Re: gnat-3.10

[Ronald Cole <ronald@ridgenet.net>]

Andrew Lynch <lynch@cci.de> writes:
> I see being able to get a free version of GNAT as a privilege, not a
> right. If GNAT does not work for me, then tough shit.

I am under the impression GNAT is under the GPL because it was
developed with government money and uses the gcc backend.  The "tough
shit" part really isn't true because source is made available.

> I can either have a look at the sources and try to fix it myself, or
> I can *pay* ACT to make me a version that works for me, or I can
> wait for the changes that somebody else *paid* ACT to make become
> available in a public release.

The latter is what I'm doing since Robert has convinced me that I
probably should charge him for any fixes I make to his product.  And
since it'll be a cold day in hell when that happens, I won't be able
to distribute fixed versions.  It looks like quite a bit of work to
make the pa-risc more robust with Ada.  It also seems like no one is
willing to pay the high cost to fix it.  Of course, I'm not screwing
around with any of the gcc-2.8 pre-releases so I may be completely
off-base here, but no one has provided any information to indicate
that it will change in this area.

I would be very willing to pay for support once I saw that the product
had matured on that platform.  Robert has yet to convince me with a
factual argument that GNAT for the pa-risc has.

> For my personal use of GNAT I usually go with the third option. If
> you want to use GNAT in your company I would suggest you go with
> option two and carry on these arguments via email.

No client of mine would be willing to pay what it would cost to make
GNAT more robust on the pa-risc platform.

> Maybe ACT could fix every problem free of charge, but for how long?
> At some point all ACT employees will have starved to death and then
> where will you send your bug reports..?

The Hacker Ethic that spawned the GNU Manifesto has both feet in the
coffin.  I contributed a considerable amount and time and effort to
develop COFF support for and get it into g++-1.32 almost a decade ago.
It was the right thing to do and I don't regret not profiting from it
at all.

-- 
Forte International, P.O. Box 1412, Ridgecrest, CA  93556-1412
Ronald Cole <ronald@ridgenet.net>        Phone: (760) 499-9142
President, CEO                             Fax: (760) 499-9152
My PGP fingerprint: E9 A8 E3 68 61 88 EF 43  56 2B CE 3E E9 8F 3F 2B