discuss-gnustep
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: elementary OS


From: Gerold Rupprecht
Subject: Re: elementary OS
Date: Mon, 10 Feb 2014 22:30:31 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20131103 Icedove/17.0.10

Hi Liam,

Thanks for your remarks. I have been supporting GNUstep for a long time,
although mostly lurking the last three years.

The desktop has been created in a rather ambigous manner over the years,
various bits and pieces appeared as it scratched certain contributors
itches.

There has never been an overall decision to actually "create" a desktop.
It has been a rather anarchic creation over the years. In fact there is
also Etoilé, so there are actually two GNUstep desktops using your
definition.

As far as distributions are concerned, Riccardo has had some success
with the HURD operating system, which is also in a long slow development
showing that the code is highly portable (despite the almost
insignificant usage). Getting better packages in Debian is definitely a
laudable goal. What can be done to improve the current situation?

WindowMaker is a great piece of software, lightweight, efficient for
what it does and still usable after all these years, but the original
author has moved on some years ago. Could this be the basis for a better
integration of the other GNUstep apps? Probably, but at what cost and
what is the strategic value?

I think GNUstep needs to rethink along the lines of the Boston
Consulting Group and do a SWOT analysis:
Strengths,
Weaknesses,
Opportunities,
Threats

I would like to see if and how certain forks can be brought back into
the GNUstep mainline. I see Cocotron advancing in their own corner
(focusing on Windows platforms). I see Nikolaus Schaller's branch which
was necessary for certain small machines which had no floating point
hardware many years ago.

Nikolaus has written some superb software, and  launched the Simple Web
Kit which needs some features to be able to handle help viewing and then
eventually could be expanded to a full fledged web kit in objective C.

For Nikolaus to work on small portable machines with GNUstep, the
cross-compiler tools need to be made more easily upgradeable and have
automated regression tests. Right now I believe he and Riccardo almost
always are compiling natively on their smaller machines. I believe
Nikolaus is having to work with GCC for his use cases.

Right now I see more and different hardware showing up, particularly ARM
processors. Having two compilers available should make porting GNUstep
and any derivatives anywhere much simpler, more attractive.

Can the tool chain be improved so the GNUstep main contributors are able
to work in their areas of interest more effectively?

The newer syntax in Objective-C require some hacking on GCC, the
compiler collection. David Chisnell is very capable, but has been
working mostly on Clang. Getting similar features into GCC would be a
big plus. David, I would love to hear your thoughts if this would be a
fruitful endeavor to broaden the appeal of Etoilé?

 Has there been any discussion as to what compiler changes would be
needed for Nikolaus to consider merging back into GNUstep?

As for the GS office suite, the Lighthouse Design applications are still
gathering dust. I would love to see these back as open source. Maybe
another try to get this code republished would be appropriate. Perhaps
some people's attitudes will have changed with the slowdown in the PC
market, and the need to diversify to other platforms.

The above market tendency might give an impetus for the people selling
the Eggplant software to open source it. This is an excellent software
testing tool based on Objective-C. They will need to move on to other
platforms as the PC market shrinks again in 2014. Mobile platforms with
different processors is where the current market is going...

What areas of collaboration are of most interest for the contributors?
Maybe some short, medium and longer term goals can be prioritized for 2014?

Thanks to all who have contributed to GNUstep in the past and look
forward to your responses on this list.

Best wishes,

Gerold


On 10. 02. 14 17:20, Liam Proven wrote:
> On 9 February 2014 20:46, Gregory Casamento <greg.casamento@gmail.com> wrote:
>> So, I hope you'll fully understand when I say that I'm banning you from it.
>
> Shocking, petty, vindictive and wrong. Terrible, terrible moderation.
> "You are saying something that is not permitted so I am banning you."
> It's too Orwellian for words.
>
> There is more good sense and well-made points in that one post than in
> anything else I've seen on this mailing list in over a year.
>
> The GNUstep project seems to be in deep denial, about what it is, what
> it does, why it exists.
>
> GNUstep is a desktop.
>
> No, it is *not* just a set of libraries. Nobody knows or cares about
> that; every leading desktop has a set of libraries and dev tools.
>
> KDE is a desktop, GNOME is a desktop, Xfce is a desktop. What this
> means, today, is a whole set of tools, libraries, a window manager and
> desktop manager, a suite of apps sharing that look and feel, and a set
> of tools and libraries for building new apps that share that look and
> feel. That is what GNOME is, that is what KDE is, and if those are
> desktops, then that is what GNUstep is as well.
>
> But the integration is shockingly poor. The official GS WM, Window
> Maker, has its own dock and its own menus which GS ignores. The GS
> desktop seems, as far as I can tell, to offer no way to let the user
> access the WM menus, so you can't open non-GS apps. You can't even
> readily add apps to the GS dock, or move it. The packages for GS in
> the leading distros - i.e. Debian and therefore Ubuntu - are
> apparently years out of date.
>
> There is no distro with GS as its desktop.
>
> There is no GN office suite or web browser, nor even effort to integrate one.
>
> The GS devs seem to have spent years fiddling with their back-end
> libraries while the front-end demo apps are woefully neglected and
> feature-poor. There doesn't seem to be any effort to reach out to the
> distros and ensure that the bundled libraries are current. There
> doesn't seem to be any effort to actually get GS out there in a form
> which people can see and try.
>
> It doesn't matter how good your libraries are if nobody knows you
> exist. Nobody will code for a platform which cannot be tried or even
> seen, and GS is completely below the radar, and nobody involved even
> seems to notice that this is a problem.
>

-- 
Gerold Rupprecht
Mobile +41 79 914 29 52


Attachment: signature.asc
Description: OpenPGP digital signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]