perm filename BBNLET[DOC,AIL] blob
sn#049057 filedate 1973-07-17 generic text, type T, neo UTF8
To: Jim Goodwin
From: the Stanford SAIL group
Date: 18 June 1973
Copies to: J. Feldman, M. Weinstein, B. Sutherland, R. Smith
1. What is the state of TENEX SAIL? The people at IMSSS are most
anxious for you to finish up, since they really need it and have put
a great deal of effort into a TENEX IOSER, which I understand you
were planning to use. Also RFS at Xerox & some other TENEX sites
have been asking about a TENEX SAIL. In looking at your directory
at BBN, I noticed that the SAIL files haven't been written since
last April, or thereabouts. Is this because you are done? If not,
is there any information we can supply that you need to finish up?
Also, when could you expect to be ready to merge files? Bob Smith
tells me that IMSSS has been ready for the past couple months (on
about two days notice) to merge in their work, which includes the
TENEX stuff, together with some arithmetic routines.
2. It seems to me that such a merger would be very desirable, even
if the files are only to be kept to within a few months of each
other. There are several reasons for this. First, a common base is
pretty much needed if bug fixes or new features added at one site are
to be propogated to other sites and systems. Ideally, such changes
would become instantly and universally effective. However,
realistically, probabably the best we can hope for is about a three
month interval. The problems of maintaining even this degree of
compatibility will become almost insurmountable unless there is one
"official" set of files from which various sites can work, and to
which they can periodically submit modifications. Although I quite
agree about the desirability of a TENEX (or DEC 10-50) site being
able to snarf a current .SAV file set, it seems to me that it is
still very important that a set of base source files be available on
line so that various users can, if they desire, add their own local
modifications (for instance, device dependent display code) while
still keeping maximal compatibility with the world at large and
getting the benefit of recent fixes. To maintain any sort of option
for such a process, it seems to me to be imperative that at least
the basic TENEX structures get merged in as soon as possible, since
the attendant difficulties of creating such a common base are going
to increase, and our availability for doing the initial work
decrease, the longer we wait (see also paragraph 4).
3. We are finally ready to ship out the long awaited Decus release,
which will go out just as soon as the manual is printed. (hopefully
by the end of June -- the main uncertainty being the university
print shop). After this release, we do not anticipate any more major
developments except for Kurt's debugging system, since JRL, HJS, & I
will be pretty much tied up with theses. Of course, we do plan to
continue fixing bugs.
4. When we agreed last February to incorporate your TENEX changes
into our files as part of a common file base, it was with the
expectation that you would have something ready in a reasonable
time. We are still interested in making the great merge, provided
it can be done by the time the Decus release finally goes out. The
SAIL group here has more or less allocated a week for last minute
adjustments and cleanups to the source files, which will take place
in about 2-3 weeks. During this time we will be IFN'ing out the
Stanford specific hacks, incorporating the IMSSS arithmetic
routines, etc. So we feel that this would be the best time for a
merge. After that date we are much less likely to be willing to
undertake such an enterprise, since we will have to be spending much
more time on our theses. If you have anything in a "done" state or
can get something into an approximation of such a state within our
time frame, we strongly urge you to arrange with us for the merger.
In any event, we are anxious to know when something might be ready.
5. We are giving serious consideration to holding a SAIL users'
meeting here at Stanford. The date is yet to be set, but we are
thinking about holding it just before the IJCAI. Do you have any
thoughts about such a meeting, or about a date?
6. Sorry about the delay in answering your last letter, but I was
out of town until last week, and it took some time for us to
formulate a detailed reply. If you have any questions, perhaps a
phone call would be useful. The number here is (415) 321-2300 x 4971.
Russ Taylor