RE: GNUstep on the rise!

2005-08-05 Thread Sašo Kiselkov

Quoting Vaisburd, Haim [EMAIL PROTECTED]:

 Cris Vetter wrote:

 [...] [ GNUstep separate OS ]
 
  AFAIK, xMach is dead :-( So does anyone know how far Hurd is with
  respect to usability?
 [...]

 Sorry, I deleted the essential part of Cris' letter that I'm going to
 reply.

 I guess when Chris was talking exclusive GNUstep experience he meant the
 distribution, not the OS per se (kernel). It seems the obvious choice
 for that distribution would be the Linux kernel. As far as I understand,
 nothing in the kernel dictates unix standard FHS, unix standard
 configuration
 (/etc content), a type of window system or other things that might annoy
 a person who experienced NeXT ( I never did ).

 On the other hand it's well supported for several architchures and many
 periferal devices.

 Other post suggested Darwin, I believe it would be as good as Linux,
 but why to look for something else?

 Tima.


Exactly. When I did my small GNUstep OS experiment I thought long about jumping
to some more interresting platform (I tried Hurd, and Darwin and BSD came to my
mind as well), but later I realized that it is not important to have a specific
kernel (since it's development isn't connected to GNUstep anyway) and the
following are facts:
Linux ...
 - is stable
 - has much support
 - has drivers for almost everything under our Sun
 - is layout independent and flexible (it's just a kernel after all)

However, the question of file system layout and other under the hood,
complicated features, isn't important - history has shown us that NEXTSTEP,
OPENSTEP, Mac OS X and other pure OpenStep-ish systems can be Unices without
any loss of ease of use. For example, on NEXTSTEP/OPENSTEP all folders like
/bin or /usr were simply hidden by the Workspace from the simple user,
until unhidden by a user default. This is even better than trying to reinvent
the wheel and define a different file system layout - lamers don't see what
they don't need, and experienced UNIX experts find themselves like at home
immediately. Don't make the same mistakes as Microsoft does, when they
forcefuly degrade experienced people to the level of trained monkeys.

Regards
 Saso



___
Discuss-gnustep mailing list
Discuss-gnustep@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnustep


Re: GNUstep on the rise!

2005-08-04 Thread Chris Vetter

On Wed, 3 Aug 2005 09:50:47 Tabitha McNerney wrote:
 Hi all,
 I have to say that when I read this about DRM and Apple Intel machines:
 http://apple.slashdot.org/article.pl?sid=05/08/01/0421248from=rss
 it made me realize even more so how important it is that GNUstep
 continue to rise to the forefront such that it can be installed on top
 of open source systems to avoid nasty DRM!!!

That reminds me...

A few days ago I read an article mentioning, among other things, GNUstep and
how 'cool' (my choice of words) it would be if/when/whether 'these guys'
(ie. the developers of GNUstep) would decide to put GNUstep on top of their
OWN operating system instead of 'screwing around with existing systems
resulting in symlinks all over the place' (again, my choice of words) and
'imagine GNUstep on a Mach based kernel.'

My first thought was along a distribution based on (Open)Darwin. However,
with respect to the recent news about Darwin/OSX x86 and DRM, I guess this
would be a 'no go.'

AFAIK, xMach is dead :-( So does anyone know how far Hurd is with respect to
usability?

Cheers,

-- 
Chris

-- 
GMX DSL = Maximale Leistung zum minimalen Preis!
2000 MB nur 2,99, Flatrate ab 4,99 Euro/Monat: http://www.gmx.net/de/go/dsl


___
Discuss-gnustep mailing list
Discuss-gnustep@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnustep


Re: GNUstep on the rise!

2005-08-04 Thread Yavor Doganov
On Thu, 04 Aug 2005 09:17:05 +0200, Chris Vetter wrote:

 AFAIK, xMach is dead :-( So does anyone know how far Hurd is with respect to
 usability?

It is usable, at least much more usable than it used to be.  You might
want to take a look at Michael Banck's announcement [1].  For the time
being only gnustep-make is ported for Debian GNU/Hurd and there are still
some other things to do in order to make precompiled packages.  But it is
moving forward, that's the most important thing.
I'd love to see GNUstep developers paying more attention to GNU systems,
rather than trying to enhance proprietary platforms.

[1] http://lists.debian.org/debian-devel-announce/2005/07/msg6.html

-- 
Yavor Doganov
Free Software Association - Bulgaria



___
Discuss-gnustep mailing list
Discuss-gnustep@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnustep


Re: GNUstep on the rise!

