Re: Cygnus Cygwin32 Press Release 1/21/97

[Jeremy Blackman <loki at maison-otaku dot net>]

On Thu, 13 Feb 1997, Jeremy Allison wrote:

> License (1) is the GPL. This is for authors producing work that
> they share under the GPL (something I stongly encourage people
> in the traditionally 'closed' world of DOS and Windows to consider).
> Note that my original post was in error - this doesn't preclude
> commercial use of the Cygwin32 library, so long as the terms of
> the GPL are followed.

No problems here, thank you for clarifying.  Since I usually release the 
source code to something I write under Linux or SunOS, I have no problems 
with this.

> The only people who can complain about the situation are the
> 'Shareware' type authors who typically take freely available 
> UNIX source code, port it to DOS/Windows and then hide the
> source and charge people for the privilage (a certain vendor
> of Windows NT telnetd springs to mind). Those people can pay
> to use Cygwin32 if they really want. Forgive me if I'm not
> terribly concerned about their problems with the GPL....

Aaaah, but this is the part that concerns me.

What if at some point I write a piece of software, from scratch (e.g. not 
ported from UNIX) using Cygwin32, as it allows me the freedom to use the 
development system (gcc, et al) I'm most used to with a minimum of code 
differences (kudos to the Cygwin team for that achievement).

Now, say I want to release this software as shareware.  Granted, in the 
past I've always released code; let's say for some reason I didn't on 
this project.  Not being a large commercial company, I wouldn't have a 
lot of cash to spend on the Cygwin32 license... if there's just a simple 
flat fee for everyone... like several thousand... there's no way a small 
shareware AUTHOR (not porter) will be able to afford the same license 
that a company like, say, Microsoft would.

If I am faced with a cross-platform development need for something which, 
for various reasons, I don't want to release the code to, I'd be forced 
to examine that.  If the financial cost of such a solution with Cygwin 
was more than, say, with Willows, I would probably end up having to 
develop the software under Win95/NT, and port to Linux using Willows.

This is just a hypothetical question.  Outside of work (where we use VC++ 
anyway), I can't see any reason why I'd write something I wouldn't 
release the code for.  But if such a situation were to come up in the 
future, I would like to have the information to make an informed 
decision. :)

> I know how far Cygwin32 is from POSIX. I also know how close we
> can get with more engineering effort. I think the people who think
> we won't get there will be suprised :-).

Yes, but the press release states that it is CURRENTLY POSIX COMPLIANT.  
It isn't.  It could say that it is close to POSIX compliant, but without 
being fully POSIX compliant, one should not make claims to such, 
regardless of what is planned for the future.  I could claim a graphics 
engine I wrote supported various things like full-screen animations.  
Well, nice claims... and something I plan to add later, but NOT something 
I currently have, and thus not something I will claim it has the ability 
of.  :)



+---[ Loki ]------------+---------------------------------------------------+
| Jeremy Blackman       | NeonMuck coder/maintainer - Multimedia MU*!       |
| loki@maison-otaku.net | http://www.maison-otaku.net/neon                  |
+-----------------------+---------------------------------------------------+

-
For help on using this list, send a message to
"gnu-win32-request@cygnus.com" with one line of text: "help".