwiring the board for more EPROM
capacity as well as a slight change in the input drivers) are
contained in the version called P-20. This is the version
that Stanford recovered last summer. They believe they
have proprietary rights in this version. Len made these
changes after he left Stanford so it is clear that they do.
We agreed to swap the P-20 artwork which we have for P-19
artwork. Do we have the artwork on campus or does the
board manufacturer have it?
All of thier boxes have been converted to P-20s.
cisco will sell us P-20 boards if we want them. Bill is
sending a price list to us. of course, we would get the
negotiated discount.
Has the manufacturing started ofthe P-20 boards? We have to
decide whether to convert the order or work something out
with cisco for the purchase of P-20s vs. P-19s. What are
your comments.
We agreed to go ahead with the license as is; although we
will not sign until Stanford has reached a resolution with SUN.
I told Bill that we were negotiating with SUN.
Bill verified that they have a copy of SUDS which is the HP
version obtained from HP. We have to decide what to do about
HP; I don't think we can expect cisco to take a license unless
we also go after HP.
I also asked Bill to thoroughly check their technology to see
if there is anything else which should be licensed from
Stanford. He will do that.
He also volunteered to check to see if there was any other
Stanford property on the cisco premises; they will return
anything which is not theirs.
We gave Bill a draft of the latest version of the contract.
He will be out of the country for over a week and will have
their attorney study it in the meantime. It appears we
only have legal T's and C's left. However, given some
of the issues like source code etc. it may take some time
to finally sign the contract.
I feel that Bill Graves is acting in good faith to try to
resolve the outstanding issues. Hopefully this will all
be wrapped up soon.
Lisa
To: LES@SAIL, GD.WHY, RINDFLEISCH@SUMEX, HK.JXW, HK.AXA
cc: CS.EAB, CS.EJK, HK.NJR
∂03-Mar-87 1559 LES re: cisco meeting
To: CS.EJK@FORSYTHE.STANFORD.EDU
CC: GD.WHY@FORSYTHE.STANFORD.EDU, RINDFLEISCH@SUMEX-AIM.STANFORD.EDU,
HK.JXW@FORSYTHE.STANFORD.EDU, HK.AXA@FORSYTHE.STANFORD.EDU,
CS.EAB@FORSYTHE.STANFORD.EDU, HK.NJR@FORSYTHE.STANFORD.EDU
[In reply to message sent Tue, 3 Mar 87 10:53:59 PST.]
I went into Monday's meeting knowing that all of the other participants
would want to find a way to proceed with licensing negotiations with
cisco. Sure enough, you all agreed. I am still willing to be convinced,
but I do not see a way that has a reasonable chance of reaching a fair
conclusion.
The fundamental problem is: how do you negotiate an equitable agreement
with crooks? Please understand that I am not morally opposed to
negotiating with crooks, but I believe that it is essential to have
leverage to enforce such an agreement. The problem here is a practical
one -- an unknown amount of material has already been taken, Len has
consistently made been no admissions until we identify an item and
challenge him, and we have no practical way to verify compliance with the
proposed terms of agreement. When we later discover something else that
he has taken, what can we do about it? Seek criminal charges? Not if the
statute of limitations has run out.
It is fortunate that our computer tape archiving policy has preserved a
record of some of the facts in this case. If we did not have these
records, Len would have gotten away with a lot more.
Incidentally, I am no longer certain that our computer tape records are
completely uncorrupted. After the theft of a $16,000 computer from a
locked room here two weeks ago, we reviewed our key logs and found that a
large number of people had keys to the door in question, one of them being
Len Bosack, who had retained his master key. We asked for it back and he
returned it a week ago.
It is unfortunate that key control is so sloppy in the Computer Science
Department. I don't think that Len had anything to do with the computer
theft, but it is clear that he has had access at any time during the seven
months since he left to all parts of Margaret Jacks Hall, including our
computer room and tape library.
Here are my responses to certain portions of your message.
> We had a good meeting with Bill Graves this morning.
Perhaps I have a bad attitude, but if that was a good meeting, I hope that
you never have a bad one.
> cisco's modifications (which include the change that Les outlined
> in his previous em about hardwiring the board for more EPROM
> capacity as well as a slight change in the input drivers) are
> contained in the version called P-20. This is the version
> that Stanford recovered last summer. They believe they
> have proprietary rights in this version. Len made these
> changes after he left Stanford so it is clear that they do.
I am not convinced of this and, as we already know, Len has been less than
honest in his past pronouncements. I will check the '86 June 6 version of
our SUN CPU board to see if we agree. My bet is that this is another lie.
> We agreed to swap the P-20 artwork which we have for P-19
> artwork. Do we have the artwork on campus or does the
> board manufacturer have it?
Which "we" agreed to this? I certainly did not and, at present, do not
plan to. As I mentioned in my message last week, Len took ALL copies of
the CPU artwork with him when he left. He even retrieved our P-19 version
of the artwork from Proto Engineering. That was why we recovered the only
version that Proto still had. Does any of this bother you?
> cisco will sell us P-20 boards if we want them. Bill is
> sending a price list to us. of course, we would get the
> negotiated discount.
Gee thats awfully nice of them. Perhaps sometime after Hell freezes over.
> Has the manufacturing started of the P-20 boards? We have to
> decide whether to convert the order or work something out
> with cisco for the purchase of P-20s vs. P-19s. What are
> your comments.
Yes, work has started and I do not plan to stop it. Additional boards
are needed now.
> Bill verified that they have a copy of SUDS which is the HP
> version obtained from HP. We have to decide what to do about
> HP; I don't think we can expect cisco to take a license unless
> we also go after HP.
Of course, the version of SUDS that H-P has was brought there by Len. I
would be happy to help coerce H-P into getting a license, but a better
approach might be to seek an indictment of Len and some of the folks at
H-P for conspiring to steal Stanford property. I believe that I can get
Ted Panofsky to testify about the first known attempt by H-P to take SUDS
12 years ago, though it is probably too late for that one.
> I also asked Bill to thoroughly check their technology to see
> if there is anything else which should be licensed from
> Stanford. He will do that.
How will he do that? He will ask Len, of course. And will Len be just
as honest as he has been in the past?
> He also volunteered to check to see if there was any other
> Stanford property on the cisco premises; they will return
> anything which is not theirs.
How will he do that? He will ask Len, of course. And will Len be just as
honest as he has been in the past?
Remember the list in my message last week of things that we have already
caught Len trying to get away with? There was another item that I left
out -- the two multibus high speed communications boards that were bought
by Stanford and taken by Len. After we discovered that they were missing,
traced them to Len, and requested their return, he agreed to do so but
delayed for several months. I suspect that he was using them had to buy
replacements after learning that the theft had been discovered.
You can bet that Len has essentially every piece of source code for
DEC-20s that exists at Stanford. How do we verify that all copies of
these tapes have been returned or destroyed? We can't. Would you like
to take Len's word? I wouldn't.
> I feel that Bill Graves is acting in good faith to try to
> resolve the outstanding issues. Hopefully this will all
> be wrapped up soon.
No amount of good faith on the part of Bill Graves can solve the fundamental
problem. Even though I worked rather hard at getting this negotiation
started, I no longer share your optimism.
Cheers,
Les
∂04-Mar-87 1639 HK.NJR@forsythe.stanford.edu
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 4 Mar 87 16:39:43 PST
Received: by lindy.STANFORD.EDU; Wed, 4 Mar 87 16:38:03 PST
Date: Wed, 4 Mar 87 16:39:31 PST
From: Sally Hines <HK.NJR@forsythe.stanford.edu>
To: les@sail.stanford.edu
RE: Cisco mess
Les:
As I mentioned at the Monday meeting, there appeared to be
3 primary alternatives:
1. Do nothing
2. Go to court
3. Try to make the best of a bad situation
None of the 3 are palatable; the first isn't even digestable.
The second may make us feel good but would accomplish little
else. So that left us with the third course of action, which
Lisa is now following, pursuant to Monday's meeting.
Niels
cc: cs.ejk, gd.why, hk.jxw, cs.eab, hk.axa, rindfleisch@sumex
∂04-Mar-87 1731 LES Cisco Fiasco
To: HK.NJR@FORSYTHE.STANFORD.EDU
CC: cs.ejk@FORSYTHE.STANFORD.EDU, gd.why@FORSYTHE.STANFORD.EDU,
hk.jxw@FORSYTHE.STANFORD.EDU, cs.eab@FORSYTHE.STANFORD.EDU,
hk.axa@FORSYTHE.STANFORD.EDU, rindfleisch@SUMEX-AIM.STANFORD.EDU
[In reply to message sent Wed, 4 Mar 87 16:39:31 PST.]
Niels:
The point of my message was that the way things are structured, our
attempt at negotiating a license will most likely be a transition stage to
going after cisco in court. Either that or we wimp out. If the latter,
you can wave goodbye to a lot more intellectual property. In fact, I will
probably set up another business or two myself. :-)
Les
∂04-Mar-87 1837 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU Re: Cisco Mess
Received: from SUMEX-AIM.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 4 Mar 87 18:37:43 PST
Date: Wed 4 Mar 87 18:36:40-PST
From: TC Rindfleisch <Rindfleisch@SUMEX-AIM.STANFORD.EDU>
Subject: Re: Cisco Mess
To: HK.NJR@FORSYTHE.STANFORD.EDU
cc: Rindfleisch@SUMEX-AIM.STANFORD.EDU, cs.ejk@FORSYTHE.STANFORD.EDU,
gd.why@FORSYTHE.STANFORD.EDU, hk.jxw@FORSYTHE.STANFORD.EDU,
cs.eab@FORSYTHE.STANFORD.EDU, hk.axa@FORSYTHE.STANFORD.EDU,
Les@SAIL.STANFORD.EDU, Devaney@SUMEX-AIM.STANFORD.EDU
In-Reply-To: Message from "Sally Hines <HK.NJR@forsythe.stanford.edu>" of Wed 4 Mar 87 16:40:05-PST
Message-ID: <12283834217.77.RINDFLEISCH@SUMEX-AIM.STANFORD.EDU>
Friends, my fingers balk as I type this but I'm afraid I agree with Niels that
the most sensible approach is to make the best of a bad situation in working
through the Cisco relationship. Several comments seem pertinent:
1) There is so much evidence of concealment, bad faith, or worse on the part of
some Cisco principals that we should not count on preserving a long term
working relationship. For example, I expect that they will not be particularly
forthcoming about declarations relating to royalty obligations and that they
will "modify" or "develop revolutionary new models of" their products quickly
so as to be able to claim they are no longer using Stanford technology. This
suggests collecting as much as we possibly can up front (as a lump cash
payment, free copies of Cisco hardware and software products for those who
still want to use them, etc.) for the properties they are using rather than
expecting to accumulate royalties over time.
2) We have no real way of knowing what else Cisco participants have taken
without authorization from Stanford, including hardware parts or subsystems,
Stanford software (e.g., SUDS), software Stanford has licensed from others
(e.g., DEC software like TOPS-20 sources and subsystems, UNIX, C compilers,
1022 data base system, NCPCALC, etc.). We should make it clear that the
current agreement does not prevent Stanford from seeking restoration with
respect to undisclosed properties Cisco has taken directly or indirectly from
Stanford, ask Cisco explicitly to assure us they have destroyed all software
taken from Stanford without authorization, and ask them to be willing to
indemnify us against suits arising from unauthorized software use.
3) Perhaps most importantly, what can we do to minimize the likelihood of this
sort of thing happening again? Besides possible property losses, I fear the
Cisco experience has done unseen damage to Stanford in the form of creating
inhibitions against sharing ideas, information, and developments with possible
commercial value among our groups which have need to benefit from each others
work. This topic is much too long to address here but I think we should follow
up with discussions about what we can learn about relevant policies,
incentives, and management procedures.
Tom R.
-------
∂05-Mar-87 1407 LES CPU Version P-20
To: cs.ejk@FORSYTHE.STANFORD.EDU
CC: HK.NJR@FORSYTHE.STANFORD.EDU, gd.why@FORSYTHE.STANFORD.EDU,
hk.jxw@FORSYTHE.STANFORD.EDU, cs.eab@FORSYTHE.STANFORD.EDU,
hk.axa@FORSYTHE.STANFORD.EDU, rindfleisch@SUMEX-AIM.STANFORD.EDU
We have just finished reviewing the CPU design files from SAIL archival
tapes and find that version P-20 is there. The only difference we can find
between P-19 and P-20 is one wire. This confirms that Len lied again.
