AT&T 3B1 Source Code

129 views
Skip to first unread message

Forrest Aldrich

unread,
Apr 21, 2020, 6:51:35 PM4/21/20
to
Some time ago, and I'm talking many years, there was discussion and
debate about obtaining the full source code of the AT&T 3B1. I believe
at the time, there were far too many restriction and, AFAIK, the topic
was dropped.

I wonder if since then there has been any progress on this.

I'm not entirely sure what we could actually "do" with it these days, to
be honest. There is a 3B1 emulator project that is slowly coming along,
but short of porting it over to another architecture (nobody is going to
do that), it may be useless.

I personally miss the simple MGR windowing system, I thought it was
pretty neat for what it was.

Anyone have more information about the above?



_F

Grant Taylor

unread,
Apr 21, 2020, 8:46:30 PM4/21/20
to
On 4/21/20 4:51 PM, Forrest Aldrich wrote:
> I wonder if since then there has been any progress on this.

You might email The Unix Heritage Society mailing list and see if
anybody there knows anything more recent.

They would also likely have some information on ways to use it.



--
Grant. . . .
unix || die

Aharon Robbins

unread,
Apr 22, 2020, 1:26:33 PM4/22/20
to
In article <r7ntdj$v93$1...@gioia.aioe.org>,
Forrest Aldrich <for...@forrie.com> wrote:
>Some time ago, and I'm talking many years, there was discussion and
>debate about obtaining the full source code of the AT&T 3B1. I believe
>at the time, there were far too many restriction and, AFAIK, the topic
>was dropped.
>
>I wonder if since then there has been any progress on this.

I doubt it. I've also not seen anything about it on TUHS. It'd be
worth asking there, though. It's a good question as to if Convergent
even exists these days, who owns that source, and if it even exists
anywhere in machine readable form, from whence it might be recovered.

>I'm not entirely sure what we could actually "do" with it these days, to
>be honest. There is a 3B1 emulator project that is slowly coming along,
>but short of porting it over to another architecture (nobody is going to
>do that), it may be useless.

Yeah, only for the emulator.

>I personally miss the simple MGR windowing system, I thought it was
>pretty neat for what it was.

It ran also on Sun hardware. You might could find the source and
get it to run on PC hardware, but I doubt it'd be worth it.

HTH,

Arnold
--
Aharon (Arnold) Robbins arnold AT skeeve DOT com

J Booth

unread,
Oct 30, 2020, 3:08:49 PM10/30/20
to
On Tuesday, April 21, 2020 at 6:51:35 PM UTC-4, Forrest Aldrich wrote:
> I personally miss the simple MGR windowing system, I thought it was
> pretty neat for what it was.
>
Sorry for the late response on this, but figured I would mention it anyway -- I got Bellcore's MGR running on Phil's FreeBee (3b1) emulator earlier this year. I added the VIDPAL mod to the emulator so now users can write to VRAM :) I also got the C compiler working so you can compile MGR. You have to get the MGR source code into the emulator which I did by breaking it into 4 files and putting them on ms-dos 360k floppy images using dosbox and reading them using the 'msdos' program on the 3b1. And then reassembling the 4 files back into the tarball. I tried a few demo programs that came with MGR, some sort of 3d maze demo, and another QIX type demo.


Convergent MightyFrame

unread,
Dec 15, 2020, 12:22:03 AM12/15/20
to
So, I guess this entire thread brings up a very interesting question...

HAS the source code indeed been hacked, and therefore indeed obtained, by those noble among us (Phil Pemberton, Jesse Booth, Aharon Robbins & others I have inadvertently neglected to mention) who have created, developed and continue to improve this emulator?

Why obtain the source code? Well, maybe what one can "do" with it is to preserve the legacy of what it is. As a self-appointed historian of many things Convergent for the past 6 years, I find it a fascinating thing to preserve. Much like keeping detailed volumes of history in a library (or on the internet these days). Nobody could care for years to come, even decades, but once someone does, there it will have been...preserved by those few who cared.

I've learned a lot about this part of computing history that I never knew of when I obtained my first AT&T Unix PC in 2014, and wanted to learn enough to make it work, and become familiar with it. I know there are others who have this interest, both now, and in the future.

