gc-by-area and host uptimes

[Scott McKay <SWM%sapsucker.scrc.symbolics.com@RELAY.CS.NET>]

    Date: Mon, 13 Feb 89 16:59 EST
    From: Qobi@ZERMATT.LCS.MIT.EDU (Jeffrey Mark Siskind)

    I don't necessarily beleive that Symbolics (or any company for that matter)
    always has the insight to work on only the things with the widest impact.

Agreed.

    Sometimes companies misjudge the relative impact of different alternatives
    and sometimes they just plain ignore the impact issues and work on things
    based upon their whims. Now some things that Symbolics is working on
    clearly do have far greater impact than mark-sweep-gc. Like lower-cost
    higher-performance hardware like Ivory. On the other hand, there have
    been some significant development efforts by Symbolics, ones which
    I assume take more effort than mark-sweep-gc, which I think have far
    less impact than it. For example: the Fortran and Pascal compilers,

No comment on the compilers.

    Joshua, Concordia, the document examiner, Statice, most of the
    features of dynamic windows besides infinite scrolling, ...

I beg to differ with you on these, however.

If you look beyond the simple user-interface provided by the DocEx
(i.e., what you get when you type Select D), what is really provided is
a substrate for the delivery of on-line hypertext documentation, which
includes a passably good formatter.  In addition to providing our own
documentation on-line, it provides, in conjunction with Concordia, the
ability for our customers to create their own documentation.  In fact,
many of our customers have been asking for Concordia for years, back
when it was still called Writer Tools.  Using the rest of Genera to
create such documentation on-the-fly can be very powerful indeed.

Same deal for Joshua: you get to use the Lisp development environment to
rapidly produce relatively high-performance expert-systems applications.
Joshua is very flexible and very well integrated into Genera.  Combining
it with Concordia and Statice produces some very interesting applications.
Joshua implements the notion that AI techniques should just be another
part of the grab-bag that a programmer can dip into when he sets out to
create a new program.

As for Statice, we hope to make later versions of it the foundation upon
which many more advanced development and delivery tools are built.  A
high-performance object-oriented database is the right answer to many of
the current problems in Genera.  (Obvious examples: consider implementing
a mail system on top of Statice, or the namespace system, or a version
control system, or the who-calls database, etc.)

And infinite scrolling is surely useful, but empirical studies have shown
that about 50% of the effort of modern software projects go into
developing the user interface.  I find the the other DW tools,
particularly the presentation system, translators, and the command
processor, save me an enormous amount of effort in cobbling up new
applications.  DW has its problems and shortcomings, but I think that in
its entirety it is perhaps the most significant advance forward Symbolics
has made in a very long time.  For example, on how many other systems can
you implement a passable version of a Peek-like program in 30 minutes
which has nice output, incremental redisplay, and a dozen or so useful
commands on processes?  Not very many, I bet.

So there. :-)