(Sigh.)
Les
Xerox want license for 3 Mbit Ethernet board
∂10-Mar-87 1758 CS.EAB@forsythe.stanford.edu
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 10 Mar 87 17:58:40 PST
Received: by lindy.STANFORD.EDU; Tue, 10 Mar 87 17:57:12 PST
Date: Tue, 10 Mar 87 17:58:22 PST
From: Elizabeth A. Batson <CS.EAB@forsythe.stanford.edu>
To: les@sail.stanford.edu
Michael Levitt of Systems Concepts called to say that
they are still very interested licensing the
Gateway/TIP software from Stanford. He would
like an opportunity to visit Stanford and see a demo
of the software. Who would be the best person for
him to talk to? Also, have you been making improvements
since Bosack/Lougheed left? I assume that we will be
willing to license the most recent version.
Elizabeth
To: Bill Yundt(GD.WHY), Tom Rindfleisch(RINDFLEISCH@SUMEX-AIM),
YAEGER@SUMEX.STANFORD, LES@SAIL.STANFORD
cc: CS.EJK, CS.EAB
∂24-Mar-87 1353 CS.EJK@forsythe.stanford.edu Sun Status
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 24 Mar 87 13:52:58 PST
Received: by lindy.STANFORD.EDU; Tue, 24 Mar 87 13:54:26 PST
Date: Tue, 24 Mar 87 13:52:20 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: LES@sail.stanford.edu
Subject: Sun Status
Les:
Things are progressing slowly with respect to Sun. Jasper
has been in contact with their attorney and a meeting is
set up for Thursday April 16 at 1:30PM in Jasper's office.
Niels, I, you and Jasper will atend on behalf of Stanford
and Jeff Blatt, hopefully Andy Bechtolsheim and at least
one Sun management person will attend on behalf of Sun.
There is also a pre-meeting scheduled for Wednesday April 15 at
2:00PM at our office to go over our strategy.
Please let me know if both meetings are ok for you.
There is also a letter which will go out to Sun this week
confirming what we have told Sun verbally.
I have alos dreafted a memo confirming my conversation with
Bill Osborn who was handling this case on behalf of our
office at the time Andy disclosed the invention. He is very
supportive of our position.
Lisa
To: LES@SAIL
cc: HK.NJR, HK.JXW, HK.AXA, CS.EJK
∂31-Mar-87 0848 CS.EJK@forsythe.stanford.edu
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 31 Mar 87 08:47:58 PST
Received: by lindy.STANFORD.EDU; Tue, 31 Mar 87 08:49:37 PST
Date: Tue, 31 Mar 87 08:50:52 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: les@sail.stanford.edu
Les:
Thanks for tracking down that sponsorship issue on Andy
Bechtolsheim's work. When you talked to Forrest Baskett
did you ask him what his feelings were about this
situation, especially with respect to Andy not reporting
that he had filed a patent application?
Lisa
∂31-Mar-87 1140 LES Baskett's Feelings
To: CS.EJK@FORSYTHE.STANFORD.EDU
[In reply to message sent Tue, 31 Mar 87 08:50:52 PST.]
No, there was no opportunity to discuss this matter in depth. I caught
Forest just after he finished a lecture and there were a bunch of other
people around. He asked what this was about and I explained it briefly.
I tried to reach him by phone this morning to poke a bit more but was told
that he will not be in till tomorrow. I left a message.
You memo about Bill Osborn's earlier discussions with the Sun folks is
quite interesting. Until I saw that, I assumed that Andy was simply
forgetful in not notifying OTL about the patent filing. It now appears
that he too lied. I guess that I should learn to be more cynical.
Les
∂31-Mar-87 1624 LES Cisco Ethernet Board
To: cs.ejk@FORSYTHE.STANFORD.EDU, hk.jxw@FORSYTHE.STANFORD.EDU
I just chased down the rumor that Cisco is offering the 3 Megabit ethernet
for sale and learned from Bill Winfield of Xerox PARC (415 494-4375) that
two weeks ago Cisco offered to sell these boards for $895 each in
quantities of 25 or more with delivery in 4 weeks ARO. Given that the
parts cost was substantially higher two years ago, I guess that we ought
to increase our price to Cisco to around $1600 per board, as I suggested
in the first place.
Winfield also remarked that Cisco (I think Graves) told him that they hold
a Stanford license on the board. (!%#@&!) He was rather upset when I
mentioned that no license agreement has been signed yet. He is apparently
planning a time-critical project that depends on the availability of those
boards. He asked to be notified as soon as the agreement is completed. I
agreed to do that.
Les
∂02-Apr-87 0852 CS.EJK@forsythe.stanford.edu Ethernet boards: cisco
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 2 Apr 87 08:52:38 PST
Received: by lindy.STANFORD.EDU; Thu, 2 Apr 87 08:54:17 PST
Date: Thu, 2 Apr 87 08:55:20 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: Ethernet boards: cisco
Les:
Bill Graves did ask us a few weeks ago whether he could
say that they were close to signing a license on the Ethernet
boards to Xerox. Since we have agreed on the financial
terms, we said ok.
I think we should have cisco sign this license as well as
the UART board license and the MEIS license. Do you have
any problems, Jasper, if we sign these three agreements
before signing the Gateway software (Elizabeth is still
working with Bill Graves on this one) and the SUN board
(which we cannot sign until the Sun situation is resolved)?
I would like to clean the first three up!!!
Lisa
To: LES@SAIL
cc: HK.JXW, CS.EAB, CS.EJK
∂02-Apr-87 1216 LES re: Ethernet boards: cisco
To: CS.EJK@FORSYTHE.STANFORD.EDU
CC: HK.JXW@FORSYTHE.STANFORD.EDU, CS.EAB@FORSYTHE.STANFORD.EDU
[In reply to message sent Thu, 2 Apr 87 08:55:20 PST.]
I believe that we should sign no licensing agreements with cisco until
they have paid for their "acquisition" of Stanford ethernet boards two
years ago. To do so would constitute an acknowledgement that it is OK to
steal from Stanford and that such acts do not prejudice us against having
other business dealings with the thieves.
