Mail Archives: djgpp/1999/04/08/21:00:36
>It seems that much of the incentive behind the development of DJGPP was
>due directly to the inadequacies of DOS that you are ranting about.
>Wouldn't changing that underlying DOS `engine' in the ways you suggest
>just amout to reinventing DJGPP in the image of Linux? Why not just
>move to Linux, then, and abandon DOS and DJGPP?
>
From my perspective - I am happy using DOS and djgpp and Linux too. I would
think that it would be a real shame to lose djgpp, just because the
underlying OS is systematically destroyed. There is a wealth of other good
software/applications/games which is going to become unusable because MS are
engineering the OS so people are not going to be able to use the old stuff.
I personallity don't have the large quantities of money to purchase
expensive unreliable software, when a better system exists with the source
code, and is written using the GNU philosophy. Organisations such as schools
and computing in developing nations are dependant on good software like
djgpp and Linux to allow them to educate people to use computers, and I
would like to keep that avenue open for them..Some people may not want to
use Linux because they will lose the investment they have in made buying
software for DOS/Win95. This is one segment of end-user who will lose out in
the end if their existing OS is not supported.
I think that if djgpp is to survive we are going to need a DOS-based OS to
keep us going, with sources. FreeDOS is 16-bit and although closer to the
original DOS, is not the best system that we *could* have (for 386 and above
systems). If we are going to have to build a free software DOS which runs
all existing software, the I would like it to be the best version possible.
Make it POSIX compliant. Make it 32 bit. Build all of the useful stuff in at
a low level, so that everything running on the OS has access to all the
devices. Build an optional windowing system as a enhancement to run windows
programmes, rather than the only way to get things done as in Win NT. Add
Xfree and Gnome if people feel they need it. Make DOS configurable for
individual users requirements.
Providing that the system runs all people's current software, is reliable
and is supported, then people will stick with it.
>There are many of us who just want to do away completely with M$
>software. DJGPP can take you all the way there *except* for the kernel,
>and can do it with fewer system resources than can a full-blown Linux
>distribution, and it lets you keep your DOS machine. Very neat.
>
>Yes, there is lots of room for improvement in DOS. These improvements
>could be made to enhance the cooperation of DOS and DJGPP and to
>provide some of the capabilities you mention, but DJGPP won't be of
>much use without DOS of some kind. It doesn't *have* to be FreeDOS,
>but the advantage of the FreeDOS project is that their kernel is
>maturing rapidly, and it might deserve consideration as a starting
>point for building a better DOS that will still support DJGPP.
I agree with this. I think that if a djgpp DOS is to exist, it would be
silly not to take advantage of the existing code available and save
ourselves some work. Any enhanced DOS system does not need to be as
complicated as Linux (although there are some useful features which could be
added as extras such as virtual terminals) and can retain the same basic
functionality, but with the retro-fitted features of the original built as
standard..
Do people care enough about djgpp to ensure its survival by engineering a OS
to run it? Is this thread leading nowhere? I'd be interested in seeing
whether people see this as the way forward...or any other ideas they have..
---
Arron Shutt
version8 AT ashutt DOT demon DOT co DOT uk -- www.ashutt.demon.co.uk
"You can jump all you like but it's the day of the cow" - Mike Keneally
- Raw text -