COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	\\M1BAS35.FNT[XGP,ALS]\M2NGR40L\M3NGR25\M4NGR20\
C00006 ENDMK
C⊗;
\\M1BAS35.FNT[XGP,ALS];\M2NGR40L;\M3NGR25;\M4NGR20;\;
\←L\+=50;\→L\←R\-=50;\→R\←L\→.\;
\F2\CSTANFORD UNIVERSITY
\F3\CSTANFORD, CALIFORNIA 94305
\F4ARTIFICIAL INTELLIGENCE LABORATORY\←L\-R\/'7;\+R\→.\→S   Telephone:
\←S\→.415-497-3330
\F1\CJune 24, 1974



Brian Harvey Esq.
Artificial Intelligence Laboratory
Stanford University
Stanford, Calif., 94305


Dear Brian:

\JYou have seen this letter before. It is being written to bring to your
attention a new font especially
created for use in letters where the speaker has very little to say and still
wants to make a reasonable letter.

This little experience has demonstrated to me that the existing tools for font
preparation are quite inadequate. While TVFONT is a cute package it does not
recognize the particular properties of fonts that should be under separate
control. For example one should be able to specify that all linear dimensions
except for the thickness of lines (which are less than some setable value of
thickness) are to be changed by a specified ratio, or the inverse, etc.

I will probably not be able to do the corresponding italics and bold face
before I leave for two months but I will accept the responsibility for doing
this on my return. In the meantime this is probably good enough to use.
Of course if someone else would like to make these other fonts they can go right
ahead.

This is simply BASL30 blown up by 35/30 with a minimum amount of editing
using EDFONT. The font now resides on XGP,ALS as BAS35.FNT.

The attached two test sheets may help you to locate any remaining fixable
defects. If you would like to do some more polishing you are welcome to the
task. I have done my best subject to time constraints.\.

\←L\→S\←R\-L\/'2;\+L\→L

Sincerely,



Arthur L. Samuel

\←S\→L
ALS:pdp10