Les
∂02-Apr-87 1235 HK.JXW@forsythe.stanford.edu Ethernet boards: cisco
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 2 Apr 87 12:35:09 PST
Received: by lindy.STANFORD.EDU; Thu, 2 Apr 87 12:36:54 PST
Date: Thu, 2 Apr 87 12:38:11 PST
From: Jasper Williams <HK.JXW@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: Ethernet boards: cisco
REPLY TO 04/02/87 08:54 FROM CS.EJK "Lisa Kuuttila": Ethernet boards: cisco
Lisa, I think we should resolve the matter of the already
sold Ethernet boards before signing a new license or at least
find out their position. As I told you, their lawyer canceled
last week's meeting. Jasper
To: CS.EJK
cc: CS.EAB, LES@SAIL, Lisa Kuuttila(CS.EJK)
∂02-Apr-87 1458 CS.EJK@forsythe.stanford.edu Ethernet boards
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 2 Apr 87 14:57:57 PST
Received: by lindy.STANFORD.EDU; Thu, 2 Apr 87 14:59:39 PST
Date: Thu, 2 Apr 87 15:00:50 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: Ethernet boards
Les and Jasper:
Ok, I got your message....... we won't license the Ethernet
boards until the "sale" of Ethernet boards is completed.
I guess I am just getting anxious to close some of these
agreements. Has the meeting with cisco to discuss the
boards been rescheduled?
Lisa
To: LES@SAIL, HK.JXW
cc: CS.EAB, CS.EJK
∂03-Apr-87 0857 HK.JXW@forsythe.stanford.edu Ethernet boards
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 3 Apr 87 08:57:33 PST
Received: by lindy.STANFORD.EDU; Fri, 3 Apr 87 08:59:15 PST
Date: Fri, 3 Apr 87 09:00:25 PST
From: Jasper Williams <HK.JXW@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: Ethernet boards
REPLY TO 04/02/87 15:00 FROM CS.EJK "Lisa Kuuttila": Ethernet boards
Lisa, Yes, the earliest they wanted to reschedule for was
next Thursday at 2 pm. Will discuss with you and Les once
I talk to Jeff Kignston of Brobeck. Jasper
To: CS.EJK
cc: CS.EAB, LES@SAIL
∂10-Apr-87 0847 CS.EJK@forsythe.stanford.edu sun
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 10 Apr 87 08:47:42 PDT
Received: by lindy.STANFORD.EDU; Fri, 10 Apr 87 08:49:03 PST
Date: Fri, 10 Apr 87 08:49:40 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: sun
Les:
Yes, I would like to see a copy of the license between
Imagen and VLSI Systems.
I also have gotten a copy of the agreement between Sun
and cisco for the patent. It is interesting because
sales made to Stanford by cisco are royalty-free. If
anyone wants a copy, please let me know.
I will be interested to hear what Forest Baskett thinks
of the patent.
Lisa
To: LES@SAIL
cc: HK.NJR, HK.JXW, HK.AXA, CS.EJK
∂11-Apr-87 1232 HK.JXW@forsythe.stanford.edu ethernet
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 11 Apr 87 12:32:30 PDT
Received: by lindy.STANFORD.EDU; Sat, 11 Apr 87 12:32:46 PST
Date: Sat, 11 Apr 87 12:34:01 PST
From: Jasper Williams <HK.JXW@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: ethernet
Les,
I met with cisco with regard to the ethernet boards on
Thursday. The lawyer and Bill Graves came in and offered to pay
$5000, aparently the upfront license for future boards, to resolve
the matter. I explained that I would report this to you but could not
not think that they were serious. They did say that cisco incurred
expenses of overhead, warranty, and the cost of burning the
boards. I replied tdhat if they could show actual out of pocket
costs associated with the boards then I would discuss an offset
of those costs with you. But I cautioned that overhead and warranty
would be difficult if not impossible for us to accept. The cost
of burning the boards does seem like a legitimate incremental
expense. I do need some information from you on this since I
don't quite understand what burning is. Furthermore, is it true
that cisco in fact burned the boards rather than Stanford?
If so could you check as to what would be a reasonable cost for
burning this quantity of boards? Finally do you agree that if
cisco in fact burned the boards this should be an offset? --I
hope you do. They are supposed to check their costs and get back
to me. I guess their strategy was to come in with an outrageous
number--$5000 and then next time they are suposed to have
seemed to moved when they double it to a starting number like
$10,000. Jasper
To: LES@SAIL, Lisa Kuuttila(CS.EJK)
∂11-Apr-87 2311 LES re: ethernet
To: HK.JXW@FORSYTHE.STANFORD.EDU
CC: CS.EJK@FORSYTHE.STANFORD.EDU
[In reply to message sent Sat, 11 Apr 87 12:34:01 PST.]
Burn-in can be (and often is) as simple as powering up the board and
leaving it that way for a day or two. It is intended to expose any infant
mortalities in the components. For higher reliablility, burn-in might
consist of power and temperature cycling in an oven while performing
operating tests, in order to stress the board more and expose marginal
components. The more rigorous kind of burn-in is usually used only in
applications requiring very high reliability, such as for certain military
equipment.
The 3 megabit ethernet boards that Bosack took were almost certainly
burned-in at Stanford. Remember that they were produced as part of
a larger batch of such boards. I will ask Tom Dienstbier for his
recollection of what happened.
If cicso can prove that they did the burn-in, I would offer them 2 cents
per board off the $1600/board price for the electrical power that would
have been consumed.
As for their offer, let me say that if I were there I would have been
inclined to throw them out the window and into the dumpster near the
corner of your building.
Les
∂12-Apr-87 2135 HK.JXW@forsythe.stanford.edu re: ethernet
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 12 Apr 87 21:35:13 PDT
Received: by lindy.STANFORD.EDU; Sun, 12 Apr 87 21:35:33 PST
Date: Sun, 12 Apr 87 21:36:32 PST
From: Jasper Williams <HK.JXW@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: re: ethernet
REPLY TO 04/11/87 23:13 FROM LES@SAIL.STANFORD.EDU "Les Earnest": re: ethernet
Les, The thought of throwing them out the window crossed
my mind but I suppressed the urge. Thanks for the info
and let me know what Tom recollects. Jasper
To: LES@SAIL.STANFORD.EDU
∂17-Apr-87 1657 LES Ethernet Board Burn-in
To: HK.JXW@FORSYTHE.STANFORD.EDU
CC: CS.EJK@FORSYTHE.STANFORD.EDU
Tom Dienstbier, who was directly involved in the fabrication and testing
of the 3 megabit ethernet boards, says that they were all burned-in and
tested here.
