[TUHS] Re: Sad news from IBM...

[TUHS] Re: Sad news from IBM...

[James Falknor]

>
>
>I was reading Groklaw yesterday night when I came across this. It is a
>very sad thought to know that possibly tons of old/ancient code is being
>dumped in the trash bin.
>
>More so now since the advent of software patents: it may become very
>difficult to avoid a patent on a re-invention of the wheel if previous
>knowledge has been dumped.
>
>OK, the quote. It is from "the Todd Shaughnessy affidavit [PDF] from IBM 
>that Magistrate Judge Brooke Wells requested they file when they turned 
>over all the code and paperwork to SCO":
>
>	28. As I have noted above, IBM does not maintain revision control 
>	information for AIX source code pre-dating 1991. To the extent that 
>	any code for the AIX operating system (that did not duplicate the 
>	code already being produced in CMVC) was found during the search 
>	described in Paragraph 26-27 above, it was produced. Paragraphs 
>	29-31 below describe additional search efforts IBM undertook to 
>	locate pre-1991 versions of AIX code. No versions of AIX pre-dating 
>	1991 were found.
>
>	29. In the 1980s and early 1990s, IBM prepared vital records backups 
>	of AIX source code and transferred them to a remote storage location. 
>	At some point in the 1990s, the AIX vital records tapes were transferred 
>	to Austin, Texas. In late 2000, the tapes were determined to be obsolete, 
>	and were not retained.
>
>	30. The AIX development organization contacted other IBM employees who 
>	were known or believed to have been involved with the development or 
>	product release of AIX versions prior to 1991. In addition, IBM 
>	managers and attorneys asked current members of the AIX development 
>	organization whether they were aware of the location of pre-1991 
>	releases of AIX source code. No one asked was aware of any remaining 
>	copies of pre-1991 AIX source code.
>
>Perhaps we should do something to raise awareness about the relevance of 
>legacy (not only UNIX) source code. And in any case, it is a pity that all
>that historical information had been lost forever.
>
>I have always complained about this, and consider it the biggest drawback of
>closed proprietary source code: it is OK that law protects developer interests 
>with the goal of promoting innovation and the public benefit at large. But it
>is a lose for everybody whenever any such "protected" code is dumped into the
>bin banning anyone else from further benefitting from or exploiting it, and 
>opening the road for opportunists to claim they "newly invented" it.
>
>Sic. Sigh.
>				j
>

All may not be lost.

As it appears to me, TUHS has connections with Universities / Colleges 
and other types schools, as well as programmers, software engineers and 
the like.

All we need to do is put the word out that TUHS is seeking pre-1991 AIX 
source code and it's bound to surface. If all else fails, I'm sure 
someone has a pre-1991 AIX binary distribution that could be 
disassembled (that is if a binary distribution can be disassembled back 
to a rough source code).


To all TUHS members,
     As a part of the heritage of Unix, please search any and all your 
archives for pre-1991 AIX Source Code. Maybe, just maybe, a pre-1991 AIX 
Binary Distribution will suffice. Help IBM,  TUHS, and in the end, the 
heritage of Unix.

    Thank you,
    James Falknor
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20050608/bf19aaf8/attachment.html>

[TUHS] Re: Sad news from IBM...

[Peter Jeremy]

On 2005-Jun-08 23:07:34 -0600, James Falknor <james at peacemax.org> wrote:
>All we need to do is put the word out that TUHS is seeking pre-1991 AIX 
>source code and it's bound to surface.

It depends how widely spread the source code was and how may people
still have readable 15 year old backups.  I know I tend to delete old
code after a while and many of my old QIC-150 tapes are no longer
readable.

> If all else fails, I'm sure 
>someone has a pre-1991 AIX binary distribution that could be 
>disassembled (that is if a binary distribution can be disassembled back 
>to a rough source code).

I'd be very surprised if this produces anything useful.  The code will
have been compiled with a reasonable degree of optimisation and won't
have any debugging symbols in it.  It would be reasonably trivial to
turn it into something that a C compiler could understand but making
it look anything like the original is a major undertaking.

-- 
Peter Jeremy

This email may contain privileged/confidential information. You may not copy or disclose this email to anyone without the written permission of the sender.  If you have received this email in error please kindly delete this message and notify the sender.  Opinions expressed in this email are those of the sender and not necessarily the opinions of the employer. 

This email and any attached files should be scanned to detect viruses.  No liability will be accepted by the employer for loss or damage (whether caused by negligence or not) as a result of email transmission.