Re: SGI GNAT Question? (Long)

[Chris Morgan <mihalis@ix.netcom.com>]

SpamSpamSpam <spam@spam.com> writes:

> I am using a feature in Netscape called "auto-wrap" ( thismodern tech. is
> great :) so I don't see the size of the line
> as its para from my view. But I have inserted hard returns
> as requested.

Not all newsreaders (even modern ones) auto-wrap intelligently on word
boundaries, so thismodern tech. is not great, it's best with some
cooperation from the users.

> A "currently willing" caveat would fit well here as you're a
> company.Employees come and go, people sell shares and company policy
> changes according to market need.

I think you'll find ACT is unusually focussed and the employees are
unlikely to allow it to be assimilated.

[SNIP]
> > Well I am always amused by people proclaiming that they
> > know the true intent of the GPL,
> 
> And your sheer arrogance visible in nearly every post amusesme.  The true
> intent is stated the preamble in a manner the plain
> english society would be proud of.
> 
> "The licenses for most software are designed to take away your
> freedom to share and change it. By contrast, the GNU General
> Public License is intended to guarantee your freedom to share
> and change free software--to make sure the software is free for
> all its users."

I think arguing by reference to the preamble or the GNU manifesto is
tempting but doomed to failure. RMS had very lofty goals when he wrote
them, but he was already old and wise enough to pay enough attention
to the details (i.e. the detailed statements of the GPL) to make a
workable pragmatic system of licensing and distribution. We have had
other posters complaining about perceived mismatches between the
manifesto and the license, but really they carry no weight, the GPL is
all we really have to go on (unless you have some software which is
licensed by "please abide by the manifesto").

> 
> The term "copyleft" was, I believe, intended to convey the intent
> of GPL detailed in plain english in the preamble. You don't
> need briefings from corporate lawyers to understand the intent (
> even if the remaining "copyright" rights escaped you :)

The intent is one thing, but the actual position is controlled by the
GPL. 

> 
> Thats the crux of these posts and discussion of the nature of the
> commercial and public versions of GPL GNAT and the fact that
> you seek to prevent the copy and redistribution of a GPL version
> called 3.11b2, stating first that its not based on 3.11p then that
> they are both GPL but not the same, then that they atleast differ
> in version header and then that they only differ in version headers.
> If you want to concentrate/distract on my lack of knowledge of
> the finer points of  the GPL, fine. After many posts I've got
> from you, by any means, a statement that 3.11b2 is 3.11p.

The public version is never the same as the latest commercial
version. If you have a support contract you can get newer versions of
GNAT including special experimental fixes and so on immediately after
they arrive in the source tree - possibly way too soon. By the time
the unsupported users such as myself get a new version of GNAT, it is
already an old version for the supported users as each time it bit
someone ACT fixed it under paid contract until it stabilised nicely
and looked like something even the great unwashed (again, i.e. me)
could use productively. That lag is approved by the FSF. Maybe not
applauded whole-heartedly by people like egcs and the Linux kernel
team, but not a violation.

> 
> > and then complaining at
> > us that we somehow are not following it,
> 
> Not complaining, just seeking clarification regards the differencesin
> "commercial" and "public" versions of GNAT, and resolving
> how you get to the position of saying a GPL work is not for
> redistribution because its commercial.

i.e. because its usable with commercial full-time support only

The other way of phrasing it is "because it's not tested, not proven,
not stable, just not ready".

> 
> > when in fact we
> > work closely with Richard Stallman, and he knows exactly
> > how we work, and *he* is perfectly comfortable that it is
> > appropriate.
> 
> Funny. I contacted GNU.org via email sometime ago and they toldme that
> while ACT were perfectly entitled to require you
> to part with a "support" fee for a copy of their "commercial" version
> it was not in the spirit of the GPL and anyone who obtained a copy
> was still free to redistribute it. For the price of a CD for
> example.

This is true. Many moons ago I was a supported user and received a
wavefront compiler. Someone else was interested in it and we discussed
the matter by email. I concluded I had the right to give it to him,
but that I wouldn't at that time - there is the responsibility to the
community to be balanced with the rights of one user. Since wavefronts
are (I think) just automatic output of ACT's quality control system
(proof that it builds essentially) they don't seem very valuable
compared to the risk of them getting out onto the Internet and causing
a huge flurry of pointless emails and newsgroup postings. When a
release is frozen but not available because it's brethren on other
platforms (e.g. WinNT) are still having problems, e.g. install wizards,
it would be tempting for me to ask for a new version from someone who
was supported. In fact I bet a few of these releases have been passed
around, but it seems the recipients have had the sense to keep quiet
about it. In my case I haven't got to the bleeding edge of GNAT tools
where I _needed_ the newest one for a long long time (3-4 years).

[SNIP]
> > > I can't predict the future, no more than you, but I would
> > > wager that the day GNAT has a single copyright source of
> > > ACT, is the day everyone has to pay for a new version
> > > (and not just the cost of making the CD). You can
> > > disagree and almost certainly will, it's just my view and
> > > that's my right.
> >
> > You of course have no basis whatsoever for this claim, but
> > you are right, people are free to make any outrageous claim
> > they like at any time!
> 
> Well, depends what gets your goat as to what you find outrageous.The basis
> I have for this claim  are GPL versions I cannot see without
> paying a fee well above redistribution costs.  Calling a GPL work
> "commerical", restricting its redistribution and asking everyone to believe
> on
> trust ( however well place) that theres no functional difference between
> this and the "public" version is something I find outrageous.

Well, many of the supported users post here from time to time, and
from my experience the situation is _precisely_ as Prof. Dewar
says. The commercially supported people generally work from the latest
commercial version. The only difference between these and the public
versions is that if a commercial one breaks they will fix it for
you. Think of them as release candidates and think of the commercial
users as late beta testers (wavefront users are alpha/beta
testers). Several release candidates can occur within one broad
release (3.11b2 implies versioning within 3.11). When a release
candidate works on all platforms, it's good enough to be made
public. They swap in a different version number, build the compiler
from scratch and make the release. Thus technically speaking it's not
the same build as any of the commercial versions (sum gnatmake is
different) but from the users point of view it's every bit as good as
the release candidate that shares the same source profile.

In case you're wondering what the point of this explanation is, I
should add that when I was a supported user, the Ada user group in our
company suggested it would be prudent to simply stick to the public
releases on the grounds that we wanted the most stable
versions. Prof. Dewar thinks that attitude is taking things too far,
but it certainly didn't seem to us that the commercial releases
contained any goodies that we should grab asap. It was all about
defence in depth - use the public one, if that breaks get the latest
commercial release and see if ACT have solved the problem, if not ask
them to fix it and get a wavefront with the fix. Once the fix is
firmly in place move back to the most stable version available
containing the fix (I'm talking about production compilers
here). Naturally I always had the very latest version I could lay my
hands on in my own area, and now I'm unsupported I do miss the ability
to go and get the commercial releases, but I know that they are not
the unqualified advantage you seem to think.

Cheers,

Chris

-- 
Chris Morgan <mihalis at ix.netcom.com                http://mihalis.net
 "We're going to start selling Linux to single-party users very
 soon. Q: It's going to be on the menu? A: Yes. You'll go to Dell, 
 pull down "operating system," and click "Linux."         - Michael Dell