Les
∂21-Apr-87 1122 CS.EJK@forsythe.stanford.edu cisco license negotiations
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 21 Apr 87 11:22:00 PDT
Received: by lindy.STANFORD.EDU; Tue, 21 Apr 87 11:23:18 PST
Date: Tue, 21 Apr 87 11:23:12 PST
From: Lisa Kuuttila <CS.EJK@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: cisco license negotiations
Dear folks:
Just wanted to let you know that we signed 3 license agreements
with cisco yesterday, for the Tip/Gateway software, the UART
board, and the MEIS technology. The UART and MEIS agreements
were fully paid at $15,000 for both. The software agreement
was $19,300 up-front with earned royalties based on sales.
There is also a provision for equipment discounts as well as
the software updates which will be provided to Stanford.
Elizabeth will be providing more details about the nature
of the discount.
We still have two outstanding agreements to conclude with
cisco: the Ethernet board as well as the Sun board agreements.
Thank you all for the support you have provided to us during the
negotiation!
Lisa
To: LES@SAIL, HK.JXW, HK.AXA, RINDFLEISCH@SUMEX-AIM, YAEGER@SUMEX, GD.WHY,
HANSEN@SIERRA
cc: HK.PLD, NILSSON@SCORE, CS.EAB, CS.EJK
∂21-Apr-87 1352 LES re: cisco license negotiations
To: CS.EJK@FORSYTHE.STANFORD.EDU
CC: HK.JXW@FORSYTHE.STANFORD.EDU, HK.AXA@FORSYTHE.STANFORD.EDU,
RINDFLEISCH@SUMEX-AIM.STANFORD.EDU, GD.WHY@FORSYTHE.STANFORD.EDU,
HANSEN@SIERRA.STANFORD.EDU, HK.PLD@FORSYTHE.STANFORD.EDU,
NILSSON@SCORE.STANFORD.EDU, CS.EAB@FORSYTHE.STANFORD.EDU
[In reply to message sent Tue, 21 Apr 87 11:23:12 PST.]
Thank you for telling us. As I made clear to you repeatedly, including
last week, I believe that signing a licensing agreement with cisco before
we have settled the dispute over the ethernet boards that were taken from
Stanford is a mistake. I hope that the rest of the issues can be resolved
soon, but I am less certain that this will happen than I was before you
signed the agreement. -Les
∂22-Apr-87 1155 GOTELLI@Score.Stanford.EDU Cisco Computer Account
Received: from SCORE.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 22 Apr 87 11:55:00 PDT
Date: Wed 22 Apr 87 11:24:43-PDT
From: Lynn Gotelli <GOTELLI@Score.Stanford.EDU>
Subject: Cisco Computer Account
To: Les@Sail.Stanford.EDU
cc: Gotelli@Score.Stanford.EDU
Message-ID: <12296589716.11.GOTELLI@Score.Stanford.EDU>
Les, Late yesterday Bill Graves, CEO for Cisco called me regarding
the balance due on the <CSC@Score> computer account. He said
that since the account was frozen on 9/30/86 he does not feel
Cisco should have to pay disk charges for the months of Oct,
Nov, Dec and Jan. when they did not have access to the account.
Will you agree to accept payment in full from Cisco for charges
incurred prior to 9/30/86 ? If so, I will call and let him
know your decision and send him a revised invoice. Thanks, Lynn
Since CISCO was being billed every month they could have paid up
soon after the account was frozen.
Thank you, Lynn
-------
∂18-May-87 1858 LES Schnurle Letter
To: CS.EJK@FORSYTHE.STANFORD.EDU
Having received a copy of George Schnurle's letter of May 11 today
and having discussed it with Ralph Gorin, who was his supervisor
when the MEIS project started, I still do not agree with George.
Ralph acknowledges that he probably did discuss the potential
marketability of the MEIS scheme with Schnurle and that he also probably
discussed the Stanford policy on licensing inventions. He does not
recall linking these two subjects in a single discussion.
My view is that neither the MEIS boards nor the MEIS software contain an
"invention." Together they made it possible for a computer with a DEC
Massbus to communicate with other computers via Ethernet. While the
design task involved a great deal of work, it simply provided a connection
between two well-defined standard communication interfaces and required no
innovations.
For what it is worth, I believe that the Computer Science Department spent
far more on the development of this product (perhaps by an order of
magnitude) than it saved by avoiding the purchase of similar commercial
products or will ever recover through licensing. In other words, the MEIS
project should never have been undertaken. (I tried to convince Ralph of
this in advance of the undertaking, in 1980, but his "receiver" was turned
off.) All other issues aside, it doesn't make sense to me to offer
royalties on a product that cost us far more than we will ever get back.
Of course, it would be nice if Stanford had a coherent intellectual
property policy so that staff members would know where they stand and we
could avoid these recurring debates.
Les
∂27-Jul-87 1258 HK.JXW@forsythe.stanford.edu Returned mail: User unknown
From: Jasper Williams <HK.JXW@Forsythe.Stanford.EDU>
To: lesearnest@SAIL.STANFORD.EDU
Subject: cisco
cisco folks,
After discussions with Bill Yundt and Les Earnest, on
yesterday I made an offer to Jeff Kingston with regard to the
ehternet boards that cisco-Len Bosack sold to Xerox. I offered
to settle this part of the dispute for the offered $7500 and
two Gateway royters with ethernet interfacing. Mr. Kingston
said he would discuss it with cisco and get back to me. I will
let you know when I hear from him.
