[Top][All Lists]

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

Re: So, honestly, is GNUStep a viable development option?

From: Riccardo
Subject: Re: So, honestly, is GNUStep a viable development option?
Date: Tue, 13 Nov 2007 20:04:25 +0100
User-agent: GNUMail (Version 1.2.0)


this discussion is a bit sidestepping and taking up precious time to reply, still, let me make some short commrents.

On 2007-11-13 14:37:39 +0100 "Mark Grice" <address@hidden> wrote:

If UBUNTU has shown us nothing, it has shown us that.
steal the work of others, make up some bad packages which are often broken, hide everything behind some shiny desktops and setups and call it a distro?

I get a sense that most GNUStep developers today were NEXTStep (or
OpenStep) developers in the past. That is fine, but you have to
realize that you started into GNUStep with a lot more understanding of
this is pretty wrong. Although some people here used NeXT, I come from the Mac and Unix world. One day I wanted to develop. Also, I was tired of that shiny, bloggy, yummy look that most linux stuf has and which just wastes my resources. I was happy with apple but found it just un-professional looking, furthermore the more macos evlolved the less I could share some of its choices.
I speak for myself, but I know of others.
And I was a total newbie. I fo und installing gnustep extremely easy on debian: just install the packages and start the stuff from windowmaker menu. Sure that was the beginning. Quickly I wanted my own environment and reading the tutorial from Dennis I set up my environment. Easy. Easier than compiling gnome or kde! just get the release tarballs, configure, make install. Once you need to source that file or you can put it in yor shell script.
But if you want to develop, you better master these small tricks.

I have gone on Amazon and found used books on NEXTStep development.
I've ordered them and hope they will help fill the gaps I am missing
to understanding exactly what GNUStep gives me. That isn't a viable
strategy for everyone though. (For one thing, the source of used books
will one day dry up...)
OpenStep would have been better.
But many tutorials for the old Rhapsody or early macOS just apply 1:1! The dev. environment is a bit different, but gorm operates like interface builder and once you get that "make" is needed (but ProjectCenter hides that for you) you can just redo them.
Also this appears a world where you need to feed everybody.
Gnustep has an example package with some apps which cover many aspects of development. Furthermore being most applications opensource you can just take them and dissect them. Bootrap is easy, you need just a couple of things - understanding Objc. The NeXT or Apple book is perfect for that and freely available - knowing your dev. environment. Either ProjectCenter+Gorm or your makefiles with vi/emacs.
- get a general grasp of Foundation and AppKit

then just start coding, the rest comes by itself.

1. Stop the mailing list and put up a forum. That is the preferred
method of communication for most people these days. This takes about
one day to do, and costs little or nothing. I would be happy to set up
the basic structure for this if anyone wants me to, but I obviously
can't fill in all of the information. I'd even be willing to register
the domain and pay for the first year (and maybe more, but I can
easily commit to a year...)
forums are terribly time consuming. A mailing list is really better. And I use forums elsewhere, so I know what they are.
If the mailing list bothers you you can use the gmane news gateway.

There is also an IRC channel.

2. Detailed installation instructions. PLEASE. I actually had very
little trouble getting GNUStep on Ubuntu. I used the Synaptic Package
manager, clicked everything even resembling GNUSTEP, and had no
problems at all. HOWEVER -- The GNUStep.org page still includes a link
for UBUNTU that is almost 3 years old! I read that and was so
confused, I almost gave up!
help updating the wiki. Or think before giving up.

And, I still read that some people can't manage to get it installed.
This is a real problem.  We need instructions that even a Linux/Unix
noob can follow. These people are a GREAT source of new converts
because a lot of them haven't made their mind up yet about GNOME vs.
KDE, and are open to new development environments.
a developer is never a noob, so the instructions just need to be up to date. But this consumes time and resources. Users should just be able to use packages form their distro, but this requires them to be well set up.

3. Update the examples. It's nice to have them, but with no document
to describe them, and virtually no comments in the code, they aren't
very helpful. And make sure they work on the main Distros. (i.e.
Debian, Ubuntu, and Redhat)  My PC/GORM problem is the prime example
of something that should never happen to someone following a script.
I commented that above. Your problem is related to some setup or a bug or both. Gnustep runs on a much more diversified environment than KDE :)

3. More tutorials. More examples. My suggestion: Scour the net for the
new KDevelop and GLADE examples. Take one and re-do it in GNUStep.
SHOW people why working with GORM and GNUSTEP cuts down on their
programming time. I'll be happy to help out here -- but first I need
to learn this myself...
commented above.

A) Update the look and feel. If that means Cameleon, great. If it is
Etoille. Fine. But there should be ONE PLACE to get this. That should
be the MAIN ROOT, not a branch. If someone wants to keep the old style
NEXTStep look and feel THAT should be a branch. The main code has to
move forward in this area or no one will take GNUStep seriously.
the main code should remain main code. Full stop. I agree with you though that gettin the stuff means searching over scattered places. Ideally one should just to be able to install "package X" to get look X by using its package manager. I also propsoed a section of the website where to cater themes to download and each with a quick screenshot. People who love themes lvoe that kind of stuff too (think of skins of opera or seamonkey)

B) Native widgets. Why not? I used to work for Neuron Data about 15
years ago. We had a product called Open Interface that provided a
cross-platform GUI. It was great when we started -- a superset of all
windowing environments... but we couldn't keep up. Critical mass was
against us. Our developers kept telling us that there was no way to
use native toolkits then... suddenly... there was. It made a
difference. Now if I opened a file dialog box it looked like everyone
else's. That is HUGE to a user. I can't emphasize it enough...
leave the native wiedget duscussion out of this thread. What themes should support though is the emulation of a native look. This is what original openstep did, btw. It worked fine on win32.

I cut the other points since I didn't think of a godo reply to generic comments of them.

I just think ther eis no "before" and "after" stages. It is a flow. Better website developed along beter and more applications developed alongside better themes, etc.... It is even happening as now, it just takes time out of our few resources


reply via email to

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