I've assembled some notes from old manuals and other sources
on the formats used for on-disk file systems through the
Additional notes, comments on style, and whatnot are welcome.
(It may be sensible to send anything in the last two categories
directly to me, rather than to the whole list.)
The recent FOSS release of CDE has got me thinking of other Unix GUIs
from the "golden age" of workstation Unix. Obviously, stuff like
SunView and OpenWindows from Sun and 4DWM/Indigo Desktop from SGI are
pretty well known, but I've always wondered what else was out there.So
far, I've come across Looking Glass, DECWindows and HP VUE. Is there
anything else of any importance/interest out there?
Though Apollos were much more Unix-like than actual Unix, the DM
environment (display manager?) on Apollo Aegis, Domain/IX, and
Domain/OS workstations was pretty interesting, most notably for the
sophisticated interaction between the command-line and GUI. The DM was a
lot like the Moxie carbonated beverage - you either liked it or you
really wanted to spit it out. Apollo systems also ran various versions
of the X Window system, but the unique stuff was in the DM.
Site such as Toastytech's GUI Timeline
(http://toastytech.com/guis/guitimeline.html) and Typewritten Software's
Retrotechnology Media page (http://www.typewritten.org/Media/) have many
screen shots of old GUIs and apps, many on Unix or Unix-like systems.
However, the complicating factor here is, I think, that SYS V uses a
lot of code from the later editions of Unix, so relicensing the newer
research versions might cut into the license streams from them in
some way. At least that was reported at the time of the only through
7th edition licensing.
I'm not sure who would have reported that `System V uses a lot of
code from [post-7th] editions of UNIX.' I may be misled by having
had my hands and eyes mainly on the kernel and the most-basic commands
like the shell, but offhand I can't think of any System V code at all
that was adopted directly from the Research systems in the 8th, 9th,
or 10th Edition eras.
There were certainly ideas that were picked up, mulled over, and
re-implemented in changed form by the System V people, sometimes
better and sometimes worse than the original; but not straight
code transplants. The systems had diverged far too much for that
to be easy.
If anything, the licensing problem runs the other way: System V
code taken in by the Research system. For example, the C compiler
we used most was based on pcc2, developed on the System V side of
the company after Steve Johnson moved there. I think our version
of make may have been based on a System V version as well. I'm
sure there are other (mainly smaller) examples, though since we
used no source-code control mechanism, tracing the details is
None of which invalidates the basic point: there's certainly
plenty of entanglement, whether because 10/e includes ideas that
were used commercially in System V and whose mutant descendants
are still present in Solaris, or because 10/e includes some
source code directly descended from System V.
It's a shame we didn't get the several companies whose lawyers
might care to agree that there's nothing of commercial value in
the latter-day Research systems back when it was simpler to figure
out who those companies were. As I've reported here before, there
was actually some thought by certain persons here (and one who is,
alas, no longer able to be here) of doing that, some years back,
but a certain irksome legal circus about UNIX IP got going too
quickly for that to happen, and left us with the confused situation
we have today.