Jasper
To: Bill Yundt(GD.WHY), LESEARNEST@SAIL, Lisa Kuuttila(CS.EJK), Adrian
Arima(HK.AXA), Elizabeth Batson(CS.EAB)
∂06-Aug-87 1236 HK.JXW@forsythe.stanford.edu cisco
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 6 Aug 87 12:36:52 PDT
Received: by lindy.stanford.edu; Thu, 6 Aug 87 12:36:37 PDT
Date: Thu, 6 Aug 87 12:35:24 PDT
From: Jasper Williams <HK.JXW@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: cisco
cisco fans,
I just got a counteroffer from cisco to our offer. We
offered to take $7500 plus 2 gateway royters with ethernet
facing. The value of this offer to Stanford was about $15,000.
cisco's counteroffer is that they will give Stanford $18,000 of
equipment, without a discount. Since the discount is 20% the value
of the counteroffer is $14,400.
I recommend that we accept this offer. It is not quite what
we initially demanded but the dollar values are close enough to
our last offer that I am comfortable with it. It also is quite
likely that the actual value of the equipment will come within
a few hundred more or less of $18,000 so the numbers may be
closer to the $15,00 than $14,400. But basically if we would
accept $15,00 then it is not worth our while to go to court over
$600.
There are two issues that Bill and Les will have to work
out. The first is whether they can work out the internal
split of the consideration in merchandise. The second is that
in fact there is $18k of cisco equipment that we want at this time.
I have no solutions on these issues.
If these two internal issues are resolved then Bill should
decide on the equipment that he wants and I will document the deal
with a brief letter. I told cisco's attorney that the first item
of internal allocation might be a problem and that if we were willing
to accept an all equipment settlement then the next contact would
be from Bill Yundt to cisco to work specify the equipment.
I hope that this offer works for everyone. Let me have
your responses. Jasper
To: Lisa Kuuttila(CS.EJK), LES@SAIL, Bill Yundt(GD.WHY), HK.AXA
∂07-Aug-87 1600 LES re: cisco
To: HK.JXW@FORSYTHE.STANFORD.EDU
CC: CS.EJK@FORSYTHE.STANFORD.EDU, GD.WHY@FORSYTHE.STANFORD.EDU,
HK.AXA@FORSYTHE.STANFORD.EDU
[In reply to message sent Thu, 6 Aug 87 12:35:24 PDT.]
Bill Yundt indicates that he would like to buy at least $14-15k worth
of cisco systems and, being a cheapskate, is willing to give Computer
Science just $10k in cash for the privilege. Overcoming moderate
moral indignation about the whole situation, I'll buy it.
Let's try to close this deal and put it behind us. If cisco pulls any
more tricks, however, I will advocate giving them a public bashing.
Cheers,
Les
∂10-Aug-87 1109 HK.JXW@forsythe.stanford.edu re: cisco
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 10 Aug 87 11:09:46 PDT
Received: by lindy.stanford.edu; Mon, 10 Aug 87 11:09:39 PDT
Date: Mon, 10 Aug 87 11:08:37 PDT
From: Jasper Williams <HK.JXW@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: re: cisco
REPLY TO 08/07/87 16:00 FROM LES@SAIL.STANFORD.EDU "Les Earnest": re: cisco
Cheers! JXW
To: LES@SAIL.STANFORD.EDU
cc: Bill Yundt(GD.WHY)
∂11-Aug-87 1020 TOM@Score.Stanford.EDU CISCO
Received: from SCORE.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 11 Aug 87 10:20:13 PDT
Date: Tue 11 Aug 87 10:19:32-PDT
From: Thomas Dienstbier <TOM@Score.Stanford.EDU>
Subject: CISCO
To: les@Sail.Stanford.EDU
cc: gotelli@Score.Stanford.EDU, bscott@Score.Stanford.EDU,
ball@Score.Stanford.EDU
Message-ID: <12325675834.26.TOM@Score.Stanford.EDU>
I heard from the grape vine that the latest CISCO agreement has been
agreeded upon; thats good news. Also, that networking is paying us
some amount of money for the equipment that is part of the agreement.
I presume that 078 will receive this money since it is the account
that incumbered the expenses at the time of development of the product/s.
tom
-------
∂13-Aug-87 1105 GD.WHY@forsythe.stanford.edu cisco
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 13 Aug 87 11:05:49 PDT
Received: by lindy.stanford.edu; Thu, 13 Aug 87 11:05:40 PDT
Date: Thu, 13 Aug 87 11:04:13 PDT
From: Bill Yundt <GD.WHY@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: cisco
REPLY TO 08/06/87 12:35 FROM HK.JXW "Jasper Williams": cisco
Jasper,
I have spoken to Bill Graves at cisco about the equipment
we will accept in settlement and have transmitted an
expense transfer to Les Earnest in the amount of
$10,000 to a designated Computer Science Department
account to take care of the internal transaction.
For your information, the equipment that cisco will
deliver is:
2 each cisco Model AGS-2E Two Segment Ethernet Gateways
List Price - $8998.00 each Total List Value = $17,996.00.
I assume this is all the information you will need to complete
the settlement documentation.
Thanks very much for your efforts.
.....Bill
To: HK.JXW
cc: CS.EJK, LES@SAIL, HK.AXA, GD.CLB
∂19-Aug-87 1707 GD.WHY@forsythe.stanford.edu cisco settlement and related request
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 19 Aug 87 17:07:24 PDT
Received: by lindy.stanford.edu; Wed, 19 Aug 87 17:07:26 PDT
Date: Wed, 19 Aug 87 17:06:15 PDT
From: Bill Yundt <GD.WHY@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: cisco settlement and related request
Jasper,
Your draft looks fine to me. The date for delivery of
the gateway boxes should be September 21. We would
like that much time to permit them to supply the
version with the new processor board if possible.
Jasper, because of the long delay in bringing this
negotiation to a conclusion, we have only a short time
left to exercise our right under the agreements
negotiated by OTL to purchase cisco products at a 30%
discount instead of the lower (23%) level provided for
on a continuing basis. If I recall correctly, the
agreements provide that we have the right to place two
orders within 6 months of the effective date (April 20,
1987) at the more favorable discount. That gives us
only until October 20. I would appreciate it if
someone could see if cisco is willing to extend that
date to six months from effective date of the
settlement agreement, in view of the fact that the
original term for that provision was predicated on the
belief that all matters would be cleared up in April
and we have effectively lost the intervening 4 months
so have had no opportunity to test the cisco systems in
the Stanford network (purpose of first order) to
determine whether or not to buy larger quantities
(second order).
Are you the right person to explore this possibility or
should Lisa or Elizabeth look into it?