So, I just want to say a special "Thanks" to the contributors here that I've gotten to know. DoN Nichols, you are at the top the list, and although I just missed him when I started in 2013-2014, the late Thad Floryan. And there are MANY others. To all of you who brought us this far in preserving these marvelous, ingenious and still somewhat enigmatic machines, we thank you.

And thank you, Forest, for bringing up the topic.

Just my rambling thoughts late this Monday evening...

Best always,
AJ
http://MightyFrame.com

Aharon Robbins

unread,
Dec 15, 2020, 11:43:06 AM12/15/20
to
In article <95738019-e5f7-4df9...@googlegroups.com>,
Convergent MightyFrame <mighty...@gmail.com> wrote:
>So, I guess this entire thread brings up a very interesting question...
>
>HAS the source code indeed been hacked, and therefore indeed obtained,
>by those noble among us (Phil Pemberton, Jesse Booth, Aharon Robbins &
>others I have inadvertently neglected to mention) who have created,
>developed and continue to improve this emulator?

I don't have the source code. I can't say anything about anyone else.

I would love to see the source code available so that we could indeed
recompile Unix, but my hopes are not high.

Forrest Aldrich

unread,
Feb 26, 2021, 10:47:59 AM2/26/21
to
As I mentioned originally, many years ago I had corresponded with
someone who lived in Florida, who claimed to have all the source code,
but at the time it couldn't be released without legal approval.

At this point in time, I would think that the entirety of the source
code would have no patent/commercial value and "should" be made
available to the public -- under whatever license (ie: non-commercial).

What we'd need to do is identify "who" owns the code at this point.
Surely it's not with AT&T any longer -- but it's possible.

I imagine all of these companies reserve their various source codes for
posterity in some archival format, somewhere.

I would think having access to the code would open up many doors.

Thoughts?


_F

Forrest Aldrich

unread,
Feb 26, 2021, 10:57:59 AM2/26/21
to
According to Wikipedia:

In the early 1990s, AT&T sold its rights in Unix to Novell, which then
sold its Unix business to the Santa Cruz Operation (SCO) in 1995.

Most of us know what a debacle SCO ended up becoming. They still exist,
and I emailed their "info" address, explaining my interest.

In this case, I'm going to guess that the AT&T UNIX PC/3B1 source code
still exists with AT&T as a part of their archives.

I'll post here if/when I learn more.



_F

Grant Taylor

unread,
Feb 26, 2021, 11:04:10 AM2/26/21
to
On 2/26/21 8:47 AM, Forrest Aldrich wrote:
> I imagine all of these companies reserve their various source codes for
> posterity in some archival format, somewhere.

I would not hold my breath.

I've read many accounts where companies either destroyed things (IBM and
AIX/370), or lost it (HP lost a lot through fires), or didn't retain it
(migrated to a new source code repository and only copied active projects).

J Booth

unread,
Feb 26, 2021, 12:11:57 PM2/26/21
to
3b1 source code I expect would have been under the control of Convergent, who was later purchased by Unisys. Then to further complicate matters, I assume UNIX portions were licensed from AT&T... So doubtful that Unisys would want to release the code due to the former licensing involved.

Forrest Aldrich

unread,
Feb 26, 2021, 2:02:05 PM2/26/21
to
I put a feeler out to Unisys.

I wouldn't be surprised if the 3B1 source code is MIA and/or otherwise
sitting on media that's been stored, in a box forever in someone's
basement :)

Other than nostalgia, I'm not sure what we would "do" with the code if
we were to obtain it. I admit to liking the stock text-based menu
system; it was useful for certain purposes. But the rest of the system
has long since been deprecated.

Alas, nostalgia is powerful, too.


_F

dbr...@gmail.com

unread,
Aug 18, 2022, 11:39:21 PM8/18/22
to
Well, it would be nice to have kernel source and be able to put a real tcp/ip stack in that could work over the serial port with PPP -- even if only for Freebee. Being able to scp or -- woo hoo -- NFS mount would make life so much better.

Network reasons are why I gave up on my original one around 1994.
Reply all
Reply to author
Forward
0 new messages