perm filename KAHN.2[LET,JMC] blob sn#828400 filedate 1986-11-17 generic text, type C, neo UTF8
COMMENT āŠ—   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	\input buslet[1,ra]
C00011 ENDMK
CāŠ—;
\input buslet[1,ra]
\jmclet
\parindent 8truept
\vskip 30pt
\address 
Dr. Robert E. Kahn
Corporation for National Research Initiatives
P.O. Box 18658
Washington, D.C. 20036

\body
Dear Bob:

	My expenses for the Nov 9 and 10 digital library meeting were
as follows:

\tabskip=1truecm
\halign{%this is where the preamble begins
#\hfil&% column 1 is left justified
\hfil#\hfil&% column 2 is centered         
\hfil#% column 3 is right justified
\cr
Airfare&&          \$980.00\cr% first row
Round trip to SF Airport&&             10.00\cr% second row
Parking at SF Airport&&            22.00\cr% fourth row
Taxi&&                                  10.00\cr% fifth row
Sheraton Reston Hotel&&73.89\cr
&&---------\cr% this should come under the sums

&TOTAL&                                     \$1,095.89\cr
} % \halign
%[RUTIE, THE EXPENSES AND THE REST SHOULD BE ON SEPARATE PAGES].

\vfill\eject
	Let me re-iterate a few of the points that I made at the meeting.

\item{1.} Technology already exists to make an enormous improvement
in the accessibility of written material to the public at costs
far less than continuing the present system.  This technology has existed
since about 1970.
\item{2.}The primary problem is entrepreneurial and political ---
not technical.
\item{3.}It is always possible to conduct more studies and to do
more research.  In so far as this has the effect of postponing action,
such studies and research have substantial negative aspects.  Institutions
already exist to support research, and many people are programmed
to initiate studies.  One way to get support for a study is to claim
that its results are necessary before action can be taken.  This has
the afore-mentioned bad effects.
\item{4.}You and NRI are in a an excellent position to do much
of the politics required to bring about a national digital library
or at least to get some project started.
\item{5.}More precisely technology is ready for the following.
A national library in which the contents of the Library of Congress
are available on disk file and are accessible from any terminal in
the country by telephone directly and by the various networks.  In
its preliminary form, readers find the documents they want by using
the usual catalog, available as another document.  At first there
need not be pictures or even formulas, but once the project is started,
including documents for which these are essential should be an
early goal.  This involves research, but this research will
readily find support once the main task is undertaken.  The documents
would be primarily stored as character strings to permit searching and excerpting
rather than as images.
\item{6.}Ideally Congress would accept the goal of a national free
public electronic library comprising all documents in the Library
of Congress and would appropriate money accordingly.  The Library
of Congress could manage the project if it took a broad enough view
of its function.  The Library's current project based on storing
images is probably a harmful distraction, however.
\item{7.}An attempt should be made to achieve this goal.  At the
meeting the names of various Representatives and Senators who might
be interested were mentioned.
\item{8.}If it turns out that getting political support for the
grand goal of a comprehensive national library is infeasible, various
limited projects may be politically more feasible.  Here are some
that were mentioned at the meeting.
\itemitem{a.}The information that the Government must make
available according to the Freedom of Information Act should
be electronically available if the goals of the act are to be
realized.
\itemitem{b.}NTIS should make an electronic library of its
reports.  For this purpose it needs to define one or more electronic
formats in which it will receive the information.
\itemitem{}{\it In this connection it now occurs to me that the
Postscript format is probably not acceptable, because it doesn't
have a decent standard form in which the character strings are visible
so that the reader can search and excerpt.}
\itemitem{c.}The National Library of Medicine is relatively
advanced technologically compared to other Government agencies.
\itemitem{d.}The Government Printing Office should publish
many of its documents primarily in electronic form so that they are nationally
available for reading on terminals and for local printing.
\item{9.}The technical problems include
\itemitem{a.}Improved terminals for reading including very
portable ones.
\itemitem{b.}Computer system improvements that will permit handling
a large number of readers at once.  Doubtless this will involve parallel
processing at some point.
\itemitem{c.}Maintaining adequate redundancy.
\itemitem{d.}Copyright problems.  These can be postponed if
necesary by starting with public domain material and by making
arrangements for experimental use with copyright holders.  Co-operative
copyright holders can be offered title to the standard format copies
of their material in exchange for permission to use it in experimental
versions of the system.
\item{10.}I am willing to take part in a few presentations of the
idea to people who might help get the money to implement it.

	I am eager to see the NRI plans take a concrete form that
will advance these goals.

\closing
Sincerely,       

John McCarthy    
\annotations
%\vskip 1in
%Enclosure
\vskip 1in
JMC/ra 
\endletter
\end