Bill
To: Jasper Williams(HK.JXW), LES@SAIL
∂20-Aug-87 0856 HK.JXW@forsythe.stanford.edu cisco settlement and related request
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 20 Aug 87 08:56:08 PDT
Received: by lindy.stanford.edu; Thu, 20 Aug 87 08:56:10 PDT
Date: Thu, 20 Aug 87 08:54:48 PDT
From: Jasper Williams <HK.JXW@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: cisco settlement and related request
REPLY TO 08/19/87 17:06 FROM GD.WHY "Bill Yundt": cisco settlement and related
request
Bill, If you want and if you give me all the necessary
information (ie is the discount agreement written, or is
there other info in addition to this) then I can put it
into this letter. I will forward a coppy of this em to
Elizabeth so that she can let us know if she would prefer
to handle it or has some other ideas. Which one of you
had the discussions with respect to the discount anyway?
Let me know what you want to do. I'd like to get the
draft out by tomorrow. Jasper
To: GD.WHY
cc: LES@SAIL, Elizabeth Batson(CS.EAB)
∂27-Oct-87 0850 AR.PMB@forsythe.stanford.edu Future Business With Cisco Systems
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 27 Oct 87 08:50:09 PST
Received: by lindy.stanford.edu; Tue, 27 Oct 87 08:48:49 PST
Date: Tue, 27 Oct 87 08:49:39 PST
From: Pat Burns <AR.PMB@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: Future Business With Cisco Systems
Procurement has recently received inquiries into doing
business with a company called Cisco Systems. As you recall
about a year ago we discovered that this company was owned
and operated by a Stanford employee named Leonard Bosack.
On advise from the Stanford legal department we stopped
doing business with him.
Confirming our conversation of Monday, October 26,
it is now understood that Stanford may again place purchase
orders with this company. I would appreciate hearing from
anyone if they disagree with this finding.
To: HK.PLD
cc: AR.JHG, LES@SAIL, HK.JXW
∂27-Oct-87 1141 LES re: Future Business With Cisco Systems
To: AR.PMB@FORSYTHE.STANFORD.EDU
[In reply to message sent Tue, 27 Oct 87 08:49:39 PST.]
Yes, the problems with Cisco were eventually resolved through negotiation.
Be careful with those people, though.
∂27-Oct-87 1304 HK.JXW@forsythe.stanford.edu Future Business With Cisco Systems
Received: from FORSYTHE.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 27 Oct 87 13:04:34 PST
Received: by lindy.stanford.edu; Tue, 27 Oct 87 13:03:09 PST
Date: Tue, 27 Oct 87 12:49:43 PST
From: Jasper Williams <HK.JXW@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: Future Business With Cisco Systems
REPLY TO 10/27/87 08:49 FROM AR.PMB "Pat Burns": Future Business With Cisco
Systems
Pat, We have resolved a good portion of the cisco problem and
expect to sign agreements documenting our agreement this week.
I will let you know if this in fact happens. If they sign the
documents that we have outstanding, then I have no problem
with our resuming doing business with cisco. Jasper
To: AR.PMB
cc: HK.PLD, AR.JHG, LES@SAIL
∂28-Oct-87 0552 HK.JXW@forsythe.stanford.edu Future Business With Cisco Systems
Received: from LINDY.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 28 Oct 87 05:52:17 PST
Received: by lindy.stanford.edu; Tue, 27 Oct 87 15:24:23 PST
Date: Tue, 27 Oct 87 13:06:52 PST
From: Jasper Williams <HK.JXW@forsythe.stanford.edu>
To: les@sail.stanford.edu
Subject: Future Business With Cisco Systems
REPLY TO 10/27/87 08:49 FROM AR.PMB "Pat Burns": Future Business With Cisco
Systems
Pat, We have resolved a good portion of the cisco problem and
expect to sign agreements documenting our agreement this week.
I will let you know if this in fact happens. If they sign the
documents that we have outstanding, then I have no problem
with our resuming doing business with cisco. Jasper
To: AR.PMB
cc: HK.PLD, AR.JHG, LES@SAIL
∂29-Oct-87 1103 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU Re: Royalty sharing for tip/gateway
Received: from SUMEX-AIM.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 29 Oct 87 11:03:23 PST
Date: Thu, 29 Oct 87 11:04:25 PST
From: TC Rindfleisch <Rindfleisch@SUMEX-AIM.STANFORD.EDU>
Subject: Re: Royalty sharing for tip/gateway
To: CS.EAB@FORSYTHE.STANFORD.EDU
cc: Rindfleisch@SUMEX-AIM.STANFORD.EDU, Yeager@SUMEX-AIM.STANFORD.EDU,
Timothy@SUMEX-AIM.STANFORD.EDU, Les@SAIL.STANFORD.EDU
In-Reply-To: Message from "Elizabeth A. Batson <CS.EAB@forsythe.stanford.edu>" of Fri, 16 Oct 87 14:54:47 PDT
Message-ID: <12346404302.36.RINDFLEISCH@SUMEX-AIM.STANFORD.EDU>
Elizabeth, this is to follow up on the earlier messages and telephone talk we
had about disposition of the accumulated CISCO payments (about $15K through
August) for the Ethernet TIP/Gateway software. I have spoken with Bill Yeager
and we are in agreement with the split between co-developers (80% SUMEX, 15%
EE, and 5% CS). Furthermore, Bill Yeager has agreed that all of the SUMEX 80%
will go into our unrestricted funds to support further systems research in our
group. On that basis and past precedents for CSD software royalty allocations,
I will seek approval from Nils Nilsson to have the department also pass on
their 1/3 of our share to the same SUMEX unrestricted fund for research
support. We will let you know about Nils's decision asap.
Margaret Timothy will send you an account number shortly. Tom R.