2005-08-04 Thread Sašo Kiselkov
Quoting Chris Vetter [EMAIL PROTECTED]:

 A few days ago I read an article mentioning, among other things, GNUstep and
 how 'cool' (my choice of words) it would be if/when/whether 'these guys'
 (ie. the developers of GNUstep) would decide to put GNUstep on top of their
 OWN operating system instead of 'screwing around with existing systems
 resulting in symlinks all over the place' (again, my choice of words) and
 'imagine GNUstep on a Mach based kernel.'


This would, IMHO, be the biggest mistake possible: tying GNUstep to a particular
platform. I personally love that fact that I only have to write an app once and
then have it running with little porting effort on Linux, FreeBSD, Mac OS X,
and even ugly, but spread M$ Windows.

Of course, I'm not against a GNUstep-only-system (hell, I already did it as my
graduation work at high school, complete with a CD-based installer, integrated
workspace, it's own package management, etc.), but care should be taken not to
make it part of the core libraries.

 Saso




___
Discuss-gnustep mailing list
Discuss-gnustep@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnustep


Re: GNUstep on the rise!

2005-08-04 Thread Nicolas Roard
On 8/4/05, Sašo Kiselkov [EMAIL PROTECTED] wrote:
 Quoting Chris Vetter [EMAIL PROTECTED]:
 
  A few days ago I read an article mentioning, among other things, GNUstep and
  how 'cool' (my choice of words) it would be if/when/whether 'these guys'
  (ie. the developers of GNUstep) would decide to put GNUstep on top of their
  OWN operating system instead of 'screwing around with existing systems
  resulting in symlinks all over the place' (again, my choice of words) and
  'imagine GNUstep on a Mach based kernel.'
 
 
 This would, IMHO, be the biggest mistake possible: tying GNUstep to a 
 particular
 platform. I personally love that fact that I only have to write an app once 
 and
 then have it running with little porting effort on Linux, FreeBSD, Mac OS X,
 and even ugly, but spread M$ Windows.
 
 Of course, I'm not against a GNUstep-only-system (hell, I already did it as my
 graduation work at high school, complete with a CD-based installer, integrated
 workspace, it's own package management, etc.), but care should be taken not to
 make it part of the core libraries.

Well, GNUstep itself should stay as it is, cross-platform, indeed.
Having a good windows port would be rather useful, and should attract
OSX devs. But on the linux (or bsd) side, I don't see how GNUstep
(more exactly, a GNUstep-based desktop) could prevail against
KDE/GNOME -- considering their respective market share and, the fact
they are rather good now, it would be really difficult for a 3rd
desktop to really emerge.

For me the only real option would indeed to have a GNUstep OS
containing only GNUstep apps, to really have a consistent and powerful
environment that people can easily try and adopt. Plus, in some way it
would be easier than a desktop: with a desktop you need to be rather
tolerant to your host OS, while with an OS you can control everything
and thus propose original solutions (eg you can decide to use launchd,
rendezvous, etc.). If people want to adapt the desktop part on their
own os, they still can anyway.

You can see how different it feels when you have a complete GNUstep
env by trying the GNUstep livecd (http://livecd.gnustep.org) ... and
yet, the live cd is just based on debian.

Anyway, the live cd route is a good one, as it let people easily try
new environments/projects.

-- 
Nicolas Roard
Any sufficiently advanced technology is indistinguishable from magic.
  -Arthur C. Clarke
___
Discuss-gnustep mailing list
Discuss-gnustep@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnustep


Re: GNUstep on the rise!

2005-08-04 Thread reuss
Sašo Kiselkov wrote:
 Quoting Chris Vetter [EMAIL PROTECTED]:
 
 
A few days ago I read an article mentioning, among other things, GNUstep and
how 'cool' (my choice of words) it would be if/when/whether 'these guys'
(ie. the developers of GNUstep) would decide to put GNUstep on top of their
OWN operating system instead of 'screwing around with existing systems
resulting in symlinks all over the place' (again, my choice of words) and
'imagine GNUstep on a Mach based kernel.'

 
 
 This would, IMHO, be the biggest mistake possible: tying GNUstep to a 
 particular
 platform. I personally love that fact that I only have to write an app once 
 and
 then have it running with little porting effort on Linux, FreeBSD, Mac OS X,
 and even ugly, but spread M$ Windows.
 
