Mailing List Message #13585
From: Bill Cole <>
Subject: Re: eudora support and SIMS temp failure
Date: Fri, 5 Sep 2003 10:09:35 -0400
To: SIMS Discussions <>
At 3:46 PM +0400 9/5/03, Technical Support  imposed structure on a stream of electrons, yielding:
Hello, on 05.09.2003 00:52, Stefan Jeglinski at wrote:

 I can appreciate that. Is the person that wrote it still there at
 all? Would that have been dear old Vlad B?

 Correct. But I doubt he'll ever look at that code again.

 Fair enough. But since I have your attention, let's consider the
 stated facts  and some speculation:

 1. Talented programmers like Vlad are unlikely to ever look at
 the code again.

 2. You're deathly afraid to touch certain parts of the code :-)

 3. The SIMS code base is not part of CGP, otherwise CGP would have to
 be considered a mess. Also, #1 and #2 would not be true.

 So I again bring up the idea of open-sourcing SIMS. At worst, no one
 will be able to figure it out and it will just sit, as it does now.

 Can we run this up the flag pole again? I've never seen a definitive
 statement from the Commandant of Stalker as to why this cannot or
 will not be done.

Here is the official answer I got:

Stalker Software is a software development company, and not an "open-source
development" company. Currently we see no sense in openening the source code
of any of our products.

That's either encouraging or deeply disappointing...

The only rational reason to keep ahold of SIMS in a proprietary manner is if Stalker intends to do more work with the code. That would be really good news to SIMS users, as there was at one point at least a theoretical plan mentioned here to do some very interesting things with SIMS.

On the other hand, the wording of that leads me to believe that the author of that official answer really doesn't understand SIMS or the nature of open source software and why companies like Apple, IBM, Sun, and even SCO (spit) all have seen reasons to open some of their source. SIMS falls into the same category as the Unix System 7 code that SCO opened before they went totally insane. It is essentially obsolete and of interest to a small handful of people who either still use it in odd niches or have historical interest in coding techniques. That may sound surprisingly derogatory from a longtime SIMS user, but it is simply honest. The world has moved along in the past few years as SIMS has stagnated. Stalker need not get into the business of open source development in order to open SIMS. SCO is not in the developing System 7, and IBM is not developing the OS's for the S/36 that they have opened. Opening up old and essentially obsolete code (as any network-oriented code is that is written for Classic MacOS) is a way of turning code that is at best of zero value to Stalker into something that others can benefit from, without necessarily costing Stalker anything. Stalker need not guide or provide resources for ongoing development of SIMS if its source were to be opened. Given that the SIMS code is now
approaching 4 years old for the latest changes and is written for an OS that no one is still writing any serious apps for, the idea that the SIMS source contains some secrets which have proprietary value is simply absurd.  SIMS is also no longer a useful advertisement for CGP, because CGP has moved so far ahead of SIMS and the world has so changed in the last 4 years that the logical step up from SIMS is really Postfix.

From myself, I can only promise that if I have some spare time in the
nearest future, I'll try to do my best to figure out that spot in the queue
management code to fix the problem with temp rejection of addresses.

Given that SIMS seems to understand SOME 4xx codes and not others and to support  the concept of a message being delivered successfully to some recipients but not yet to others (that based on what I've seen in multi-recipient messages in the Queue) it seems to me that this may be a trivial fix: just make sure that any 4xx response to RCPT leaves that recipient's status line in the queued message tagged for retry.

Bill Cole

Subscribe (FEED) Subscribe (DIGEST) Subscribe (INDEX) Unsubscribe Mail to Listmaster