-------
∂29-Oct-87 1552 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU CISCO Royalties for TIP/Gateway Software
Received: from SUMEX-AIM.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 29 Oct 87 15:52:48 PST
Date: Thu, 29 Oct 87 15:53:46 PST
From: TC Rindfleisch <Rindfleisch@SUMEX-AIM.STANFORD.EDU>
Subject: CISCO Royalties for TIP/Gateway Software
To: Nilsson@SCORE.STANFORD.EDU
cc: Rindfleisch@SUMEX-AIM.STANFORD.EDU, Yeager@SUMEX-AIM.STANFORD.EDU,
Les@SAIL.STANFORD.EDU, Timothy@SUMEX-AIM.STANFORD.EDU
Message-ID: <12346456978.36.RINDFLEISCH@SUMEX-AIM.STANFORD.EDU>
Nils, as you recall (fondly, I'm sure), a settlement was reached last summer
with CISCO (Len Bosack, Inc.) about royalties for various pieces of hardware
and software developed at Stanford and now being commercialized by CISCO. This
message concerns royalty allocations for the accumulated CISCO payments through
August for the Ethernet TIP/Gateway software part of the agreement -- amounting
to about $15K total.
The initial design and development of these systems was done by Bill Yeager in
my group at KSL/SUMEX with additions by EE and CSD/CF. There is agreement
among the co-developing groups that an appropriate split of these software
royalties is 80% for SUMEX, 15% for EE, and 5% for CSD/CF. Furthermore, Bill
Yeager has agreed that all of the SUMEX 80% will go into our unrestricted funds
to support further systems research in our group -- i.e., Bill will waive any
personal remuneration from the royalties.
Under similar situations in the past, when such royalties have been plowed back
into research support, CSD has agreed to pass on their 1/3 departmental cut of
the royalties to the developing group. This message is to request your
agreement for a similar disposition of the departmental part of the SUMEX 80%
of the CISCO TIP/gateway royalties. This is especially important in the face
of declining federal support for these kinds of systems development efforts,
even though they remain critical to supporting our computer science research
environment.
Thanks for your consideration of this matter. Let me know if you have any
questions or would like further discussion.
Tom R.
-------
∂29-Oct-87 1713 nilsson@Tenaya.Stanford.EDU CISCO Royalties for TIP/Gateway Software
Received: from TENAYA.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 29 Oct 87 17:13:17 PST
Received: by Tenaya.Stanford.EDU (3.2/SMI-3.2)
id AA04393; Thu, 29 Oct 87 17:13:01 PST
Date: Thu, 29 Oct 87 17:13:01 PST
From: nilsson@Tenaya.Stanford.EDU (Nils Nilsson)
Message-Id: <8710300113.AA04393@Tenaya.Stanford.EDU>
To: Rindfleisch@SUMEX-AIM.Stanford.EDU
Cc: Rindfleisch@SUMEX-AIM.Stanford.EDU, Yeager@SUMEX-AIM.Stanford.EDU,
Les@SAIL.Stanford.EDU, Timothy@SUMEX-AIM.Stanford.EDU
In-Reply-To: TC Rindfleisch's message of Thu, 29 Oct 87 15:53:46 PST <12346456978.36.RINDFLEISCH@SUMEX-AIM.STANFORD.EDU>
Subject: CISCO Royalties for TIP/Gateway Software
Let me think about your proposal for csd to waive its royalty share,
Tom. Here are my biases as I begin to think: I have seldom been able
to get SOE to waive its royalty share---they prefer to accumulate
these and then make decisions about how to spend them in the best
interests of the School. I feel somewhat similar about CSD's share.
In my view, the various research groups have not been very generous
(compared to CMU's groups, for example) at contributing to
first-year PhD student support. Yet, without someone supporting the
first-year students, we wouldn't have any second-year students. So far,
the Dept. has assumed that responsibility. I think an excellent case
could be made for csd's share of royalties to be devoted to first-year
PhD student support---or to other things that benefit all research
groups. Comments? -Nils
∂30-Oct-87 1714 RINDFLEISCH@SUMEX-AIM.STANFORD.EDU Re: CISCO Royalties for TIP/Gateway Software
Received: from SUMEX-AIM.STANFORD.EDU by SAIL.STANFORD.EDU with TCP; 30 Oct 87 17:14:53 PST
Date: Fri, 30 Oct 87 17:16:16 PST
From: TC Rindfleisch <Rindfleisch@SUMEX-AIM.STANFORD.EDU>
Subject: Re: CISCO Royalties for TIP/Gateway Software
To: nilsson@TENAYA.STANFORD.EDU
cc: Yeager@SUMEX-AIM.STANFORD.EDU, Les@SAIL.STANFORD.EDU,
Timothy@SUMEX-AIM.STANFORD.EDU, Rindfleisch@SUMEX-AIM.STANFORD.EDU
Message-ID: <12346734139.36.RINDFLEISCH@SUMEX-AIM.STANFORD.EDU>
Nils, meetings and preparing to travel to a conference next week have conspired
to delay getting a reply to you -- in fact, it might be better to sit down and
talk about all this in person since the issues are complex. While I understand
your problem with 1st year graduate students, I would make the following
counterarguments:
1) Some of us have tried to be as helpful as possible in supporting early
students and shouldn't necessarily come under the general indictment of being
stingy. The SUMEX project, over its 15 years of existence and under the
constraints of NIH funding for AI applications, has done quite a lot in this
regard for both PhD and MS-AI students. Thus, I would advocate dealing with
the generosity vs departmental need issue on a case by case basis, even though
this makes administering the policy difficult by forcing individual decisions.
But how often does this royalty issue come up?
2) We have tried to foster an attitude among KSL faculty and staff of
contributing funds that otherwise might be taken as personal remuneration, back
to support university research. In the midst of start-up fever (and some
Stanford people still feel screwed by having shared software in good faith with
Bosack while at Stanford and then finding CISCO underway without any discussion
or share in it), we have felt the need to provide an incentive for developers
to contribute their share back to the university. For people who have not
benefitted personally from either the commercialization/start-up or the
university royalties, that incentive comes from doubling their contribution to
their group's research activities through getting the matching departmental
share.
I look forward to talking more about all this when I get back the week of 11/9.
Tom R.
-------
∂11-Nov-87 1205 ARK gfs
I saw someone logged into GFS account yesterday from host MATHOM.CISCO.COM.
I don't remember what file (thru E) the person was reading.
Arthur
∂11-Nov-87 1210 ARK gfs files
the files were pofile[1,gfs] and parts[1,gfs]. I forget the extensions now.
the first I saw user gfs reading, the other is listed by dr with sort on
-ref
Arthur
∂23-Feb-88 1402 AIR
Carol Gilmer, 38079, about CISCO
Cisco investment close Dec. 22, 1987.