 Of course, I'm not against a GNUstep-only-system (hell, I already did it as my
 graduation work at high school, complete with a CD-based installer, integrated
 workspace, it's own package management, etc.), 
is it downloadable?

but care should be taken not to
 make it part of the core libraries.
 
  Saso
 
 
 
 
 ___
 Discuss-gnustep mailing list
 Discuss-gnustep@gnu.org
 http://lists.gnu.org/mailman/listinfo/discuss-gnustep
 
 



___
Discuss-gnustep mailing list
Discuss-gnustep@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnustep


Re: GNUstep on the rise!

2005-08-04 Thread Chris Vetter

On Thu, 4 Aug 2005 12:26:47 Nicolas Roard wrote:
[...]
 Well, GNUstep itself should stay as it is, cross-platform, indeed.

Absolutely.

[...]
 For me the only real option would indeed to have a GNUstep OS
 containing only GNUstep apps, to really have a consistent and powerful
 environment that people can easily try and adopt. Plus, in some way it
 would be easier than a desktop: with a desktop you need to be rather
 tolerant to your host OS, while with an OS you can control everything
 and thus propose original solutions (eg you can decide to use launchd,
 rendezvous, etc.). If people want to adapt the desktop part on their
 own os, they still can anyway.
[...]

Again, I agree. However, rather than creating another 'new' OS ('new'
because it would certainly be another derivative version of UNIX), I'd
preferably go with a 'stripped down' version of Darwin.
That is, 'stripped down' as in 'without DarwinPorts' where possible.

-- 
Chris

-- 
GMX DSL = Maximale Leistung zum minimalen Preis!
2000 MB nur 2,99, Flatrate ab 4,99 Euro/Monat: http://www.gmx.net/de/go/dsl


___
Discuss-gnustep mailing list
Discuss-gnustep@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnustep


Re: GNUstep on the rise!

2005-08-04 Thread Sašo Kiselkov
Quoting [EMAIL PROTECTED]:

 Sašo Kiselkov wrote:
  Of course, I'm not against a GNUstep-only-system (hell, I already did it as
 my
  graduation work at high school, complete with a CD-based installer,
 integrated
  workspace, it's own package management, etc.),
 is it downloadable?

Well, sorry to disappoint you, but it's been more than a year back since I
stopped working on it, and it's been my personal love affair anyways - nothing
public. I never got to the stage where I would have to courage to release it,
because from my point of view it simply wasn't mature enough. On top of that, I
realized that it would have been quite nonsence trying to compete with projects
like Debian in terms of number of packages available for my system, so I cut
out the desktop part and continued work on that. It's available at
http://openspace.adlerka.sk , but I don't have any time to continue working on
it right now. Maybe in the future. If somebody wishes to continue it, I'd
appreciate that - I use it daily myself ^_^.

Sorry for being a bit off-topic here ... ^_^;

 Saso




___
Discuss-gnustep mailing list
Discuss-gnustep@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnustep


Re: GNUstep on the rise!

2005-08-04 Thread Trevor Fancher

On 2005-08-03 14:50:47 -0500, Tabitha McNerney [EMAIL PROTECTED] said:


Hi all,

I have to say that when I read this about DRM and Apple Intel machines:

http://apple.slashdot.org/article.pl?sid=05/08/01/0421248from=rss

it made me realize even more so how important it is that GNUstep
continue to rise to the forefront such that it can be installed on top
of open source systems to avoid nasty DRM!!!

-Tabitha


You may be interested in reading http://osnews.com/comment.php?news_id=11455 .

--
Trevor Fancher
http://trev0r.org/

___
Discuss-gnustep mailing list
Discuss-gnustep@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnustep


RE: GNUstep on the rise!

2005-08-04 Thread Vaisburd, Haim
Cris Vetter wrote:

[...] [ GNUstep separate OS ]

 AFAIK, xMach is dead :-( So does anyone know how far Hurd is with 
 respect to usability?
[...]

Sorry, I deleted the essential part of Cris' letter that I'm going to
reply.

I guess when Chris was talking exclusive GNUstep experience he meant the
distribution, not the OS per se (kernel). It seems the obvious choice
for that distribution would be the Linux kernel. As far as I understand,
nothing in the kernel dictates unix standard FHS, unix standard
configuration
(/etc content), a type of window system or other things that might annoy
a person who experienced NeXT ( I never did ).

On the other hand it's well supported for several architchures and many
periferal devices.

Other post suggested Darwin, I believe it would be as good as Linux,
but why to look for something else?

Tima.


___
Discuss-gnustep mailing list
Discuss-gnustep@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnustep