GORM 1.0 Release to Take on GNOME/KDE?

qa'lth writes "Today marks the occasion of the release of Gorm 1.0, the Interface Builder for the GNUstep project, and with its release, comes the obsolesence of the GNOME and KDE projects. Finally, today, Free Software users can enjoy the power of a well-designed, powerful object-oriented system derived from OpenStep, legacy to the acclaimed MacOSX, through GNUstep, our loving reimplementation of the OpenStep standard."

Comments

Riiight.

[coolGuyZak]

with its release, comes the obsolesence of the GNOME and KDE projects

Riiight. 'Nuff Said.

Re:Riiight.

[molnarcs]

Yeah - I don't know which is worse: 1) making such claims just for publicity (flamebait?) 2) or truly believing in it. In either case, the first screenshot you bump into will discredit their claim immediately. Compare it with anything trolltech has to offer with qt4 (or kde4's plasma efforts, koffice kids, etc.) and their development tools... I don't mention GNOME development tools because I'm not familiar with them, but I don't think they will be "obsoleted" either.

Re:Riiight.

[GreyWolf3000]

I think it was meant as a joke.

However, gnustep has been themed, and it can look pretty good. From a UI perspective, I really like how consistent and polished the interface is, even when it's in the default "prosaic" grey. And it's not only easy to learn, it's also easy to use. From a usability perspective, I think it's much more intelligently designed than Gnome or KDE.

Re:Riiight.

[borgheron]

My fault, that screenshot is rather old. :) The look has improved quite a bit since then. I'll have to upload an new one.

GJC

Re:Riiight.

[Anonymous Coward]

First, theming engine inclusion into the main GNUstep sources is in progress. Second, GNUstep appearance was lower priority than functionality until now. Third, development is still mainly community based, compared to KDE/Gnome with large sponsors. Knowing this, I would say that the community is doing good work.

Re:Riiight.

[minus_273]

the screenshots on that page are horrible. look at these for a batter idea:
the dock

like GTK, Everything looks better once it is themed

look at this

The new icons are really nice too
gorm
help
installer

Re:Riiight.

[molnarcs]

I think it was meant as a joke.

Then the guy must have a very good sense of irony ;) It is interesting to see how (at the moment) slashdot tries to decide whether it was a joke or not :)) I wonder which reading of this story will win out :) Currently I can't decide - but yeah, at first sight, it didn't occur to me that it was a joke :)

I'm not a developer - but follow the development of various desktop closely (mainly KDE and enlightenment). I'm also a thinkerer, and I like to try out alternative desktops once in a while, including Afterstep, windowmaker, and the likes (that follow the same UI paradigm seen in gnustep) - and I noticed that there was very little or no development at all of these desktops in the past few (3-4) years. I have to admit that both wmaker and afterstep are different from other desktopts, but I won't apostrophize that difference as revolutionary. And I don't see where it would take (even with rapid development tools) the current desktop paradigms (or how it compares to future ones, like plasma). In short: I don't see the vision, the why this is soo cool aspect. You wrote: "From a usability perspective, I think it's much more intelligently designed than Gnome or KDE." - yeah, but whose usability? It is really really difficult to define an objective usability perspective. I don't dispute your claim, I just don't understand it :))

Re:Riiight.

[sznupi]

"Their" clame? Are YOU on drugs? Just some "qa'lth" said it... BTW, not only is the appearance skin-deep (and apparently now they want to be faitfull to NextStep), but judging useability of programming tools by screenshot...I've lost my words here.

Re:Riiight.

[Anonymous Coward]

>>I'm also a thinkerer

Does that mean you tinker around with thinking? :)

Re:Riiight.

[wangmaster]

I love it. I swear people take these things way too seriously. I get the feeling that the vast majority of people on slashdot are computer geeks with no social skills, hence the inability to get a joke.

That said, the OPENSTEP/NEXTSTEP UI has largely been considered one of the most elegant and usable interfaces to have been created. It's extremely intuitive, and while the GNUSTEP work isn't there yet, the "completion" of Gorm.app is a very good sign, as the interface builder is the foundation to creating the wonderful UI of classic NeXT applications. NeXT spent a ton of money hiring some of the best UI designers in the world, and the UI shows it. People laud over OS X's ability to hide the Unix from the newbie user, but I don't think many of them know that this had existed since the '90s beginning with NEXTSTEP.

Apple broke alot of the inherent intuitiveness of the NeXT UI when they moved to OS X, which isn't necessarly bad though, as intuitiveness != familiarity and their changes were mainly to make existing Mac OS users comfortable.

I'm impressed that the GNUStep project is still able to have milestones like this.

Re:Riiight.

[diegocgteleline.es]

Indeed. And I'd like to know why it's "well designed", "and powerful object-oriented", too. I mean, just because mac os x managed to write the best desktop platform from openstep doesn't means qt and GTK aren't capable of the same.

Re:Riiight.

[gentlemen_loser]

Trolltech is a COMPANY - not an open source trust. As much as I like KDE and Qt, you need to keep in mind that they can go away in an instant if Trolltech is bought, the license changes, ect. The counter argument here is that the license is written in such a way that it provides for such an event. Maybe so - but the threat of a (or actual) lawsuit (even if it results in a loss in the long term) would be enough to kill the project or at least result in companies running from it.

GNOME's APIs suck (or at least they did a year or so ago) - its that simple. Unless there has been (or will be) a major restructuring - its NOT fun to play with.

GNUStep gets us closer to a completely open stack. Its well written and well thought out.

Re:Riiight.

[Anonymous Coward]

thinkerer? apostrophizes?

Welcome to slashdot, Mr. President!

Re:Riiight.

[filesiteguy]

Agreed. I looked at the video. My first though is that is is as pretty and user-friendly as OS/2 2.0.

Next!

Re:Riiight.

[Archibald Buttle]

Yes folks.

As we all know, how pretty your user interface looks is the best barometer of how easy it is to use, how simple it is to program for, how much leverage applications get from the underlying API, and how powerful applications written using the underlying toolkit will be.

American readers should append the word "NOT" onto the end of the last paragraph.

Gorm and GNUstep are mostly about the underlying API. It's possible to rapidly build incredibly powerful applications using Gorm - significantly faster than you can with the KDE or GNOME toolkits.

Re:Riiight.

[ErichTheWebGuy]

HA! I literally laughed out loud at that screenshot.

Re:Riiight.

[Anonymous Coward]

Ya know, being able to come up with very good justifications for something sucking doesn't change the fact that it sucks.

Re:Riiight.

[Ulrich Hobelmann]

You made the mistake to judge something by its superficial impression, not its power.

The very same windows as in that screenshot can run just the same on Mac OS X and then they suddenly look cool :)

Newer pictures of GNUstep I've seen also look quite good, definitely comparable with the ugly Gnome (pre-Clearlooks) or KDE default themes.

Re:Riiight.

[Anonymous Coward]

Just my $0.2 here, but...

I follow the GNUstep-dev mailing lists (big fan, myself), and I can assure you the poster was making a joke. Among the GNUstep developer community I do not think there is any misconception as to how "complete" the system is. I use it, and I write apps for GNUstep using Gorm and ProjectCenter, but there is a lot of ground to cover before obsoleting the likes of GNOME.

Amusingly, there was recently a thread about trying to get the release of Gorm 1.0 announced here. Plenty of folks said that they had submitted the story numerous times, only to be rejected. There is perhaps a feeling that the editors deny the existence of anthing that isn't KDE/GNOME, Microsoft/Google/Apple, etc., and perhaps the poster just didn't have very high hopes that the story would even get through.

Re:Riiight.

[swillden]

As much as I like KDE and Qt, you need to keep in mind that they can go away in an instant if Trolltech is bought, the license changes, ect.

No, they can't. If Trolltech stopped dual-licensing under the GPL, either because of a decision on their own part, or because they were bought out, that would not change the existence of the already-released source code. Trolltech could release a snazzy new version with a lot of attractive features and do it under a different licensing structure, but the code that has already been released under the GPL is here to stay.

Oh, and Trolltech doesn't own KDE, and can't make any decisions about the licensing of that code.

GNUStep gets us closer to a completely open stack. Its well written and well thought out.

No argument there. NeXTstep was a beautiful environment and the NeXT Interface Builder was perhaps the best development environment I've seen. Some of the modern C++ and Java environments have arguably caught up to it, and perhaps even surpassed it in some areas, but only just.

That said, even if we had a complete open source implementation of NeXTstep, I doubt I'd move away from KDE quickly... NeXTstep was awesome for its time, and it's still extremely elegant, but compared to modern desktop environments it's feature-poor. That can be fixed, of course.

Re:Riiight.

[/ASCII]

I love it. I swear people take these things way too seriously. I get the feeling that the vast majority of people on slashdot are computer geeks with no social skills, hence the inability to get a joke.

I thought it was pretty funny as well, but it is well known that irony doesn't travel well over the Internet, so the author probably should have phrased things differently.

Re:Riiight.

[Elranzer]

From a usability perspective, I think it's much more intelligently designed than Gnome or KDE.

Uh oh, don't say that here! There goes your karma...

Re:Riiight.

[Anonymous Coward]

Right. It's good that GNOME and KDE are no longer needed, since they were a major roadblock to users adopting Linux as a desktop OS. I'm happy to see projects taking on these roadblocks, as opposed to say, oh, cooperating with each other and taking on Windows instead as a community. Way to go!

Re:Riiight.

[flacco]

I'm also a thinkerer, and I like to try out alternative desktops once in a while,

just a little more attention to your spellering and your grammarishness and you'll have the complete package.

Re:Riiight.

[drinkypoo]

gnustep is a whole hell of a lot more than a desktop, though. we're talking about a whole method of software development.

Re:Riiight.

[molnarcs]

You made the mistake to judge something by its superficial impression, not its power.

Yeah, I admit that - the problem is, that most WMs I saw that were based on gnustep's concepts weren't much different (afterstep, wmaker). Now I can accept that there is a possibility to make something as beautiful as e17 with gorm, I just don't see it. Well, we have to wait and see... Also, I used wmaker for a while, and although it was intriguing (it was different afterall) I don't share the apparently common view that it was something revolutionary, or more advanced than $your_desktop_of_choice UIwise.

Re:Riiight.

[Anonymous Coward]

> Uh oh, don't say that here! There goes your karma...

...and there goes your joke. ID subscribers don't believe in karma and could give a crap about a nerdy atheist /. mod whos just reached puberty. Karma is for wussies and the posters that use it. Only an AC has teh hairy ballz to say it like it is.

check deez nutz foo!

Re:Riiight.

[Ulrich Hobelmann]

Well, yes, Enlightenment rules :)

The cool thing about GNUstep and Gorm is the object oriented development environment, and the libraries for Objective C. The e17 libraries are structured in a totally different way, and I think they only use pure C. Maybe one could write ObjC wrappers for the e17 libs and integrate them into Gorm...

Wmaker is cool; I mostly liked its performance, clean looks and the quick workspace switching (alt-1 etc.) though. Curious how e17 will turn out (as it is, it doesn't compile on my Mac)...

Re:Riiight.

[NoMoreNicksLeft]

What, you must mean apps? Well, I doubt someone will bother redoing firefox in GORM, but even for those KDE and gnome apps that you just must use, get a gnustepish theme. Gtk2step (gnome) and newstep (kde) both come to mind. Much nicer than the fugly widgets they have by default. I've even been working on an opera skin, myself.

Future projects of mine: patches for gimp/inkscape (ala gimpshop) that will move vertical scrollbars to the left, and replace file dialogs with Next-styled ones.

And for those that want things just a bit more OSXish than NeXTish, try out skippy (exposé) and kxdocker (OSX dock). Both work well in windowmaker (kxdocker needs you to upgrade to the latest windowmaker and you have to edit the configuration manually). My desktop looks somewhat weird, what with the NeXT dock on the left side for dockapps, and kxdocker at the bottom to lauch apps (had 4 minidocks at one point). Heh, and the orange on black xterms so that they look like old monochrome amber terminals...

Re:Riiight.

[Synic]

Wow! I had no idea that GNUStep could be themed at all! Artists please make-over and demand GUI features to fully make over this system into a modern looking piece of beauty! ^_^

Re:Riiight.

[orcrist]

No kidding. Talk about the a way to get Gnome and KDE zealots on to team up on you ;-)

-chris

Re:Riiight.

[PlacidPundit]

Here's the latest Camaelon theme engine. There's a really nice theme called "Nesedah" out there.

Re:Riiight.

[WuphonsReach]

I can't decide whether the poster was trying to coin a new word ("thinkerer") along the lines of being a smarter tinkerer, or if it was a simple spelling error.

Huh... Google turns up hits for "thinkerer".

Re:Riiight.

[TheRaven64]

GNUstep is not a window manager. GNUstep is a set of core libraries for developing applications. It uses Objective-C, which is a wonderful language to work with. Things like run time introspection are provided for free, while things like Qt need to hack them on top of a language that simply wasn't designed to work with them. The syntax is very simple - it only adds a bout half a dozen syntactic elements on top of C, and provides a true object oriented language in the Smalltalk sense.

The GNUstep API is that of OpenStep. You may have heard of this. It was created as a collaboration between NeXT (who already had a first-rate development environment, and who produced the first mainstream RAD tools) and Sun. It was used by Tim Berners Lee to write the first web browser - he later said it wouldn't have been possible with any other environment.

GNUstep is split into two main components. The Foundation library provides things like strings, basic data structures, and all of the things that are needed to write the back-end of an application. The AppKit provides all of the tools for writing a GUI.

GORM, discussed here, is an object modelling tool. You create instances of classes (including GUI classes), join them together, and save them as a serialised object graph. You can then load and manipulate this graph in your code. One example of this is to create a document window with connections to your document controller object as a GORM file. You would then simply instantiate a copy of this object graph in one line of code to create a new document window with a document object attached. This is a very powerful way of programming.

GNUstep works well on *NIX, and fairly well on Windows (support is being finished - more or less everything works, but some things are a little rough around the edges). Code written for GNUstep will almost always compile on OS X with Cocoa (which is another implementation of OpenStep) with almost no modifications.

Programming with OpenStep is a joy to do - you end up with nicely abstracted code built around good design patterns almost without even thinking about it.

Re:Riiight.

[molnarcs]

Thanks for the information :) It would be interesting to see if gorm can accelerate developments of WMs based on gnustep. I understand now that gnustep is not a wm, but those WMs that are described as "based on gnustep" saw very little development in the past few years. At least that was my impression while trying out afterstep and windowmaker.

Re:Riiight.

[Arandir]

I get the feeling that the vast majority of people on slashdot are computer geeks with no social skills, hence the inability to get a joke.

GNU has a very poor sense of humour. GNU takes everything way too seriously. GNUstep is a GNU project. Hence the assumption that it was not a deliberate attempt at humour.

Re:Riiight.

[rsheridan6]

They should really make one of those the default. I just rejected GNUstep a week ago when I was setting up a system for my wife based on those screenshots. I thought it looked just like WindowMaker, but uglier, so there was no reason to use it. When you're quickly evaluating several different WM/DEs, you're not going to dig deeply into any of them and find out what they look like themed. I ended up going with XFCE and she seems happy with it.

Re:Riiight.

[Anonymous Coward]

The interface builder is only one part of many in the IDE. Unfortunately, the GNUstep people don't have many screen shots of GORM for you to compare with Qt Designer, so you would probably have to download it and try it yourself. While it seems to me GORM does have some catching up to do with Qt Designer, I don't know how much. I've seen the interface builder that comes with Apple's Xcode. While it's not as impressive looking as Qt Designer, I doubt it's lacking much of anything that Qt Designer offers you, and it's probably much simpler to use. So a screen shot isn't necessarily going to give you the whole story. But please don't compare GORM with the entire Qt tool set, that's just plain silly.

Re:Riiight.

[roard]

Well actually, neither Afterstep or WindowMaker are based on GNUstep -- they are written in C, not Objective-C, and they don't use at all the GNUstep libraries.

Re:Riiight.

[mad.frog]

But is it possible (or practical) to use it with anything *except* Objective-C?

I realize that many people seem to love it, but I really, really dislike it... the syntax is wonky (IMHO) and far from my favorite language.

If I could code in something like EcmaScript/JavaScript/ActionScript, that might make it much more interesting... but any kit that requires I use Objective-C as my primary language is a nonstarter as far as I'm concerned.

Re:Riiight.

[jcr]

NeXT spent a ton of money hiring some of the best UI designers in the world, and the UI shows it.

Umm.. They got a world-class designer, but NeXT didn't spend a ton of money on Keith Ohlfs. He got rich on his WebTV shares.

-jcr

Re:Riiight.

[vettemph]

Well i just paid $400.00 USD on an LCD monitor so i'll stick to something that going to look good.

PS- if this GORM stuff is able to look good it should have been shown that way in the screendump.

Re:Riiight.

[Art Tatum]

A lot of people in the GNUstep community are religious about the old NeXT look and don't like to put themed screenshots up.

Re:Riiight.

[DianeOfTheMoon]

Just my $0.2 here, but...

Why thank you for your generous tip, but around here, ideas go for at most $.02...

You'd think that with with this being a bastion of open-source, you wouldn't have to pay anything at all, but that's capitalism for you...

Re:Riiight.

[kitzilla]

These are really, really nice -- as good as anything out there, including the commercial GUIs.

Congrats to whoever designed such a clean and thoughtful theme.

Re:Riiight.

[minion]

Yes folks.
 
As we all know, how pretty your user interface looks is the best barometer of how easy it is to use, how simple it is to program for, how much leverage applications get from the underlying API, and how powerful applications written using the underlying toolkit will be.
 
American readers should append the word "NOT" onto the end of the last paragraph.
 
Gorm and GNUstep are mostly about the underlying API. It's possible to rapidly build incredibly powerful applications using Gorm - significantly faster than you can with the KDE or GNOME toolkits.
 
  Lets not forget the speed at which GNUstep runs. Why do we all accept the fact that computers keep getting faster, but our desktops keep getting slower? I can guarantee you my 386DX/33 with 8M of RAM would navigate within Windows 3.1's menus, and default programs FASTER than WinXP on a modern machine. Its sad that none of us complain more to the GUI developers (Gnome and KDE are NOT exempt from this either!). There is no reason for the bloat that we're all accustomed to.

Re:Riiight.

[Jambon]

Could something explain to me how the QT and GTK are different from things like Enlightenment? I understand they make different themes and such, but how does application building work? Would GNUstep be compatible with something like enlightenment?

Re:Riiight.

[grammar fascist]

As we all know, how pretty your user interface looks is the best barometer of how easy it is to use, how simple it is to program for, how much leverage applications get from the underlying API, and how powerful applications written using the underlying toolkit will be.

American readers should append the word "NOT" onto the end of the last paragraph.

Why? Is it different elsewhere in the world?

No wonder you're all screwed up.

Re:Riiight.

[Anonymous Coward]

> Agreed. I looked at the video. My first though is that is is as pretty and user-friendly as OS/2 2.0.

> Next!

You're completely right. NeXT was the company that created the OpenStep standard

Re:Riiight.

[RevAaron]

I agree. I am of the mind that NeXTSTEP and OpenStep were the nicest OSes I've ever used. The GUI was the most elegant, with so many darn handy features.

Re:Riiight.

[Anonymous Coward]

last time i checked, enlightenment was still a powerful GUI window manager, and libs for GUI things. GNUStep is a implementation of the OpenStep API standard, its a good standard, ill tell you. So they are quite diffrent, esp since GNUStep isent exactly pretty, or as themeable as enlightenemnt would like.

In any case, GNUStep comes with Gorm to build GUI's, its like glade from GNOME, GNUStep also comes with ProjectBuilder, its a good way to setup new projects, but last time i checked a few months ago it was still lacking in a number of areas to be able to use it to really handle and edit a full project.

QT and GTK are toolkits, they draw on the screen, and are easer to program for then Xlibs, and also are platform independent. Enlightenment is just a desktop environment (or is trying to become one), like KDE and GNOME are desktop environments.

Re:Riiight.

[Thing 1]

That's way too many smileys.

Are you on drugs?

Then give me some!

(RIP Mr. Kinison)

Re:Riiight.

[Lproven]

Not eveyone shares your tastes for flashier coloured interfaces. Personally, I think all your suggested changes are much less pleasant to look at than the original. NeXTstep is a thing of beauty, restrained and subtle and elegant. Adding themes to it is like draping Michelangelo's "David" in pink lace ruffles and frills - it merely detracts from the original beauty.

It's themable. Go play to your heart's content, but don't force your changes on anyone. The original is simple, plain, unfussy and easy on the eye. You may find it boring but to lots of people plain and simple is actually attractive.

For my two-penn'orth, the original scheme looks ten times better than any theme on any Linux GUI I have ever seen. KDE is pug-ugly, GNOME is only marginally better. The only half-way professional looking themes out there are Redhat's Bluecurve and Ubuntu's default muted earth-tones. All the others look like tart's handbags.

Re:Riiight.

[DariaM84]

Yeah. People aren't going to, at the drop of a hat, abandon a GUI that's working for them. My philosophy is always, "If it ain't broke, don't fix it."

Re:Riiight.

[Dave2 Wickham]

GNU takes everything way too seriously?

Whiskey Tango Foxtrot!!!

[FreeLinux]

with its release, comes the obsolesence of the GNOME and KDE projects.

WTF??? Not even Microsoft would dare make such blatant and patently false claims. I'm all for marketing but this is unadulterated bullshit and I don't even want to look at something that starts with BS like this!

Re:Whiskey Tango Foxtrot!!!

[Elektroschock]

Guess this was a joke. :-) However, is it possible to reimplement the Mac OS X APi based on GNUStep?

Re:Whiskey Tango Foxtrot!!!

[tezbobobo]

Actually, it may not be BS. Time will tell that. It isn't so much 'patently false.' so much as 'unsubstantiated.'

And actually it is very similar to microsoft claims. Think of times they made claims about there killer software which had not even been released yet!

Re:Whiskey Tango Foxtrot!!!

[grahamlee]

Both OS X's Cocoa and GNUstep are based on the same API specification; "OpenStep" jointly written by NeXT Software and Sun Microsystems in the mid-1990s. Code which targets OpenStep will work on both Cocoa and GNUstep - however from thereon there are divergences...GNUstep has classes which Cocoa lacks and vice versa. It would be possible to reimplement the missing classes on the other system, just it hasn't been done.

Re:Whiskey Tango Foxtrot!!!

[SavvyPlayer]

GORM also obsoletes XTerm, Vista, Web 2.0, the Automobile and the Universe in general.

Re:Whiskey Tango Foxtrot!!!

[sznupi]

I haven't noticed that GnuStep guys said it. Just some individual named "qa'lth" did it...

Re:Whiskey Tango Foxtrot!!!

[Alioth]

GNUstep is exactly how we ported Oolite (an open source game for Mac OS X) to Linux and FreeBSD. There were a few small issues to contend with (and we eventually went from using NSOpenGLView for the graphics to SDL) but 99% of the code is identical on OS X and Linux.

It's a pity that GNOME was written way back when instead of GNUstep being the free desktop of choice - had all that effort gone into GNUstep, it would have been pretty easy to target both Mac OS X and Linux/*BSD instead of having to write separate UI code for each.

Re:Whiskey Tango Foxtrot!!!

[1336]

Ladies and Gentlemen, step right up! Who will be the first to gaze upon the amazing GORM?!? Descended from the same stock as the legendary Apple! So powerful, it can beat both Gnome and KDE with one hand tied behind its back! Setp right up! Step right up!

(and then you get inside and its just X11 with a horn glued to its forehead ):

Re:Whiskey Tango Foxtrot!!!

[JudgeFurious]

What was so hard about picking up the intentional humor in that submission?

  Before I even opened the link I thought it was pretty clear that he was joking about that and seeing the screenshots only confirmed that.

Re:Whiskey Tango Foxtrot!!!

[LizardKing]

It's a pity that GNOME was written way back when instead of GNUstep being the free desktop of choice - had all that effort gone into GNUstep, it would have been pretty easy to target both Mac OS X and Linux/*BSD instead of having to write separate UI code for each.

Miguel de Icaza and the other founders of the GNOME project did consider GNUstep. Their conclusion was that the amount of work needed to finish GNUstep, and the added burden of staying compliant with future OpenStep specifications weren't worth the effort. They weren't to know that NeXT would find a massive new audience as the new MacOS ...

Re:Whiskey Tango Foxtrot!!!

[Anonymous Coward]

GORM may obsolete the Universe, but KDE obsoletes the Multiverse, so I guess KDE wins.;-)

Re:Whiskey Tango Foxtrot!!!

[Elektroschock]

So how many manyears have to go into GNUStep?

Re:Whiskey Tango Foxtrot!!!

[SEE]

They weren't to know that NeXT would find a massive new audience as the new MacOS ...

NeXT was purchased by Apple in 1996, with the public announcement that NeXT's OPENSTEP operating system would be the basis for the next-generation MacOS. The GNOME project was announced August, 1997. There were lots of reasons not to go with GNUStep, but that OpenStep was the API of the future MacOS was well-known.

In fact, it's prety clear that cross-platform development wasn't a major consideration in toolkit selection. When GNOME was announced, the OpenStep API was already available on Solaris and, as OPENSTEP Enterprise, for Windows NT 4.0.

let the flames...

[know1]

..BEGIN

That's confidence for you...

[meringuoid]

and with its release, comes the obsolesence of the GNOME and KDE projects.

Nothing like a little optimism, eh? Quick question, what are you smoking and who's your dealer? I gotta get some of that stuff...

Re:That's confidence for you...

[Anonymous Coward]

Nothing like a little optimism, eh?

Did you mean: opium

Re:That's confidence for you...

[b100dian]

At least he's not smoking cracks for their own software...
Err.. wait..

Re:That's confidence for you...

[Anonymous Coward]

Personally, I think it's hilarious that the main developer of Gorm is complaining that Slashdot was rejecting his announcement.

http://heronsperch.blogspot.com/2005/11/slashdots- incessent-rejection-of-gorm.html

In case he decides to take it down/delete it/have it 404... Here's the full text:

I'm sorry, but am I the only person in the world getting pissed off at slashdot for rejecting my posts? Does it seem like slashdot only worries about certain peoples posts and indeed only posts on certain subjects?

The site has increasingly become an GNOME/KDE site over the course of the past few years. It's become apparent to me that no other competing API toolkit will be able to edge it's way into that site so that it can get some attention.

It's time to make a better slashdot.

Re:That's confidence for you...

[Anonymous Coward]

It's time to make a better slashdot.

To be released in 2013!

Re:That's confidence for you...

[roard]

Well, don't you think it's a bit sad to actually get _this_ troll article instead of a proper article, with links and good explanations about Gorm and GNUstep ?

.. hmm perhaps not actually, that way people can have fun on /. while some gnustep devs are here trying to explain their project...

Desktop wars.

[ExtraT]

When will it end? :(

Re:Desktop wars.

[molnarcs]

"Desktop wars When will it end? :("

Hopefully never! As long as it lasts, there is competition, meaning rapid pace of development, choice, etc..

Re:Desktop wars.

[DaPoulpe]

You have to pay attention son, they just said it was over 'coz OpenStep won :)

Re:Desktop wars.

[Anonymous Coward]

Because I can't use a desktop unless when windows open, they look like bunched up fabric gradually smoothing out, and when they close, they evaporate into puffy vapors. Desktop environments are so much BETTER than old command line / text menu systems. You wouldn't believe how much higher your productivity is NOW.

who cares

[Anonymous Coward]

How about decent and WORKING drag and drop?

everyone is busy with eye candy and other useless add-ons and ignore basic operability and useability.

Re:who cares

[grahamlee]

GNUstep has drag and drop, and doesn't have much eye candy.

Re:who cares

[Taladar]

Drag and drop is a solution looking for a problem IMO (outside of graphic or 3D design apps where you move vertices, selections,... that is). Requiring the user to either rearrange windows or wait seconds until the taskbar realizes you didn't just place your cursor on the minimized window by accident is just retarded from an efficiency point and getting users to understand the concept is much more difficult than for any of the alternatives.

Re:who cares

[ScootyPuffJr]

I disagree here. Drag and drop can be incredibly intuitive if it's implemented in a sensible way.

I use drag and drop a lot when using Mac OS X simply because it works so well. Dragging files to the trash, dragging files to applications to open them, dragging images off web pages to save them etc.. it just comes naturally after a while (whereas digging through menus to find features like that has to be learnt every time). And with Exposé, you never need to rearrange windows.

Just because Windows (and therefore Linux, as sadly the linux desktops have heavily copied windows as opposed to OS X) can't do drag and drop effectively doesn't mean it's inviable. It just means that it's been made inviable through poor design.

Re:who cares

[Anonymous Coward]

> How about decent and WORKING drag and drop?

Good point

> everyone is busy with eye candy and other useless add-ons and ignore basic operability and useability.

But if you really had a clue about usability you should be able to spell the word.

Re:who cares

[genglish]

That's like a Windows user saying the command prompt sucks because all they're used to is DOS. Any great idea can be implemented poorly.

Re:who cares

[SCHecklerX]

I agree with your disagreement. In fact, I really don't know wtf gnome and kde have their own way when xdnd has been around for quite some time, and doesn't require any daemons running in the background to work. In addition, you have xds (think drag/drop in reverse...instead of dragging a file to a program to open, think of dragging from a file dialog within a program to a location to save/change icon/change theme/whatever). Rox uses this very well, and it is incredibly intuitive and useful. Again, another nice way of doing things that I wish was *THE* standard on X11 desktops. Oh well.

Re:who cares

[m50d]

KDE has awesome working drag and drop. Drag a link to a playlist into a media player and it starts playing. Drag an email into an address book and the sender gets added. Drag a file onto an IM contact and you send it to them. Etc, etc.

Re:who cares

[Anonymous Coward]

> GNUstep has drag and drop, and doesn't have much eye candy.

The tranny desktop? A DE pretender dressed in widget "drag", something unexpected "drops" when you touch the wrong button, and the eye candy really looks like something from the GTK1 Tammy Faye Baker School of Cosmetics. Good call there with the analogy to a tranny in drag. +5 Insightful.

Re:who cares

[advocate_one]

intuitive my arse... dragging disk icons to the trash to eject them??? I, and no doubt many others, would expect that to erase them

Re:who cares

[Watts Martin]

OS X does that to some degree, although I'm not sure it's as thorough about it as Rox is. (It's been years since I used Rox, so my memory is very fuzzy of it.) OS X does have the interesting idea of the "proxy icon" in the window bar; drag it to a Finder window to create a copy of the file (or an alias to it), drag it to a message window of just about any e-mail program and create an attachment, heck, drag it to another program icon to open it in that program, too. A lot of programs even let you drag and drop colors if you have a palette window open.

I think people who dismiss the usefuleness of drag and drop really haven't used an operating system (or window manager) that supports it effectively. Something you'll often notice is that Windows users tend to run nearly every application full-screen, while Mac users tend to run them all in, well, windows; I've always suspected this is because Windows has historically given you very little benefit from having multiple windows visible, whereas Mac OS has historically allowed a range of actions between windows.

Re:who cares

[SCHecklerX]

the examples you cite sound a lot like OS/2's WPS back in 1993 :) Too bad the two most popular environments on linux/X11 choose to chase windoze ui conventions rather than do things as powerfully as they *could*. Rox is a little guilty of this (Send To), but overall It's the best example of a file manager done right that I've seen in a very long time. What's really nice is that when you select something in rox, it goes into your cut buffer, so you can use it to make yourself more efficient even with programs that don't support xdnd...just highlight, then when opening the "do foo with file" dialog, just mid-click. This is a huge timesaver with using M$ Sharepoint here at work.

Re:who cares

[arthas]

File management (with GUI) is actually one of the most useful applications of drag and drop technique. Having two or more small file manager windows open and dragging and dropping files between them is very intuitive and comfortable. This, in my opinion at least, requires spatial file manager (like Nautilus, OS/2 Workplace Shell, Finder (in spatial mode), Gnustep GWorkpace.app (in spatial mode)) to be really usable. Gigantic MS Explorer windows are really not good enough for this type of task. You may want to say: "But that's what copy/cut-paste is for" but I have to say that copy-pasting files between file manager windows always feels uncomfortable, inelegant and most of all: inefficient (this is naturally just be my personal opinion, but still). I would also much rather just drag and drop text/pictures/whatever than copy/paste it.

I would like to see dnd techiques better supported in Linux desktop apps. It would be really nice if I could, for example, drag a picture from a web browser window into a mail client or word processor window... (didn't work with Epiphany and OpenOffice (gnome 2.8))

Re:who cares

[cloudmaster]

1) Linux desktops have almost all been around since before OS X, and could not have copied that interface.
2) OS X was partialy inspired by NeXTStep, which also inspired GnuSTEP (through the OpenStep standard). Which is *what the article's about* and which existed before OS X.

BTW, Expose just rearranges windows for you, so saying that it makes it unneccesary to rearrange windows is a bit of a stretch.

Just for the record, Drag-n-drop is only useful for simple actions - it becomes a nuiscance for more complicated tasks, or is simply non-intuitive in some situations. What if I drag a file and drop it on the icon of a text file - do I want to concatenate them, do I want to run the perl script in that text file, do I want to rename a file, do I want the guy down the hall to spontaneously combust? I drop a file in the middle of a word document. Do I want to insert the file there, do I want to open a new window, do I want to replace the current window, so I want the file name inserted at the drop point? What's intuitive to every user? What's sensible?

Re:who cares

[Uerige]

KDE has awesome working drag and drop. Drag a link to a playlist into a media player and it starts playing. Drag an email into an address book and the sender gets added. Drag a file onto an IM contact and you send it to them. Etc, etc.

Wow. That's really quite... what I expect?

Re:who cares

[Blakey Rat]

MacOS has had full support for drag&drop since version 7.0. MacOS 7 is roughly equilivant to Windows 95. (And that's more-or-less the same time Windows got drag&drop support, or at least Office did.) It's not just an OS X thing. It also has had clipping support that long... if you select some text and drag it to the desktop, it creates a "text clipping" file, which is pretty much exactly what it sounds like. Unfortunately, the OS X Finder handles clipping files in an intensely stupid way, but at least MacOS can still create them.

And I think the feature is very useful. The easiest way to save an image from a webpage (a task I do a lot) in MacOS is to grab that image and drag it to the desktop, or a folder, or where ever you want it. And creating clippings is handy as hell... I have something like 5 years of interesting dialog and quotes from a MUD I play all stored as clippings in a folder. If I see something interesting, I select the text and drag it in there... DONE! No copy, open SimpleText/TextEdit, paste, save, choose path, quit SimpleText/TextEdit.

Re:who cares

[cloudmaster]

Windows 3.1 had drag and drop in the GUI, though not all applications supported it. Mac OS did too, prior to 7.x Regardless, the post asked why Linux desktops didn't copy OS X, which is why I even mentioned it... :)

Anyway, I was just messing around on an XP workstation, and found that it's not letting me create text clipping files anymore. I know I could do that in '98, and will have to try it on a 2K workstation. If you copy some text, go to a folder (including the desktop) and right-click to select paste (or ctrl-v), a file gets created containing the clipboard contents. Similar to McBehavior.

As far as saving images, I personally don't like the Mac way of navigating folders while dragging - the delay between hovering and a folder popping open is just never right, and the time between when the folder icon blinks like it's gonna open and when it actually opens is always too short. I don't know how many times I've dropped into a freshly-opened folder which opened on top of the window I *wanted* to drop into because I hovered over a folder too long while I made the decision. OS X is better about that, esp. when using the view whose name eludes me now, but involves panes in a single window expanding out to the right as folder navigation depth increases, but it still irritates me often enough. Windows, of course, is even less useful - but it doesn't get your hopes up, either. :)

It's as quick or quicker, and more precise, to right-click on the image (ctrl-click), select "save-as", and navigate to the target in the resulting dialog. That also affords the opportunity to rename the file - and the ability to look at another program or something (where did that text file say to save this again?) before doing the actual save. Using "save as" also makes the behavior more clear in the case of an image which is a link to another web page - do you want to save the image, the link, or the linked page's contents?

Re:who cares

[Burning Plastic]

In OSX (as opposed to OS9) the trash turns into a nice eject symbol when dragging a disk to it...

Re:who cares

[Anonymous Coward]

Lucky the trash icon changes to an eject icon when you drag a CD, then, isn' it?

Re:who cares

[MightyYar]

Haven't used OSX yet, huh? There's a little "Eject" button next to any removable volume in the Finder. You can also drag to the trash if you are hard-core System 6. If you are feeling extra windozey, you can right-click (or Ctrl-Click for the right-button-disadvantaged) and choose "eject". For some reason, there is an "eject" button on the keyboard that spits out the optical disk. There is also the command line for the deranged: hdiutil eject -force /dev/diskxxxx.

Re:who cares

[Anonymous Writer]

OS X does have the interesting idea of the "proxy icon" in the window bar; drag it to a Finder window to create a copy of the file (or an alias to it), drag it to a message window of just about any e-mail program and create an attachment, heck, drag it to another program icon to open it in that program, too.

There's also a useful function wherein you can drag a folder to an open/save dialogue window and it automatically sets the path to that folder. I've found myself using it a bit. I presume that this is standard in other terminal applications though, but being able to drag a file or folder onto the Terminal and have it suddenly appear as a text path on the command line as if you typed it all in is really useful.

Re:who cares

[Anonymous Writer]

I should have mentioned this in my previous comment but a good example of the window proxy icon drag and drop function in conjunction with the Terminal is this; if you have a Finder window open and you would like to navigate to that directory in the terminal, you just type the "cd " command in the terminal, then drag the Finder window icon onto the Terminal window, and hit return. The Terminal then has the command line path set to that directory. It has a nice feel to it, combining Finder window navigation with the powerful Unix-level commands available through the Terminal.

Re:who cares

[ady1]

aah yes... here is an example of the usefulness of drag and drop in windows http://carcino.gen.nz/images/index.php/04980e0b/5e 7e783a

Thats one ugly interface

[Anonymous Coward]

It doesn't even look all that functional, lightweight yes, but it still looks ugly has hell, i mean fluxbox looks better. So how is this supposed to compete with gnome and KDE

Re:Thats one ugly interface

[stivi]

First: ugly appearance is not equal to non-funcionality. Flashy and colorful does not mean functional. If you want it to be pretty, use Camaelon as depicted here. The Camaelon bundle is in the progress of being included into the main GNUstep distribution. Themability was low priority to the project, compared to functionality.

Besides that, if you would like to know, then I have seen many expert applications that are "ugly" because of some obsolete GUI toolkits. And guess what? The experts use it and do not care about the appearance at all, but about the functionality.

And how it competes with KDE? By different application development approach, by different point of view on objects (C++ objects vs. Objective-C/Smalltalk-like), by another objective runtime, by different paradigms... If it had everything that KDE has, it would be KDE, not GNUstep. It is different so developers can have an alternative. Can you imagine a world where everyone would be the same?

Re:Thats one ugly interface

[Anonymous Coward]

Beauty is in the eye of the beholder ... I think it looks as good as or better than most/all other interfaces I've seen ... it's a very clean, simple, uncluttered look without the irritating 'eye candy' a lot of intrerfaces seem to insist upon.

Re:Thats one ugly interface

[JStrike]

Camaelon also looks really ugly. In fact it just looks like the other screenshots, but with colour. Not many people will be impressed with the early 1990's look, no matter what colours you dress it up in

Re:Thats one ugly interface

[shancock]

Well....This may be all good and true but for my workstation, my everyday interface on the computer I do want it to look
as good as it functions. In short I want both. Is this not possible? Can't we have a well funtioning desktop that is not ugly? Experts may accept an ugly interface to get functionality that is not availale elsewhere but given a choice between two gui's with the same or similar funtionality, I bet they choose the one that is not ugly. In this case there does not seem to be so very much difference between any of them (Gorm, Gnome, KDE, etc). It really comes down to personal, subjective choices. Would Firefox be so popular if it were ugly? I doubt it. Gorm is ugly and does not seem to offer anything I really need that is not available in Gnome or KDE (or am I missing something important here?).

Re:Thats one ugly interface

[cbv]

Camaelon also looks really ugly.

Of course it does, because Camaelon is "just" a theme engine. You're supposed to install a theme and actually use it, otherwise Camaelon will default to the original NeXT'ish look.

So, use a theme that pleases you, just as you (in general) do using any other desktop environment.

Re:Thats one ugly interface

[JStrike]

Theme engines dont make a desktop look better. They just look like variations of the same thing. Look at the Gtk theme engines. They dont make that much difference. It all looks pretty much the same at the end of the day. (That being said, Gnome looks pretty famn decent. The theme engines only make the slightest difference)

Re:Thats one ugly interface

[Clevo]

Would Firefox be so popular if it were ugly?

The default theme for Firefox is butt-ugly. I'm all for slick desktops (hey, I have several OSX boxes along with my many Linux boxes) but Firefox is a clear case of a product gaining popularity because it's functionality and security are superior to it's competition.

Re:Thats one ugly interface

[crivens]

When will Camaeleon be folded in and available as LiveCD? I'd be interested in trying it.

Re:Thats one ugly interface

[Art Tatum]

There's quite a bit of difference between GNUstep on the one hand and KDE/GNOME on the other. But it's not a difference for end-users so much as a difference for programmers. Building a more attractive user interface for GNUstep is not that difficult, however, since there's a theme engine in Beta. We just need usability experts and artists to help design good themes and desktop systems.

Re:Thats one ugly interface

[Art Tatum]

This is, quite simply, false. GNUstep bundles can change as much behavior and appearance as desired. It's not just "changing colors." You can completely redesign the whole behavior and appearance of widgets if you like. For instance, there's a bundle called WildMenus that ditches the NeXT vertical menu and replaces it with a horizontal Mac menu at the top. Just an example.

The Real Question is...

[Markus_UW]

Does it run on Linux?

Re:The Real Question is...

[Lodragandraoidh]

Afterstep runs under Linux -- a cousin of OpenStep...

Re:The Real Question is...

[Anonymous Coward]

Unfortunately Windows 3.1 already has RAD builders on this level, such as Visual Basic 3.0 and Borland Delphi. Maybe someday Linux will catch up.

Re:The Real Question is...

[tarzeau]

No, Afterstep is only a window manager, that is only working on X11. See here, what the OpenStep specification is about. While you do so, you might also read up the Savoy 1992, Booz-Allen Study results

Graphical Object Relationship Modeller

[stivi]

Gorm is also a RAD application that allows one to create user interfaces and various application object models in very intuitive way, benefiting from highly dynamic features of the Objective-C language and runtime. Flash videos can be seen here. More information can be found on this blog. Interesting is, that the application could never be done in C++, check out why.

Re:Graphical Object Relationship Modeller

[tgv]

I say: nonsense. Everything that can be programmed can be programmed in C++. Or C. Heck, even in Perl.

I know Objective-C (I do my modelling in Cocoa), and I know how the dynamic bit works, but to say that it cannot be done in C++ is BS in principle. It cannot be done in the same way, but it surely can be done.

Re:Graphical Object Relationship Modeller

[Anonymous Coward]

Oh, I remember the ObjC versus C++ language war. There was an unconditional surrendor in like 1993.

That blog is sorta like those Japanese soliders that held out on remote islands for 20 years because nobody told them the war was done.

The real question is, does this bring anything to the table that isn't available in C# environments?

Re:Graphical Object Relationship Modeller

[Anonymous Coward]

Interesting is, that the application could never be done in C++, check out why [blogspot.com].

That's crap, and if I had a blogger account I'd tell him where to go.

4) Dynamic binding: so you make a baseclass with all the common interfaces you'll need. If that's not good enough, C++ does support dynamic downcasting with run-time typing information.
1) Categories: I'm not 100% sure what he means but these sound like C#'s attributes. So build this kind of mechanism into your baseclass.
3) Protocols - WTF? If that means 'interface', well, yeah C++ does those well.

Re:Graphical Object Relationship Modeller

[Anonymous Coward]

Interesting is, that the application could never be done in C++, check out why.

Of course it can be done in C++. Maybe less quickly, maybe with a different internal design, but all programming languages are equivalent at a fundamental level and any program that is possible to write in one language is possible in another.

Re:Graphical Object Relationship Modeller

[b100dian]

I think that anything can be done in asm!

Heck, actually everything is done in asm!

Re:Graphical Object Relationship Modeller

[Anonymous Coward]

Interesting is, that the application could never be done in C++, check out why [blogspot.com].

Arguments #1, #3, and #4 are all solved by COM and Corba, which you can use with C++. It's not built into the language, but it's not supposed to be with C++.

"could never be done in C++" ... What a bunch of crap. You might as well say it will make KDE and GNOME obsolete... oh wait...

Re:Graphical Object Relationship Modeller

[Anonymous Coward]

Exactly.

The only question you need to ask:

  Is it Turing complete?"

Re:Graphical Object Relationship Modeller

[sznupi]

And soon (via converter) it will be compatible with .nib files, meaning very easy poerting of Cocoa apps (or you could save to XML in Interface Builder...)

Re:Graphical Object Relationship Modeller

[hunterx11]

Saying that it couldn't be done in C++ sounds like flamebait. A much better (and true) claim would be that Objective-C is the right tool for the job, and C++ absolutely the wrong tool for the job in this instance.

Re:Graphical Object Relationship Modeller

[Anonymous Coward]

Good grief, Are you guys literalistic or what?

Plainly 'could never be done in' means could not be done practically (ie would take too long/require too much effort/manpower).

Of course anything that can be done in one computer language can be done in almost any other, but practically everyone knows that, so it's not an interesting point, and it's therefore rather dumb to interpret the statement literally.

Re:Graphical Object Relationship Modeller

[tgv]

When you've got the Core Data bit done, call me.

Re:Graphical Object Relationship Modeller

[Lodragandraoidh]

I think the point he is trying to make (after RTFL - the relevant link) is that while you can impliment equivalent functionality in any language, you will take a performance hit because other languages are not optimized to handle some of the items he mentions (class indexing etc...)

That is a tenuous argument at best; nonetheless without knowing the OpenStep standard there might be some validity to it - particularly if the standard specifies such functionality as is native to Objective-C.

Having never used Objective-C I am probably completely wrong about this - just making an observation based on what I read.

Re:Graphical Object Relationship Modeller

['nother poster]

How do you figure? Perl doesn't have an intermediate step that has assembly code, and neither does Ruby. Compiled languages go from their language to machine code, once again, no assembly. Now, with a disassembler you can turn machine code into asm, but it's usually pretty ugly.

Re:Graphical Object Relationship Modeller

[dnaumov]

I say: nonsense. Everything that can be programmed can be programmed in C++. Or C. Heck, even in Perl.

So, can you point to some operating system kernel written in Perl?

Re:Graphical Object Relationship Modeller

[Anonymous Coward]

but now those Japs have combined with the COBOL veterans of the Civil War and the ADA holdouts from the Spanish American War and are going to kick C++'s butt!

Re:Graphical Object Relationship Modeller

[SassyDave]

After watching these videos, I must disagree that Gorm is going to cause "the obsolesence of the GNOME and KDE projects". Gorm is about where Qt Designer was 3 years ago.

Re:Graphical Object Relationship Modeller

[Anonymous Coward]

Of course it can be done in c++. There are all sorts of theoretical papers on the minnimal number of constructs to program any solution and trust me it doesn't require the dynamic bit or anything else specific to object c. The only barriers to various languages come when specific hardware is inaccessible to the language but there are usually ways to extend the language (for the programmer that doesn't require access to the parser/complier code) parser/complier through multi-language programing and independently compiled modules there by making the hardware available from there on. That's not to say that this isn't a better way to do it or that it can't be done in the same manner but the idea that one language can do things others can't is pretty silly give all the mathematical proofs and cs papers written in this area. The damn marketers/fanatics should do some research.

Re:Graphical Object Relationship Modeller

[unborn]

It'd be rather silly for a compiler to produce binary code directly. All compilers I know produce assembly code and then assemble it with an assembler into binary code.

If it's bytecode or interpreted languages that you speak of, then the VM or the interpreter are written in binary (assembled in some way) and without them the programs they interpret wouldn't run.

Re:Graphical Object Relationship Modeller

[Taladar]

You do know that "assembler" and "machine code" are one and the same, just displayed in a different way? Saying something uses machine code and not assembler is like saying some printed text is not "written", just "typed".

Re:Graphical Object Relationship Modeller

[hackstraw]

Thanks for the flash video link. I thought I would never have said that, but I did.

Reminds me of the days when I used WindowMaker. WindowMaker was also a GNUStep project, and a damn good window manager at the time.

I welcome a more "UNIXy" desktop environment vs KDE or Gnome. To me, those are way too much like the Microsoft Windows interface. Personally, I don't like any of them. Honestly, all GUIs nowadays seem really dated when compared to OS X. Yeah, yeah, call me a fanboy or zealot, but for the look and feel, OS X is pretty slick.

Now that I mentioned OS X, what does it take to come up with a good interface like that? Its pretty much completely different than other GUIs I have used. Things like the way each application has all of their windows grouped together. An application can only be launched once, and doing something like launching an application that is already running, but has no windows in it will open up a new window. Working drag and drop. A working clipboard. A common look and feel between applicatons. The list goes on.

GUI land on UNIX has always been a little lacking. I've been writing and talking about this for many years now. There are many issues, but here are some that I can think of:

Fragmented GUI widget sets. This is big. You have KDE and Gnome. Motif. Xlib. Xaw. Xaw3d, XawWindows95wannabe. You name it. This has to stop.

The whole X server thing. Its a great idea, and has lasted for a long time, but I think its done. The only saving grace I think it has is the ability to remotely run an application and display it and interface it locally. The downside is that local applications are pretty much like remote applications. The graphics are not smooth and jerky at times. They flicker, and just are not as good as graphics on MS Windows or OS X or even a console game. I think X should die. Keep around a compatible way to run X apps like you have for Windows or OS X (hopefully, much better than the OS X way), but there needs to be a new and better integrated and featurefull GUI environment. Kinda like the KDE or Gnome "desktop environment", but a little more low level, and better. It should only have X ability for backwards compatibility, but ditch it from there.

Gotta work...

Re:Graphical Object Relationship Modeller

[j-tull]

Just because everything can be programmed in *insert language here*, that doesn't mean that it has or even should be. The fact that an OS (or any other specific program for that matter) hasn't been written in a certain language says nothing to if it can or can not to done.

Yes, it would have been easier for me to simply mod you down (since I do have the points to do so today), but I thought a little exposition and explanation would provide more value to the /. community in the long run.

Re:Graphical Object Relationship Modeller

[Kupek]

Perl doesn't have an intermediate step that has assembly code, and neither does Ruby.

Sure it does. It's just that there's another layer (the perl program, in Perl's case) inbetween your Perl source and the resulting code. While your Perl program may never exist as a machine executable, there certainly is a stream of instructions associated with your Perl program. That was the parent's point.

Compiled languages go from their language to machine code, once again, no assembly.

Not true. When you invoke gcc, for example, there four distinct phases: preprocessing, compilation, assembly and linking. The output of the compilation process is assembly code for the architecture gcc was compiled for. It then invokes as, the GNU assembler, to convert the assembly code to machine code. Passing gcc the option -S tells it to not call the assembler, and the output is assembly code. Most compilers are going to have these distinct phases for reasons of portability, code maintenance, and ease of applying optimizations (some optimizations are applicable at the source code level, others at the assembly level).

But there really is a one-to-one mapping of assembly to machine code, so I don't consider the difference to be significant. Assembly is really just human-readable machine code. Even then, most processors don't actually execute the machine code as-is. Pentiums break down x86 instructions into micro-operations which are what really get executed on the processor.

Re:Graphical Object Relationship Modeller

[Anonymous Coward]

Actually, QT Designer is a much more primitive design ... it basically just allows you to place widgets and generate the code to draw them.
Gorm is much more sophisticated than that ... it creates objects on the fly and saves them to be loaded in to an application (not just generating code) ... and allows those objects to be linked together so that messages from one object trigger an action in another (QT Designer has a crude version of this in 'signals and slots' but Gorm provides the power of dynamic Objective-C messaging here).
Most importantly, the objects linked together need not be gui objects ... you can connect backend objects like database, timers etc together. Obviously the video is going to show the simple gui stuff (easiest to demo) ... but don't assume that Gorm is as limited as QT-Designer on that basis.

Re:Graphical Object Relationship Modeller

[Anonymous Coward]

http://gscoredata.nongnu.org/

Re:Graphical Object Relationship Modeller

[Anonymous Coward]

Turing completeness is bullshit. Yes, you can perform any calculation in any Turing complete language. That's quite valuable. But that doesn't mean all Turing complete languages are equally functional.

The problem is IO. If your program never does IO, it's a space heater. This covers both what sorts of IO your code needs to do (text, graphics, sound, time, swapping virtual memory), as well as when that IO needs to be done. You can encode MPEG2 video in JavaScript, but can you grab frames from a webcam and burn the result to DVD in real time?

Re:Graphical Object Relationship Modeller

[ajs]

Well, someone was working on it in 2001, but thankfully he never finished. As others have pointed out, it's a horrible idea. Of course, Perl's current compiler/interpreter/runtime would need massive work—possibly a full re-implementation—for such a project, but none of that has anything to do with the language.

Let's be clear, here. When someone says, "you can't write foo in language x," they are full of shit, and they can take the matter up with Turing.

If they say, "language x is more suited to problems of the same class as foo than many/most/all other choices," then they might have a point, depending on the specifics.

The first claim is what was made by the original poster, who is... in case you lost me on this point... full of shit.

Re:Graphical Object Relationship Modeller

[Anonymous Coward]

the application could never be done in C++, check out why.

To begin with, Objective-C (like C++) was originally implemented as a translator to plain C (+ a runtime library.) It is absolutely not true that Objective-C can do anything, ever, that C cannot do. (I only bother to point this out for the benefit of those who don't comprehend "Turing complete.")

It is, of course, easier to do certain things in one language than another: there is nothing that can be done in Lisp that can't be done in C either (for the same reason), however the C program is likely to be huge, take forever to write, and be hard to understand and maintain. Objective-C does have advantages for certain types of programs (like UI's), but not on the scale of Lisp/Haskell/your favorite super-language.

As far as the usual reasons given for OC's "superiority":

• Categories: I'd list this first too. There is no simple equivalent that is remotely as flexible or easy to use. Still, pluggable interface objects can do much the same thing, without the "hidden superclass" problem.
• Loadable bundles: generally do nothing that ActiveX controls can't. Granted, IB palettes are very easy to create and use, while COM is among the worst designed and implemented ideas in the history of computer science. Still, it works, and there are other, better alternatives out there too.
• Protocols: same thing as interfaces. Objective-C adds some handy but non-vital introspection capability.
• dynamic binding: essentially the same thing as messaging in most modern GUI frameworks. OC hides most of the support code in its runtime library and adds some sugar to make it easier to use. On the down side, messaging is much slower than virtual methods, even w/ Apple/*Step's caching scheme, which is one reason why it's sparingly used in C++ programs.

"Faking" Objective-C features in C++ also emphasizes that many of them aren't necessary all the time. OC jams everything in to every object, while a C++ program can add features where they're needed. Every NSObject, for example, carries the weight of OC's memory management (improved ref counting) scheme; my current C++ version not only has much lower overhead, but can be selectively added to the (few) places it's actually needed. Likewise messaging: most frameworks allow the plumbing for message sending and receiving to be added specifically to objects that needed it, not loaded onto every single object.

Re:Graphical Object Relationship Modeller

[The Bubble]

Everything that can be programmed can be programmed in C++. Or C. Heck, even in Perl.

Everything that can be programmed can be written in assembler.

:wq

Re:Graphical Object Relationship Modeller

[Hatta]

That's great for developers, but what appeal does it have for users? Is the fact that my menu bar, desktop icons, etc are objects going to help me manage a large number of xterms efficiantly? If so, how?

I've tried pretty much every WM and DE I could find, and eventually settled on fluxbox with XFCE* a close second. What features does GORM have that would make me want to use it? How is work flow in GORM different than workflow in other environments?

*XFCE walks the middle path between functional WM and featureful DE very well. Users with a preference for either extreme should check it out. Just use your own file manager, XFFM sucks.

Re:Graphical Object Relationship Modeller

['nother poster]

Well, maybe things have changed in the last 30 years, but assembly was a low level human readable language. You typed it into a file, or punched it on cards, and it was then used as input to a program called an assembler which turned it into machine code. The machine code file was not the same as the assembly code file.

Re:Graphical Object Relationship Modeller

[arkanes]

Let's be clear, here. When someone says, "you can't write foo in language x," they are full of shit, and they can take the matter up with Turing.

In many cases people use that phrase as shorthand for "you can't write foo in language X without re-writing it as language Y first", which is more reasonable and correct. You can't have Python style duck-typing in C++ without first implementing a new dynamic type system and then running your code in a VM-like environment, executing some other language that isn't C++. And then you really aren't talking about writing your application in C++ anymore. People call this Greenspunning, after his famous claim that any sufficently advanced program contained half of Lisp, but I think that Lispers fall into this trap as much (or more) than anyone - Lisp simply provides an environment where implementing your own language is simpler and considered appropriate.

Re:Graphical Object Relationship Modeller

[Coryoth]

The whole X server thing. Its a great idea, and has lasted for a long time, but I think its done. The only saving grace I think it has is the ability to remotely run an application and display it and interface it locally. The downside is that local applications are pretty much like remote applications. The graphics are not smooth and jerky at times. They flicker, and just are not as good as graphics on MS Windows or OS X or even a console game. I think X should die.

You are aware that the ability of X11 todisplay remote applications transparently has no significant effect on local applications right? Local applications effectively uses shared memory, the same way OS X and Windows works. Will people please get through their thick heads that the remote capabilitis of X11 have nothing to do with their percieved performance issues of X11. Getting rid of X11's network transparency will not suddenly stop the "flickering". If you want to do that, try messing with Keith Packards kdrive and similar - yes they are still X11, but a different implementation of the X11 protocol that's smoother and has a lot of the facnier effects you might expect.

Jedidiah.

Re:Graphical Object Relationship Modeller

[hackstraw]

Local applications effectively uses shared memory, the same way OS X and Windows works.

As I said, I'm not a GUI guy. I don't know the underlying whatever that makes Windows, OS X, or X11 do its thing. However, I do know that to run an X app, you need an "X server" to display it on. From what I do know, the X server is the guy that is responsible for getting the hardware information like keyboard and mouse input and relaying it to the application. It is also responsible for displaying the information onto the display.

Being that all X apps run the same, right out of the box either locally or on another machine, I was assuming that the "glue" between the app and the server as to why X apps look bad. If that is not the case, then my argument for ditching X is even more strong. Its just not good anymore.

I've got a relatively high end 4 Itanium CPU box with dual graphics cards that are pretty high end. these guys. They retail for about $800 a piece.

Even on that box, X apps, even OpenGL ones that you can run on any box, don't look as good.

If a $40,000 computer can't make X look good, then lets bury the old dog out back with the others.

Re:Graphical Object Relationship Modeller

[Synic]

Blame ATI. Their drivers are complete shit for X11.

Re:Graphical Object Relationship Modeller

[jemfinch]

Turing completeness is a useless and irrelevant measure of the power of a language. A better measure is was put forward by Paul Graham: One language is "more powerful" than another if, in order to implement the features and capabilities of the latter language, one must write an interpreter in the former language.

That is to say, C simply *can't* do the same kind of runtime object introspection that Python can. In order to achieve that feature using C, a programmer would have to write an interpreter for Python in C. Therefore, Python is more powerful (in this useful way, not in the turing completeness way) than C.

Jeremy

Re:Graphical Object Relationship Modeller

[Coryoth]

I can't honestly tell you what you're doing wrong, but X11 works fine for me with no flicker - but then I have nVidia and I hear their Linux drivers are far superior to the ATI ones. The biggest issues I ever had with X11 was with XFree86 and their inability to actually do anything, and the resulting stangantion. Now that Xorg has effectiovely taken over as the premiere OSS X11 implementation things are starting to move again and we can expect to see plenty of improvements down the line.

In the end X11 is a protocol, not an application - there are lots of different implementations. X11 can definitely be a stunning graphical environemnt: IRIX used to use X and you can hardly say SGI was all about shitty quality graphics. You seem to be suffering under a poor driver for your Xserver. Wanting to scrap X11 because of this is like wanting to scrap HTML because there's a page that takes a while to load, or wanting to scrap KDE because someone wroite a crappy application for it.

Almost all the hatred of X11 is largely misdirected. Scrapping X11 is not going to magically make everythign better. Using a better X11 implementation (again, check out Keith Packard's work, which is slowly making it's way into Xorg) is the answer for the problems you're raising, and we don't lose the network transparency which, to be frank, is a truly wonderful thing.

I suggest you actually take some time to learn about what X11 is and how it works before you call to scrap it. There are some legitimate complaints that can be made about X11, but I don't think they're sufficient to start from scratch, and they certainly aren't anything like the ill-informed arguments you are touting.

Jedidiah.

Re:Graphical Object Relationship Modeller

[Arandir]

Interesting is, that the application could never be done in C++, check out why.

There's no bigot like a language bigot. Qt/C++ does everything GORM/ObjC does. But since it isn't GORM/ObjC, it of course doesn't count.

Re:Graphical Object Relationship Modeller

[Arandir]

WindowMaker was also a GNUStep project

Not really. While it is the "official" window manager of the GNUstep project, it itself does NOT use GNUstep, Objective C, or GORM. It is a standard X11 window manager written in plain old-fashioned C. Once GNUstep gets their own Object-C/GNUstep window manager, they'll drop WindowMaker as fast as GNOME dropped Enlightenment.

Re:Graphical Object Relationship Modeller

[GileadGreene]

Funny. I regularly use Apple's X11 server - which adds another layer on top of the standard Mac OS graphics subsystem FFS - to run X11 apps on my Mac, and I don't experience any of the problems that you are alluding to. My only gripe is that the integration between X11 and Aqua isn't quite as nice as I would like, but that's a UI issue rather than a performance one (and it's pretty much a function of the different UI models, so I'm not sure how it could be fixed). As others have already pointed out, perhaps your problem isn't X11 at all...

Re:Graphical Object Relationship Modeller

[listen]

I have to say, nothing you said there can't be done with Qt Designer.
Also, QT Designer doesn't generate code. It generates a .ui XML file which can be loaded at runtime.
Sorry.

Re:Graphical Object Relationship Modeller

[Arandir]

Precisely. If you wanted an pervasively object oriented desktop for *users*, OS/2 Warp was it. But just like NeXT, the customer wanted something else.

Re:Graphical Object Relationship Modeller

[TheRaven64]

A very good definition. For example, last month I wrote some code in Objective-C that enumerated all of the classes that implemented a particular interface or were subclasses of a given class, and presented the user with a choice of the one to use. This was about a dozen lines of code.

Doing this in C++ would be possible. Doing this using the C++ class model, would not. I would have to write a set of introspection methods into C++. The most important thing, however, is that my Objective-C code worked without my having to modify anything else in my class hierarchy. This ensures that when I use this feature of the language, I am compatible with other people's code (important, in my case, since it was a system which was designed to be extensible, and I didn't want people adding new classes to have to register them with the UI or anything to make their lives harder).

Systems like Qt and COM add some Objective-C-like features to C++, but these features are not accessible to code is based on a different framework.

Re:Graphical Object Relationship Modeller

[jallen02]

Technically..... sure. I could very easily use a JavaScript interpreter written in C and provide the JavaScript with all of the hooks it needs to do its job. I do believe that the Mozilla/Firefox people have a nice JS langauge implementation.

It is not that language, but the implementation that you are saying is not equally functional. Functional in the sense they can get at the underlying operating system and libraries easily. Now.. the power and expression capabilities of two languages, you have a point there. It all depends on your problem domain.

Jeremy

Re:Graphical Object Relationship Modeller

[deaddrunk]

Assembly is really just human-readable machine code.

For a certain value of $HUMAN

Re:Graphical Object Relationship Modeller

[colinrichardday]

Hmm. . . Maybe, but I'll pass on trying to write the Linux kernel in Fortran.

Re:Graphical Object Relationship Modeller

[Anonymous Coward]

Categories are simply method additions to classes that can also be
loaded as bundle. It may sound strange to you to do things like this,
but it makes sense if - for example - you want to extend a class
in the standard class library (which you usually don't want - and often
aren't even able - to modify.)

Yes, every programming language is Turing-complete, but not
every language is as dynamic as ObjectiveC, and allows introspection
that goes this far.

Btw, enabling dynamic binding in C++ means that you have to
mark every method 'virtual'. Statically bound method calls
may provide a performance improvement in high-performance
scenarios, but after all developing with a completely dynamic-bound
language is much easier and makes classes extendable even if they
aren't thought to be so. (One may of course argue that programmers
shouldn't extend classes that aren't thought to be extended, but
looking at the programmers I know (me included), I can ensure you
that most classes won't be designed to be extendable - just because
you often can't think of a scenario where this makes sense for you.)

(Sorry for the long answer. I didn't want you all to have to read so much. :-))

-Guenther

Re:Graphical Object Relationship Modeller

[hackstraw]

IRIX used to use X and you can hardly say SGI was all about shitty quality graphics.

FWIW, the box is a SGI Prism. It runs Linux though.

Re:Graphical Object Relationship Modeller

[zsau]

I welcome a more "UNIXy" desktop environment vs KDE or Gnome.

What's that meant to mean? You want a DE that is built from a lot of small parts that each do their own thing, instead of a single monolithic application? Depending on how you interpret "small parts", either KDE xor Gnome probably fits the bill quite nicely. ROX is even leaner (with a Gnomish interpretation thereof), but then you tend to use Gnome (or generic) apps to fill in the gaps. GNUStep I think tends towards the Gnomish interpretation too.

Perhaps you mean you want to treat everything as a file rather than masking it behind a menu that manages all your applications. If you like the MacOS X interpretation of that where you start applications by clicking on an icon that actually literally contains the app, then you'll like ROX, which does basically the same thing (insert above disclaimer here).

Note that ROX isn't remotely application-based. You could make a hierarchy something along the lines of ROX-Gnome-Windows-MacOS X, where on the left-most edge, you tend to focus on files and windows as the focus of your interface, whereas on the other end you have applications. This isn't a case of one being better than the other, but just different paradigms that work for different people. Personally, I'm typing this on a Ubuntu/PPC box because I couldn't get myself to like the app-based interface. If, however, you consider this important, note that GNUStep is based on the same base as MacOS X and inherits the application-based interface. In combination with Window Maker, you might find you quite like it.

On my computer, drag and drop and the clipboard both work; they also do on all other computers I've used that run a free desktop. I use drag and drop and X's two clipboards on a regular basis, depending on what exactly I'm doing. I always seem to be able to move & copy stuff around.

Mac OS X doesn't have a common look, with the whole brushed metal vs aqua thing. Mac OS X also doesn't have a common feel, with Carbon- and Cocoa-based applications doing different things at times. Also, try clicking on an unfocused window some time. Different programs do different things when you do that; even the same program does different things when you do it to different *parts* of a window!

With free desktops, the different toolkits and looks-and-feels are in competition with each other (of sorts) so that it's usually quite easy to avoid programs that don't use your preferred toolkit. On Mac OS X, though, it's just one whole, so you have to cope with the different raise/focus/do on click behaviors.

I'll never for the life of me understand why the toolkit "fragmentation" "has to stop". So many people have said that before. I think that's the only reason they keep saying it.

Re:Graphical Object Relationship Modeller

[tgv]

That seems put wrongly. I surely can write object introspection in C without resorting to an interpreter. I would, just like Python, add some kind of identifier to each object, during allocation or whatever. Just like the dynamic rtti and virtual functions in C++.

What he probably *should* have said is that a language is more powerful with respect to a certain task if it gives you features to avoid constructs necessary in the other language.

The need for an interpreter is nonsense since it can always be avoided.

Re:Graphical Object Relationship Modeller

[tgv]

Ok, it's absolutely not working yet, but it's a lot more than I could have imagined!

Re:Graphical Object Relationship Modeller

[jemfinch]

The need for an interpreter is nonsense since it can always be avoided.

All you're doing in your proposed situation is partially evaluating the interpreter by hand. I.e., instead of implementing a generic scheme for handling all objects in any program in the way you propose, you only implement the scheme for the objects your program actually instantiates.

Your argument is also trivially defeated by the use of any libraries whatsoever. Given a random void pointer from a random library function, I can't do the same kind of runtime introspection in C that I can do in Python.

Jeremy

Re:Graphical Object Relationship Modeller

[tgv]

"Evaluating the interpreter by hand" is not interpreting, which is what you demanded. And no, the scheme would apply to all objects.

The run-time library doesn't contain language specific objects, but even then name lookup is possible since most compilers support debugging information.

And that's not the point. The point is that in Python the language gives it to you for free, whereas in C you need to add some (or quite a lot) of code to obtain the same, without interpretation.

And btw, the reverse also holds. I'm pretty sure you can't write bitfield structure members in Python, nor can you do the remarkable pointer arithmetic C is capable of. Ergo, C is a more powerful language than Python.

Hmm, now let me think: C > Python && Python > C. There something wrong here... Ah, I see, it's the operator. It hasn't been defined really properly, has it? Consequently, saying one language is more powerful than the other is really asking for trouble.

whuh????

[bazmail]

"...and with its release, comes the obsolesence of the GNOME and KDE projects"

Are these guys on crack?

Re:whuh????

[sznupi]

"These"? Just some "qa'lth" said it...

GuuuuuuuuhhhHHH!!!

[Anonymous Coward]

I GOOOOOOOORRRRMMM!!!!!! GORM SMASH!!!!! KDE EAT!!!! SMASH GNOME!!!! GUUUUUhhhHHH!!!! GrrrRRRRR!!!! I GOOOOOOOORRRRRMMMMM!!!!




alternatively:

Gorm nuts!
Corn nuts!!
Gorm nuts!
Coooorn nuts!!
Gooorm nuts!!!
Cooooooooorn!!!!!!
Gooooooooooorrm!!!!!

Re:GuuuuuuuuhhhHHH!!!

[Anonymous Coward]

Who aaaaarrreee these people?
Hoo aaaaarrrrrr deese beebles?

Re:GuuuuuuuuhhhHHH!!!

[slavemowgli]

Why am I reminded of this immortal dialogue...

Lahndon!
Essex!
Lahndon!
Essex!

Re:GuuuuuuuuhhhHHH!!!

[rubens]

Slashdot, the only place where you get modded "Score: 5, Funny" with a post like that...

Re:GuuuuuuuuhhhHHH!!!

[Anonymous Coward]

*sigh*

does anyone else miss OOG THE OPEN SOURCE CAVEMAN? i know i do!

Re:GuuuuuuuuhhhHHH!!!

[Castar]

Gooooorrrm. Nice woody word that. Gooorm. Not at all tinny. Gooorrrm.

Re:GuuuuuuuuhhhHHH!!!

[Anonymous Coward]

I assume this is a reference to something?

Knock em dead GORM!

[gihan_ripper]

Maybe GORM will make GNOME and KDE obsolete, but first their server will have to withstand a righteous Slashdotting!

Re:Knock em dead GORM!

[GweeDo]

Looks like KDE and Gnome have some time then.

Re:Knock em dead GORM!

[tux_fairy]

Why is the ./ effect misunderstood by so many? It's almost never the server, but the low internet pipe. This is so obvious on slow loading static objects, like images for instance.

Half a marketing jobbie

[Elfod]

Having set the expectation with references to OSX, why don't they have any drop dead gorgeous screenshots instead of one very dated looking one? I had high hopes that got dashed with WTF?

Re:Half a marketing jobbie

[Art Tatum]

Sadly, there is a strong contingent within the community that likes the old NeXT look and doesn't want anything else to be seen. But there is a theme engine and a couple of themes that look pretty good.

-1 Flamebait

[ticklejw]

So why can't we moderate articles too?

Re:-1 Flamebait

[Anonymous Coward]

Why, because /. is owned.

Many have been asking them for that for years.

Re:-1 Flamebait

[10101001 10101001]

Better yet, why isn't there relative modding, where people who mod similarly to you auto-mods posts (ie, just like the suggested mechanism to overcome poisoning a p2p network; trolls are out to poison this blog, right?). And yes, I could use the source. So, clearly I'm just as insightful as the parent to be bitching about why things aren't done already the way I want.

Maybe... No.

[Ageless]

I'll be the first to admit that Interface Builder (in OS X at least) is an incredible, life changing piece of software. If you've never written a GUI using Cocoa with Interface Builder you can't even imagine how easy it can be.

And Gorm is supposed to be Interface Builder for GNUStep.

That said, it's not GNOME or KDE. You've still got to write that whole boring desktop thing. Gorm might make it a lot easier to write all the stuff that's still missing but saying it made GNOME and KDE obsolete is just plain bullshit.

Re:Maybe... No.

[bullitB]

You've still got to write that whole boring desktop thing.

That's what GNUStep is for. The whole point here is that now the GNUStep project has a complete, released desktop development environment.

That said...this is essentially the equivalent of announcing GNU/Hurd, 1.0, thus making the Linux kernel obsolete. :)

Re:Maybe... No.

[TheRaven64]

That said...this is essentially the equivalent of announcing GNU/Hurd, 1.0, thus making the Linux kernel obsolete. :)

The Linux kernel was obsolete before it was started. Didn't you read Tanenbaum's post?

Over the top

[BenjyD]

Before hyping up your toolkit and predicting the death of all other OSS desktops, it's generally best to make sure your toolkit doesn't look like crap in all the screenshots.

This is a joke, right?

[schestowitz]

But it doesn't have the Monty Python foot. Will I ever drop KDE for that ugly thing?

Gorm!

[piyamaradus]

It sounds very 'tinny'...

Re:Gorm!

[Anonymous Coward]

No, definitely 'woody.' Gooooorm. Goooooooorm.

MacOSX?

[Anonymous Coward]

I assume they mean Aqua. In any case, I can't say I much like Aqua. It is just too inflexible. It is Apple's engineers saying they know better than I how I best I work.

Put the menu back in the window, bring back focus-follows-mouse, and allow typing into background windows without them popping to the front....pleeeeeaaaasse. ...or at least allow it be configured that way, as can be done on Microsoft Windows.

With that stupid menu at the top of the screen, the Mac is destined to stay a one or two monitor system.

Re:MacOSX?

[Anonymous Coward]

[BLOCKQUOTE]Put the menu back in the window, bring back focus-follows-mouse, and allow typing into background windows without them popping to the front....pleeeeeaaaasse. ...or at least allow it be configured that way, as can be done on Microsoft Windows.[/BLOCKQUOTE] Umm...put that stuff back? Last I checked, the things you mentioned have been that way on the Mac since its original release in 1984, and considering that it was the first (available to the public) GUI, it's not a matter of putting the menu back into the window or bringing back FFM. Those would be major departures from the way the Mac has always worked. There are some of us that actually like the Mac's UI specifically because it's so very flexible with regards to different ways to accomplish the same task. I find FFM annoying for the most part; the menu bar at the top of the screen adheres to Fitt's law much better than top-of-window menubars, and to be honest I've never minded having to click to activate a an application (as is necessary in OS X and Windows).

Re:MacOSX?

[xouumalperxe]

With that stupid menu at the top of the screen, the Mac is destined to stay a one or two monitor system.

Aaawwwwwwww! JUST two screens. Seriously, how large a percentage of users do you expect to use more that one (let alone more than two!) displays? The OSX interface actually follows some pretty solid rules for interface design. It was microsoft that nabbed defeat from the clutches of victory more than once with their designs (Though, to be fair, they've been learning).

Here are a few exercises for you to try out for yourself.
1. Fire up windows.
2. Open as many instances of your favorite program as your machine will allow. Maximize some, shuffle the rest about
3. Now close them. That's right, close them. Use a variety of methods. Namely, file > exit, X button on the corner, move/resize/close menu, application miniature context menu.
4. While doing the former, note a few things: The ease with which you reach the X button when the window is maximized (corner), the time it takes to reach the miniature (edge), the time it takes to reach the move/resize/close menu when the window is maximized (another corner), the time it takes to actually aim for the file menu (not on an edge), the time it takes to reach X and the move/resize/close menu when the window is NOT maximized (NOT on a corner or edge).

You'll probably note that with a teensy little bit of practice, the latter is actually much slower than all the others. However, you should also note how windows doesn't supply visual clues indicating that the corners work. The program borders/buttons don't reach the borders.

Now, MacOS has always used the top border for the menus, and with a little bit of honest effort on your part, you'll have to reach the conclusion that the edges of the screen are a VERY FAST place to reach,while things that are physically closer are not necessarily (or even usually) faster.

In OSX, with exposé, the corners of the screen have potential use as really good multi-tasking accelerators. Even though I don't own a Mac (yet), the scant minutes of use I've accumulated from some of my friends' Macs indicate that you get used to the "throw to the corner" motion of exposé remarkably quickly. KDE also places the menus on the top of the screen if you want, though unfortunately that only works with Qt applications (so no firefox, amongst other).

Gnome, while not putting the menus on the top edge, allows for large amounts of costumization regarding the panels. I have a set of launchers on one edge, the taskbar on another (as usual), the start menu equivalent on the top right (encompassing the corner), show desktop on the bottom right corner, trash bin on the bottom left, and a list of currently open applications on the top right. A desktop pager/switcher is placed on the bottom, next to show desktop. Almost all the edge space is used with useful things (except for the left edge, that I find a bit more cumbersome to reach).

Windows, however, does all it can to prevent you from using the easiest places to reach on your screen. Why is that?

Lighten up

[Ur@eus]

They GNUStep guy announcing this was just trying to have some fun, why the hell to people get some riled up by the obsoleting GNOME and KDE statement, have people completely lost their sense of humour? Congratulations to the GNUStep team on their Gorm 1.0 release! nuff said

Re:Lighten up

[scooviduvoctagon]

They GNUStep guy announcing this was just trying to have some fun,

Right-on brutha...

I know it's early morning in the U.S., and gas prices suck... but lighten up, people!

Re:Lighten up

[seanellis]

Even I could see the tongue in cheek from here.

I thought that we on Slashdot are supposed to intellgent beings, not the kind of L33T D00DZ who have to have obvious humor put in <joke></joke> tags, with a liberal helping of :-) ;-) after everything and topped off with a LOL!!1!11!!

But look like I'm wrong (joke, LOL!ll111!11, :-), rimshot, etc. Sigh.)

Re:Lighten up

[molnarcs]

"They GNUStep guy announcing this was just trying to have some fun, why the hell to people get some riled up"

Because, incidentally, this is also a good way to create publicity for your pet project. Some would argue that this is page hit "whoring". Not that it is not a legitimate way of creating interest, but I understand those who have issues with this kind of "humor". In all honesty, you have to attribute a very good sense of irony (self-parody?) to the author to take the "joke" - and I don't know him enough to do just that. I don't exclude the possibility that what he said was in jest, but I understand those who get "riled up".

Re:Lighten up

[idontgno]

I dunno. Most times, humor is...um...funny.

I don't really care if the original poster was going for funny. There's a little TLA for this situation, usually reserved for fr1st psot lusers: YFI.

Besides the whole thing reeks of "haha only serious." A real case of Napoleon Chihuahua going on here.

Re:Lighten up

[nahdude812]

I thought that we on Slashdot are supposed to intellgent beings, not the kind of L33T D00DZ who have to have obvious humor put in tags, with a liberal helping of :-) ;-) after everything and topped off with a LOL!!1!11!!

You must be new here... ;-)

Re:Lighten up

[Archibald Buttle]

Here here!

One slight jab at the precious Gnome or KDE and you're bound to get over-reactions from people who don't understand what GNUstep and Gorm are all about.

GNUstep is all about the API folks. It's a dynamic object oriented API. Alien concepts to GNOME and KDE folks who are used to their static C/C++ APIs. It lets very cool things be possible, as can be seen in the demonstration videos for Gorm.

The downers are it's a bit of a different paradigm, and effective use of the API requires a dynamic OO language such as Objective-C or SmallTalk. Obj-C is very easy to learn though - a few hours and you'll be flying.

Re:Lighten up

[mpiktas]

If the only advantage of a project is its API, then it is doomed. The majority of the users do not care about API at all. They could care about new ways about doing things, things that fix long standing problems of other projects and etc. Gnome and KDE have produced working desktops, what did Gnustep produce? API and a tool to ease developing. Excuse me, I'm not developer, and I need to work, not to code a desktop in order for me to work.

The downers are it's a bit of a different paradigm, and effective use of the API requires a dynamic OO language such as Objective-C or SmallTalk. Obj-C is very easy to learn though - a few hours and you'll be flying.

In my opinion people should be flogged in public for saying things "oh TheNextBestThing is very easy to learn (use)". It maybe is easy for someone, but not for everyone. For someone who used "TheFormerBestThing" it is probably easy, for someone who never used anything it may be impossible.

Re:Lighten up

[senocular]

Because its the slashdot way. Haven't you learned that yet?

I'm still trying to imagine a beowolf cluster of GORM...

Re:Lighten up

[ajs]

"why the hell to people get some riled up by the obsoleting GNOME and KDE statement, have people completely lost their sense of humour?"

If Gnome were to release 3.0 and declare that, "GORM and KDE are now obsolete," I don't think the reaction from the GORM and KDE camps would be any different (permute the parties to taste).

It was a cheap marketting stunt, divisive, arrogant, and exactly what the proprietary software world expects of Open Source.

Re:Lighten up

[mad.frog]

The downers are it's a bit of a different paradigm, and effective use of the API requires a dynamic OO language such as Objective-C or SmallTalk. Obj-C is very easy to learn though - a few hours and you'll be flying.

See, there's the rub... I don't *want* to use Obj-C or Smalltalk. (And yes, I've used Obj-C on a major shipping product back in the NextStep days, so I don't speak out of ignorance.)

Give me a way to use it with a language I'm comfortable with, or you've lost me. C++ would be nice. EcmaScript/JavaScript/ActionScript would be nice too. Maybe even Java.

Re:Lighten up

[Kelson]

Well, some things are only funny if you're in the right frame of mind, which can be encouraged by hearing the speaker's tone of voice, seeing facial expressions and/or body language, etc. But if you reduce it to plain text, that's all gone. End result: if you happen to be in one frame of mind, it's funny (though in this case not very), and if you're in another frame of mind, your reaction is more likely to be "WTF?" or "whatever."

When I was in school we were told to write essays or creative writing assignments without bold, italics, or any other formatting (yes, we had bold and italics available, I'm not that old). It's tricky to learn how to get the right mood across with words alone. I won't claim to have mastered it myself, either, and I often resort to <sarcasm></sarcasm> tags and the like on Slashdot.

Re:Lighten up

[Archibald Buttle]

I don't *want* to use Obj-C or Smalltalk.

And you don't have to either. It's possible to use Obj-C classes from plain C.

Give me a way to use it with a language I'm comfortable with, or you've lost me. C++ would be nice. EcmaScript/JavaScript/ActionScript would be nice too. Maybe even Java.

Well, on the Mac Java is a supported language for most of Cocoa. Obj-C++ lets you mix in C++ code with Objective-C without needing to use the plain C calling conventions. Apple's widget stuff also gives some bridges to Cocoa for JavaScript.

GNUstep will catch up with Apple's stuff eventually.

open step vs cocoa

[minus_273]

I think three things will really help GNUstep. the first it the ability to read nib files. If GORM can load a OSX nib file, it will allows people to port the thousands of OSX apps they have made to GNUstep. Second, if they chnaged their target to Cocoa (link it to a version of OSX and release new versions with each release to add features).
The third and final thing is the appearance. GNUstep will never be popular looking thw way it does now. The default look looks too much like 1994 and unfortunately, many people will judge it based on that.

OSX + linux cross platform development would be a HUGE boost to linux.

Re:open step vs cocoa

[borgheron]

The target is currently Cocoa. GNUstep includes many Cocoa classes.

Re:open step vs cocoa

[TheRaven64]

the first it the ability to read nib files. If GORM can load a OSX nib file, it will allows people to port the thousands of OSX apps they have made to GNUstep.

There is a tool to convert nib files to GORM files. Native support for nibs is being added to GORM - a process made much easier by the fact that modern nib files are XML, and thus easier to reverse engineer than the older binary ones.

Second, if they chnaged their target to Cocoa (link it to a version of OSX and release new versions with each release to add features).

GNUstep is tracking Cocoa, and includes several non-OpenStep features from Cocoa, such as KVC and a few classes. I believe that 10.2-compatible is going to be the goal for GNUstep 1.0, which should be fairly soon.

The third and final thing is the appearance. GNUstep will never be popular looking thw way it does now. The default look looks too much like 1994 and unfortunately, many people will judge it based on that.

There is a theme engine for GNUstep, and a lot of effort was put into the design of the Nesedah theme for it.

Re:open step vs cocoa

[grahamlee]

The problem GNUstep is going to have in providing 10.2 compatibility is that 10.2 only had to run on Darwin 6.x, whereas GNUstep supports Linux, some BSDs (including Darwin), Solaris, Windows...take one of the 10.2 frameworks such as NSMailDelivery, it should be simple to supply that atop Pantomime. However, now look at DiscRecording.framework...

Re:open step vs cocoa

[TheRaven64]

To clarify, GNUstep provides Foundation and AppKit implementations. A few other things (like Address Book) are also available. GNUstep does not intend to provide all of the APIs that are available with OS X, however.

Re:open step vs cocoa

[grahamlee]

Flicking through a nearby Cocoa in a Nutshell for class names I don't recognise from OpenStep, the differences between Foundation on 10.2 and Foundation on OpenStep/Yellow Box/10.0 include NSAppleEvent*, NSAppleScript, NSScript*; all of which seem too platform-specific to be able to provide compatibility with on GNUstep. Similarly AppKit's NSQuickDrawView (which frankly should die under Cocoa anyway), even providing complete AppKit/Foundation compatibility at the 10.2 level seems overkill. I'm waiting for FreeBSD to work on my PowerMac G4, so that I can get back up to speed with GNUstep and try implementing NSNetService/NSNetServiceBrowser :-)

Re:open step vs cocoa

[TheRaven64]

NSAppleEvent*, NSAppleScript, NSScript*

These are, as far as I recall, to do with scripting. GNUstep provides StepTalk, which is a scripting framework similar in capabilities to Apple's implementation, but with a different interface. The default language for StepTalk is Smalltalk, but it would be relatively easy to add an AppleScript bundle. I believe StepTalk also runs on OS X.

NSQuickDrawView (which frankly should die under Cocoa anyway)

Agreed.

I'm waiting for FreeBSD to work on my PowerMac G4

Wait no longer. PowerPC support stabilised some time during the 5.x series. It's not widely advertised, and I think the PowerPC web page hasn't been updated for about two years, but it does work. Failing that, I run OpenBSD on a Mac Mini, and it works very well.

Re:open step vs cocoa

[grahamlee]

The patch to get PowerMac G4s working properly was too late to get into 6.0, but there's now a 7.0 snap which ought to work (I've tested the kernel, but not actually tried installing the OS yet). Hopefully....:-)

the bright side?

[amrust]

Don't be too quick to judge, guys.

From what I see, it looks every bit as eye-pleasing as OS/2!

/sarcasm

Re:the bright side?

[Hosiah]

as eye-pleasing as OS/2

No, nobody can possibly make that claim. OS/2 was the max, the ultra, the pinnacle of perfection that no other OS can ever match. People who didn't like it just weren't 31337 enough to understand it.

Oh, wait. The shrooms wore off!

I hope

[Anonymous Coward]

I hope it's prettier^H^H^H^H^H^H er, more stable, than the web server.

Microsoft Stole Gorm

[diablo-d3]

According to this insider's blog, Microsoft has stole Gorm, among other software.

Re:Microsoft Stole Gorm

[ScrewTivo]

That blog blows me away! A self proclaimed Microsoft employee stating he is on a team to take FOSS and put it into MS own products! I just can't beleive this.

GNUstep/Windowmaker

[Anonymous Coward]

I would love for GNUstep to really make a comeback...
Windowmaker used to be a really strong environment; Nowdays environments like xfce tended to scratch that itch.

GNUstep is another choice, not a replacement.

[Anonymous Coward]

Hhhmmmmm.

Me thinks that this poster is a bit sarcastic.

But whatever. GNUstep is a mature and well thought out system for power users. Not my cup of tea, but in absence of Gnome 2.4 and newer software I would probably be using it.

It's also great for systems with lower resources. X terminals, Pentium 2 machines, and the like. Very nice and is picking up the slack that KDE and Gnome leave as they race to beat Microsoft Vista (hopefully before Vista reaches critical mass 2-3 years after it finally gets released (MS still saying it's end of next year?))

If your like me and KDE makes you twitch nerviously, or unlike my you don't have a gig of RAM to deal with Gnome's concept of "simplicity thru complexity" then definately give GNUstep-based systems a look. (GNUstep is actually the API stuff, other projects do the desktop bits)

The nice thing about GNUstep that may attract people is that it's a implimentation for OpenSTEP.

Software previously developed for the Openstep API is what Apple used to create the 'modern' Cocoa half of OS X. (were as the 'older' half is Carbon which follows along the lines of OS 9 and OS 8).

Effectively this makes Cocoa a extended version of Openstep. GNUstep and Cocoa then share a high degree of API compatability. That means that if you write for Cocoa you can much more easily port your applications to run in Linux on Gnustep API and visa versa.

Re:GNUstep is another choice, not a replacement.

[Keyslapper]

I'm glad you cleared that up :)

As a general rule, I generally like to take the tack opposite the majority of responders, but this one is kinda hard. I generally like lightweight, but I don't see anything there that would make me drop my Fvwm2 desktop. I'll probably switch to KDE long before I bother trying this out.

Man, now I think of it, CmdrTaco was probably tossing grenades with this post just to see what happened. Ever see someone throw a long tailed cat into a room full of rockers then step back with an evil grin to wait for the chaos?

Man after my own heart. :)

Re:GNUstep is another choice, not a replacement.

[Anonymous Coward]

Beh!

xfce blows it out in speed, useability and prettiness.

gnustep is a bad rehash of the past.

Re:GNUstep is another choice, not a replacement.

[m50d]

I tried it and just found it looked horrible. Really really horrible. It was like I'd gone back 10 years to when the best option was CDE on solaris, only worse. Yes, it doesn't really affect the functionality, but for the typical user, looks matter as well as what it can do. I'll stick with KDE with beautiful shiny shiny widgets.

Re:GNUstep is another choice, not a replacement.

[linguae]

GNUstep looks like a very nice desktop environment for those who miss the NEXTSTEP/OPENSTEP look, feel, and functionality. (WindowMaker and AfterStep gives you the look and feel, but not the functionality). Now that the development tools are completed, now they can start getting to work on the applications.

Some applications that they do need are:

1. Web Browser
2. Word Processor
3. Spreadsheet
4. Multimedia Tools
5. Full-Featured Text Editor (kind of like KDE's Kate. Ports of Emacs and vi would also be nice)
6. Graphics Tools (like GIMP)

Some command line-applications (like mplayer and emacs/vi) written in C just need graphical wrappers, so porting them would be quite easy. Some of the GTK applications are written in pure C, so they can be ported to Objective-C by removing the GTK dependencies and replacing the code with GNUstep classes. (OS X users will also be pleased, as porting from GNUstep to OS X is just a compile away). Once we get the Objective-C++ sources from Apple (where you can mix Objective-C and C++), then we can get applications such as Camino and AbiWord, which have already been ported to OS X (using Cocoa), and porting them to GNUstep isn't too difficult from there.

Re:GNUstep is another choice, not a replacement.

[Anonymous Coward]

I agree that more GNUstep apps are needed. One current GNUstep multimedia app I like is an mp3/ogg player called Cynthiune. I use it as my regular music player. It's nice and simple and works fine.

Re:GNUstep is another choice, not a replacement.

[Anonymous Coward]

Well some of the applications you were asking for already exist in gnustep versions:
mplayer - MplayerGS (simple gnustep gui with playlist support that starts cmdline mplayer)
emacs - there is an aqua/gnustep version of emacs which is imo one of best emacs versions out there

for another incomplete list of apps check:
http://www.gnustep.org/GSWeb/GSApps.woa

check out the livecd for some examples:
http://www.linuks.mine.nu/gnustep/

happy gnustepping ;)

Trail Mix?

[MrWiggum]

I thought gorm was that damn granola stuff that my parents always made me eat on a hike.

Re:Trail Mix?

[MrWiggum]

Man you know your joke sucked when you get a response like this. Sorry to have inflicted it on all of you.

Re:Trail Mix?

[AnonymousKev]

> I thought gorm was that damn granola stuff that my parents always made me eat on a hike.

An all too common mistake, Mr. Wiggum. The word you're thinking of is 'gort', the tasty, yet completely moistureless mix of nuts, rasins, and small freeze-dried squirrels. The term 'gorm' is actually the species name of a Star Trek lizard (Capt. Kirk owns a pair of gorm skin boots).

But don't worry, the 'gor' words are often confusing. For example, many people forget that 'gorp' is the name of an alien robotic policeman who destroyed life as we know it in the classic SciFi movie It's a Wonderful Day to Stand Still on the Earth. And the word 'gork' (meaning "I knew that") was coined by writer Robert "Bob" Heinlein for his novel Strange Sex in a Strange Novel. There are hundreds of other 'gor' words (well, 26 max), but I don't have time to cover them all!

Hope this little chat has cleared up your lexographic apostrophe. I know it's helped mine.

Re:Trail Mix?

[schon]

The term 'gorm' is actually the species name of a Star Trek lizard

Actually, that's GORN, not Gorm.

I think "Gorm" is a bacterial culture that was marketed circa 2001 in the UK as a meat substitute. :o)

Re:Trail Mix?

[FuckTheModerators]

Like the AC says, trail mix is gorp, acronymized from Good Old Raisins and Peanuts.

Re:Trail Mix?

[FuckTheModerators]

I think your meat substitute is Quorn.

Re:Trail Mix?

[Art Tatum]

I think your meet substitute is Slashdot. :-)

Re:Trail Mix?

[Piquan]

Why did this get modded redundant? It's exactly right, and I didn't see anybody else say such a thing.

Anyway, Gorp can be remembered with the backronym "good old raisins & peanuts".

moderation

[tezbobobo]

This article seems a little 'flamable.' What is needed is a system whereby as you submit a response, you can choose whether you agree with the choice of the article. Once there are say, 300 responses, and >50% are negative, it is removed from the front page. This would make for a boring April 1st, but there'd be no more Dvorak articles (not to mention I actually like the guy).

Mirror?

[MiliusXP]

Someone have the mirror link ? The main link seem Slashdotted!!

Re:Mirror?

[GraemeDonaldson]

Gorm
GNUstep

Re:Mirror?

[Anonymous Coward]

nyud mirror:
http://www.gnustep.org.nyud.net:8090/experience/Go rm.html

Re:Which version of Max OS X?

[dloose]

Obviously he's referring to 10.1.1. What else would he be referring to? The current version of the software? That just wouldn't make sense, would it? PS you're dumb.

Death of KDE and Gnome

[everphilski]

Nooooooooooooooooooooooo!!!

-everphilski-

Re:Death of KDE and Gnome

[Mercano]

Kahhhhhhhn!

Re:Death of KDE and Gnome

[everphilski]

gah. it was a reference to the Star Wars article preceding it. But whatever. :)

-everphilski-

Looks like I'll have to stay...

[dkf]

GORMless

Gorm Videos Demonstrations

[roard]

I recorded a few videos (flash..) demonstrating Gorm ...

It's a bit tedious to explain with words what Gorm is all about -- it's much simpler to actually *see* it :-)

If you have only one video to see, check the one about the custom palette -- but the other are interesting too :-) (the StepTalk one demonstrate a creation of a simple calculator *entirely* in Gorm, using the StepTalk palette, which let you code in various languages).

Re:Gorm Videos Demonstrations

[Jason Hood]

Nice demos, however....

This has the look and feel of gnome 1.0. If they(you) want to get this product moving, they need pretty colours and components. Not in year or two when it might be more usable to grandma, but right now. In software development, impressions get set early and are very hard to break.

GUI builders are pretty standard nowadays although I still prefer not to use them.

Re:Gorm Videos Demonstrations

[cbv]

If they(you) want to get this product moving, they need pretty colours and components.

If you do not like the NeXT'ish look and feel that GNUstep defaults to, you're free to use themes.

Re:Gorm Videos Demonstrations

[ObsessiveMathsFreak]

If you have only one video to see, check the one about the custom palette -- but the other are interesting too :-)

Whew! After all that mousewaving I finally got through it.

Basically, what I'm seeing here is Visual Basic, with object orientation. Not to knock it because of this. It seems like a good system, but the VB paradigm was never my cup of tea. I'm sure VB programmers would love it though, if it was ever ported to Windows.(And they could understand Objects!)

The concept of a Gui creator is of course nice. No one should be coding GUI's in c or c++ or whatever. However I would still feel that XUL offers greater long term benefits. The power of HTML on the client side, and highly portable. Javascript is still a little iffy of course. I wish they'd improve it.

Re:Gorm Videos Demonstrations

[hey!]

Well, (i) the easy stuff should be easy without (ii) encouraging developers to paint themselves into a corner.

VB wins on on (i) but lose on (ii); most Java frameworks lose on (i) but sometimes win on (ii).

This is why RoR has made such a splash. It doesn't make hard things easy, it makes easy things easier without making hard things more complicated than they need to be. I believe this is high praise indeed.

Re:Gorm Videos Demonstrations

[roard]

Whew! After all that mousewaving I finally got through it.

Did you notice that there's actually sound (as in me, talking and making sense of all the mousewaving) in those videos ? :-)

what I'm seeing here is Visual Basic, with object orientation. Not to knock it because of this.

Frankly, no, it's vastly different. Or, in a way, yes, it's "VB with object orientation"... but:

• Properly done
• With easy integration of custom objects, custom views, custom widgets
• Not tied to graphical objects (you can also manage and connect non-graphical objects)
• Less code to type
• Emphasis on a good design (MVC)
• Cross platform
• Target/Action paradigm
• Wonderful programming framework (use of delegates, etc)

So yes, it's "VB with OO" if you want to see it that way :-D but it's FAR from following the "VB paradigm".

Re:Gorm Videos Demonstrations

[petaflop]

Well, I was impressed. I don't know if it makes GNOME/KDE obsolete, but it makes them look pretty gormless...

(sorry)

Re:Gorm Videos Demonstrations

[Dominic Burns]

I watched the second video while listening to System of a Down - I must be tired, but it took me about 5 minutes to realise that there should've been sound with it. Up till then I figured the mouse user suffered from a medical disorder.

Re:Gorm Videos Demonstrations

[Arandir]

the StepTalk one demonstrate a creation of a simple calculator *entirely* in Gorm

BFD. I can do the same thing entirely in Qt Designer.

The rumors of KDE/Gnome's demise..

[Mille Mots]

Gutsy statement, but I regret to inform Gorm's developers that Gnome and KDE went the way of the dodo as soon as I woke up from my 'desktop wars' fueled dreamstate and realized all I need is a way to launch Firefox, Evolution and maybe VMWare from time to time. Blackbox does that for me, with minimal overhead (at least compared to KDE and Gnome) and without getting in my way. I'd go find out if Gorm is as lightweight as Blackbox, but the site is aleady /.-ed. Somehow I doubt that it is, though, what with talk of 'Object Oriented Desktop' and making Gnome and KDE obsolete. :\

Why waste time trying to make my desktop work and act like Windows(tm)(r)(C)(and possible 666) when all I really want is to get my work done without all the bling?

--
A random sig
Dynamic
Saying nothing

Re:The rumors of KDE/Gnome's demise..

[bastardsquadmuzz]

The GNUStep desktop doesn't work or act at all like Windows. It acts like NextStep, which Apple incorporated into Mac OS X. It is fast, light, and very intuitive and pleasing to use. Gorm is a development environment for GNUStep, similar to Kdevelop/QTdevelop (I forget which it is called) for KDE or the Interface Builder for Mac OS X.

Re:The rumors of KDE/Gnome's demise..

[meowsqueak]

Gorm isn't a desktop environment - it's a development tool like Glade or Hypercard (I think - never used that, but I have used Pythoncard) and is used to design GUIs for the GNUStep environment. The power of this environment is in Objective C and the closely related NEXT libraries, which provide a well-designed framework for developing consistent interfaces and code.

If you want to know what a GNU/NEXTStep environment sort-of looks like, try the 'windowmaker' window manager.

Re:The rumors of KDE/Gnome's demise..

[An Onerous Coward]

For the most part, I agree with you. I mostly ignore the desktop and use a couple of applications. Mostly I manipulate files and launch apps from the command line.

Yet I'm stuck on Gnome, and for possibly the silliest of reasons. A while back--and only in Gnome--I figured out how to make Firefox use Emacs-like keystrokes. When typing on a form, I can use Ctrl-(P,N,F,B) to move the cursor, Ctrl-E to yank everything from the cursor to the end of the line to the clipboard, etc.

It's a little thing, but it makes filling out web forms much more convenient for me. Since I don't know how to make this work on other desktops, I use Gnome. Most people have a favorite desktop, and the reasons for their preferences seem arbitrary to people who aren't them.

Re:The rumors of KDE/Gnome's demise..

[Anonymous Coward]

Blackbox! Hah! I use the far more basic FVWM. And not any 'ole FVWM. No, I use FVWM 1, which hasn't been updated since 1996. This is an extremely productive setup for me (albeit not easy-to-learn; but ni modo because I already learned it in 1995). And very lightweight--it's even more lightweight than Windows XP. I guess X has so much overhead that Windows can do a lot more at the same level of CPU power.

Gormless

[fuyu-no-neko]

with its release, comes the obsolesence of the GNOME and KDE projects.

I was about to call that a gormless prediction, but then I realised that that was the whole point... -.-

So they mean...

[TheMadPenguin]

this is going to take on modern GNOME/KDE? I'll try to contain my fear... But with that I must say I will welcome our 1995 overlords with cookies and milk that has past its expiration

Re:So they mean...

[An Onerous Coward]

People more familiar with the project might wail in anger when I say this, but my impression is that GORM isn't really the point. It seems like GORM is just an excuse to show off the GNUStep toolkit (which bears the same relationship to GORM that GTK does to to Gnome, and QT does to KDE). Since GNUStep is pretty similar to Apple's GUI toolkit (Cocoa), there may be something promising here.

Most likely, as GNUStep gets more popular, Gnome and KDE will be able to run that toolkit's apps (much as they run each others' apps today). That will be cool, because it means that you can write a Linux app whose GUI ports easily to Mac.

Note: I probably don't know what I'm talking about. Anyone with more of a clue, please correct me.

ITYM gorp

[Mille Mots]

Granola, oats, raisins, peanuts. Although Wikipedia's entry says it might also mean Good Old Raisins and Peanuts.

Apple?

[tezbobobo]

Ignoring the solid relationship to apple:

                  FTA - After creating the interface, objects can be linked using mouse operations. Also Gorm features interactive testing of interfaces.

This sounds very much like what Apple has had for a very long time with its XCode tools. Whilst its great that apple has had this for a long time, it will provide something very new for KDE and Gnome users. Perhaps it will provide some serious competition for these.

Re:Apple?

[John Allsup]

IIRC XCode, in its previous guise as Project Center and Interface Builder has had this since before Apple even got hold of Next. Though I may be wrong.

Having used XCode I must say that its a very good way of doing things, though once in a while I'd like to know what's actually going on (i.e. what's in a .nib file and what happens as it is loaded), and XCode tends to hide this. And there's still much that Xcode can learn event from (dare I say it) M$VC.Net.

Anyway, it's a little sad that Objective C is routinely getting dropped from modern distros (e.g. the OpenSuse DVD doesn't even have it as an installable option) and this makes GNUstep (especially GNUstep base IMO) less accessable than it should be. Given the size of GNUstep it should really appear as an optional developer package on most distros.

Bull stuff.

[Anonymous Coward]

No kidding! Looking at the screenshot it's an even more astounding claim. This Gorm is simply an interface designer that looks like a steaming pile when compared with the likes of QT Designer.

Re:Bull stuff.

[Anonymous Coward]

QT Designer looks pretty good. Does it run on Windows?

Haha, only marketing

[idontgno]

...the obsolesence of the GNOME and KDE projects

-1 flamebait, or maybe -1 troll, or -1 stupid advertising.

I know, I know, "it's a joke". Yeah. Right.

I use Openstep some. It's nice. But a really tricked out Accord won't be obsoleting BMW any time soon, either.

Re:Haha, only marketing

[BushCheney08]

But a really tricked out Accord won't be obsoleting BMW any time soon, either.

Wait a second. Which one is the Accord and which one is the BMW?

Re:Haha, only marketing

[maxume]

But a really tricked out Accord won't be obsoleting BMW any time soon, either

Acura TL very nice car!

Gorn?

[kmcardle]

I thought this was a Star Trek story at first look. I was hoping for a good spinoff series where the Gorn conquer the Federation.

Re:the obsolence?

[Anonymous Coward]

>"...and with its release, comes the obsolesence of the GNOME and KDE projects"

>you're joking, right?

Yes it is a joke....
It seems that slashdoters that not have sense of humour ! :)

GORM...

[segedunum]

...less? I think they could have picked a better name. It's worse than Gnome, or should I say, 'G'-nome.

1.0 already!?

[b100dian]

Since when free software software begins its releases with 1.0??!

Re:1.0 already!?

[Anonymous Coward]

I don't know ... the first formal release of Gorm was 0.0.2, but I guess you weren't watching.

Re:1.0 already!?

[Art Tatum]

Heh, Gorm was actually first released in December of 1998 and has been constantly worked on since then.

enlighten me please

[confused one]

I don't understand. From what I see, this doesn't give you much over using the old Athena libraries (visually) or using TK (as in Tcl/TK or Perl/TK). How is this an improvement over using KDE/Qt or Gnome

Re:enlighten me please

[Julian Morrison]

OpenStep (the standard GnuStep is cloning) comprises a high/medium/low level library, GUI widgets, standard look and feel, IPC, and basically all the functions of a *DE. As such it's more equivalent to Java (which is also a bundle of everything) than KDE or Gnome (which lean heavily on preexisting libs).

However, it's also 1990s tech, and it shows. Like one of those hand-wired Crays, it's a beautifully built museum piece.

Re:enlighten me please

[be-fan]

GNUStep is based on OpenStep, the same programming framework OS X's Cocoa is based on. The biggest difference between Cocoa and GNUStep is that the former has had a once-over by all of Apple's artists :)

Re:enlighten me please

[ScottyH]

Not true. This object model is the best I've ever worked with, hands down. The ease at which complex tasks can be performed is astounding. It may be 1990s tech, but every other framework has a lot to learn from this one.

What fresh hell IS this?

[Hosiah]

Why is the icon from the top of my Window-Maker dock at the top of Slashdot? I thought I was the only one who knew that desktop existed (it's my second favorite after Fluxbox!). What's Gorm? Why is a GNU app, or ANY Open Source app, touted like a World Wrestling cheesehead brought into the show in the second half? I don't care if your open-source app is the second coming of Buddha, though shalt not challenge thy fellow FOSS! What's this with WM's lineage? I thought it was descended from NextStep? I can't reach any of the links. Has OpenStep just thrown away it's only chance for popularity with this dishrag post?

Re:What fresh hell IS this?

[borgheron]

It is the GNUstep icon, not the WindowMaker icon.

The symbol that is used for WindowMaker's dock was originally created for the GNUstep project by Pyrros Atheos and was later used by the WindowMaker window manager team.

GJC

Re:You're close

[Anonymous Coward]

Fuck Mac OS X. I have a Mac and still prefer Gnome or KDE (I have used both extensively and like both). OS X is a sorry excuse for a UNIX. It doesn't even have a working native terminal program (one which allows PGUP/PGDN), and Fink's repository is very poorly maintained. Running X apps doesn't work all that great (why don't they show up in the Dock?). Maybe with a huge amount of effort (and buying a few shareware apps), I could get OS X to work how I want it. But why would I bother when I can have Linux working out-of-box the way I want it? Why would a want a non-free OS when a free one works just as good if not better?

There are a few apps that are really great on OS X, but that is all I use it for. Linux is my primary OS. Just because all the pussy, wannabe users fled to OS X, doesn't mean that OS X is better. It's better than Windows, but that's about all that can be said for OS X.

GORP

[Anonymous Coward]

I am not worried about Gnome or KDE...

Am I the only one who views this is a over the top
attempt to edge out peanuts in favor of M&M's

In other news ...

[lheal]

The VIM development team just announced that emacs is obsolete.

Re:In other news ...

[PlacidPundit]

That's not news!

Re:In other news ...

[daeley]

That's not news!

You must be new here! ;)

Re:In other news ...

[sik0fewl]

That's hardly news.

GNUStep.........naaaa!

[sheepoo]

Frankly speaking I do not think GNUStep is something which an everyday user can use. A geek maybe, but not an everyday user. KDE and GNOME are way ahead of GNUStep in terms of usability so I do nto think that there is any chance of 'obsolescence' there :)

Re:GNUStep.........naaaa!

[Art Tatum]

Frankly speaking I do not think GNUStep is something which an everyday user can use.

You probably don't realize how right you are. :-) GNUstep isn't a desktop like GNOME and KDE are. It's a development environment, more akin to QT for KDE or GTK and assorted libraries for GNOME.

More information, missing links

[tarzeau]

The OpenStep Standard. And the object oriented C. And how Interface Builder looked on OPENSTEP. And the live CD with the older version (soon to be updated).

install

[Anonymous Coward]

I'll be sure to install it next time it's 1989

I smell a lawsuit coming..

[Steeltoe]

The Gorm developers better wait this one out 'till Avalon is out the door. Eolas will seem like pancakes compared to the lawsuit Microsoft is going to face for this..

The real beauty of it all, is when all is said and done at the end of the day, the GPL will ensure Avalon becomes GPL and Free Software for all to use. Just think, the simplicity of Gorm and the beautiful Avalon, a happy merger for all enthusiasts. Ironically, this WILL spell the end of GNOME, KDE, XFce and even fvvwm!

Microsoft will regret the day they Gormed their Desktop. It's like playing with nitroglyserin, except you always explode (for those who've seen Lost, you know what I mean. It ain't pretty..)

Microsoft dies into a puddle of blood.
KDE lies in shatters on the floot.
GNOME runs and hides under the carpet.
Gorm cackles with insane glee while spinning maniacly in small circles.

A new Gorm-era is in the making, as the wheel of time keeps spinning, making everybody pretty dizzy.

Re:I smell a lawsuit coming..

[Anonymous Coward]

...that is a fairly obvious parody.

Quick Kirk!

[JabberWokky]

All Kirk needs to do is lash together some bamboo filled with coal, sulfur and saltpeter so he can fire diamond chunks and he can kill it!

--
Evan

obsolutely

[Doc Ruby]

And the Hurd makes GNU/Linux obsolescent.

Why don't these developers want to take over the world by first taking over their "competition"? Like making GORM run on GNOME/KDE? Even GNOME and KDE apps run on each other when their libraries are installed. If GORM really is that good, it should interoperate, and suck up all the oxygen to emerge as the best. That's the kind of innovation hothouse open source allows.

Obligatory

[dos_dude]

I, for one, welcome our new flamebaiting overlords.

Finally a GUI. This is revolutionary !!!!!!!!

[JustBen]

Wow you mean I can finally harness the power of an out dated 15 year old GUI on my brand new computer. It's really to bad that there has not been any development in the GUI market since the early 90's.

In other news
Who gives a shit.

Pronunciation...

[Anonymous Coward]

Why do people call it Gee-nome? You don't hear people say "My neighbour has Gee-nomes on his lawn." or how about the "Underwear Gee-nomes were mad at Kenny for going commando!". Gnome is pronounced with a silent G. You are making us all look bad - and you know who you are!

Overkill

[Anonymous Coward]

Who needs more than a X window manager like mwm, Firefox, an Xterm with emacs?

K5 has that

[Kadin2048]

So basically like Kuro5hin, just without the politics or inane trolling..?

Re:You're close

[Savage-Rabbit]

There are a few apps that are really great on OS X, but that is all I use it for. Linux is my primary OS. Just because all the pussy, wannabe users fled to OS X, doesn't mean that OS X is better. It's better than Windows, but that's about all that can be said for OS X.

Let me guess... you sat on a cactus this morning and now you are venting your pain in words?

And where are the [Python|Perl|Ruby] bindings?

[zapadoo]

Don't tell me only 'C' developers need apply. Talk about limiting your audience.

With Python I can develop responsive cross platform UI using wxWidgets, PyGTK, QT... can I do that with Gorm?

(Gorm sounds like the noise Gollum makes when choking on his tongue)

Re:And where are the [Python|Perl|Ruby] bindings?

[ceoyoyo]

Hopefully PyObjC will run out of the box, or nearly so with OpenStep. Python/PyObjC/Cocoa on the Mac is THE best development system I've ever used. OpenStep and a way to use the nibs (or XML export of them) on other platforms would give me back platform independence... well, except for the CoreImage stuff.

Seriously, check out Python and PyObjC. Write your 90% code in Python, write your 10% gotta-run-fast stuff in ObjC and it's automatically wrapped for you. The best of both worlds. My apps have interpreted parts, parts running on the CPU and parts running on the GPU, all for free.

Re:And where are the [Python|Perl|Ruby] bindings?

[Art Tatum]

I don't know if PyObjC works with GNUstep, but there are Smalltalk, Ruby, and Java bindings for GNUstep in the core CVS.

Re:And where are the [Python|Perl|Ruby] bindings?

[bsartist]

CamelBones is getting there. It compiles, at least on my system, but not all of the self-tests pass.

Legacy?

[Anonymous Coward]

I think it's the other way around, isn't OpenStep more like the predecessor to MAC OS X?

Um,

[Minwee]

The problem with reading press releases for so long is that I have lost the ability to tell when someone is being serious or just trying to be sarcastic.

It used to be a simple thing. If something is presented in a fairly straightforward, just the facts manner it was probably serious. If it makes ridiculous claims about how it will make all of its competitors obsolete, cure seventeen fatal diseases and then get you a beer while walking your dog then you could be pretty certain that it was meant to be a joke.

I used to even be able to laugh at the joke press releases, knowing that they were nothing more than way-over-the-top satire of the dumbest PR pieces in the world. Now... I have to look carefully to make sure that I'm not actually reading the object of that satire instead.

Thank you, PR flacks of the Internet, for lowering the bar so far that we need a shovel just to see the dent that it left.

Re:Um,

[Art Tatum]

Well, several more straight-forward and subdued attempts were made to post this story (about 12 or so, by different individuals) and they were all rejected, frustrating the hell out of the developers. The only way to get it through seems to be the use of Slashdot keywords like GNOME and KDE. Slashdot has become your one-stop shop for sensationalism and outrageous claims since that's all that can get the attention of the posting team.

Re:Um,

[Minwee]

Judging by some of the stories that do get posted here, promises of sex can help too.

The Young And The Gormless

[Dystopian Rebel]

Few North Americans will get the joke. And the name of this project is further proof that they're mostly gormless about internationalization.

Re:The Young And The Gormless

[borgheron]

I'm sure most of you gimps won't mind.

As a "mid-level" user...

[Kootaphor]

..by far, the best "pseudo-programming" environment I've ever worked in was LiteStep for Windows. I just made the switch to Linux last year, and I have yet to find a tool as valuable as LiteStep, and frankly, I was surprised that there weren't more modular, easily modifiable interface creation systems like this one. Both Gnome and KDE felt like a step back when I started with the penguin. A s much as I enjoy learning about programming, as a lawyer, I really don't have time to learn about stacks and arrays and pointers and object classes. But an environment like LiteStep, where I can create and specify scriptable actions for things like buttons in "dang-near-english--" e.g. ButtonXLength 10 ButtonXWidth 20 ButtonXText "Press to start firefox" ButtonXAction !Execute "C:/firefox.exe" This is perfect. And also the kind of thing that I imagine hardcore programmers hate, but it's precisely this level of control that us Linux folks (yes, I'm counting myself among you's guys) can use as a real big selling point OVER M$. I will be trying out this Gorm ASAP.

Re:As a "mid-level" user...

[Anonymous Coward]

So YOU'RE the one COBOL was created for!

GNUstep targets Cocoa

[Kaseijin]

Code which targets OpenStep will work on both Cocoa and GNUstep - however from thereon there are divergences...GNUstep has classes which Cocoa lacks and vice versa. It would be possible to reimplement the missing classes on the other system, just it hasn't been done.

I doubt they've implemented 100% of Apple's classes, but the GNUstep developers have made Cocoa compatibility an explicit goal.

Re:GNUstep targets Cocoa

[moo083]

How does GNUstep deal with the fact that some of the code behind Cocoa is closed source? The interfaces are made with Interface Builder which creates a .nib file which has your interface. But you cannot actually see the source for it because I believe it comes as, I guess, precompiled you could say. How do they expect cross-compatibility with this in mind?

Re:GNUstep targets Cocoa

[Archibald Buttle]

Good luck to them with the Cocoa compatibility.

The problem is that Cocoa is a moving target. Mac OS X 10.3 had a major update when they added in Data Bindings along with associated technologies of Key-Value Observing, Key-Value Coding and Key-Value Validation. 10.4 has had the addition of Core Data - there's quite a few new classes in that lot.

With a bit of luck though Gorm will get people more enthusiastic about GNUstep and bring more developers along to work on the core APIs, helping to fill in the gaps in Cocoa compatibility.

Re:GNUstep targets Cocoa

[Anonymous Coward]

Umh, because only the public face needs exact replication. If there is no public face, it's not an interface programmers can use, and therefore it's irrelevant for apps.

Re:GNUstep targets Cocoa

[tyrione]

Apple is making nib files regular xml based for the next release of OS X.

Re:GNUstep targets Cocoa

[plazman30]

Apple stated that with the 10.4 Tiger release, that the APIs are pretty much fixed now. I know that can always change, but I tend to think the APIs will not be in the same state of flux they were under all previous OS X versions. That's part of the reason newer versions of Safari, and apps like iPhoto required newer versions of the OS. They use APIs that did not exist in older versions of OS X(It also helped sell newer versions of OS X, so Apple isn't totally blameless for this practice...)

I hope GnuSTEP gets good Cocoa compatibility. This would be a big plus for Apple and GnuStep. A lot of the really big apps such as the Adobe Suite, MS Office, FileMaker, etc are all written in Carbon and could not be ported to GnuStep, but there are plenty of small developers writing Cocoa apps, that would probably recompile under GnuStep just to see if it works, and might possibly release those products.

GnuStep could become to Apple, what Wine is to Microsoft (though somehow I think that Apple would embrace GnuStep a little better than Microsoft treats Wine...)

Andy

Re:GNUstep targets Cocoa

[TheRaven64]

GNUstep already has KVC. I believe KVO is done, but not yet committed to the main tree. Core Data is a work in progress, but seems to be doing quite well. They do have GDL2, which is similar to EOF - the object-relational component of WebObjects which has a lot in common with CoreData.

Re:GNUstep targets Cocoa

[Elektroschock]

Sure, but that would mean more developers are needed to step in. We currently do not even have diffs between Cocoa and GnuStep. So nobody can observe what isn't there yet.

Re:GNUstep targets Cocoa

[jcr]

Apple stated that with the 10.4 Tiger release, that the APIs are pretty much fixed now

When did they say that? I was working at Apple when Tiger shipped, and that's news to me.

-jcr

Tongue firmly planted in cheek...

[borgheron]

All,

For all of those of you who can't take a joke, tongue was firmly planted in cheek regarding the "KDE/GNOME obsolesence" bit of the post. While I didn't write the post, I know who did and that part, at least, was meant as a joke. Also look on it as something of a commentary on slashdot itself: sometimes it's impossible to get anything on here unless it's sensationalistic or overly stated.

I, personally, tried posting 6 times before giving up. Imagine my suprise at seeing this when I woke up this morning!

Later, GJC

all eyes on GORM

[se7en11]

Eventhough they may not meet up to thier claim, everyone seems to be at least taking a look at it. I work with a marketing agency who does direct marketing mailings. If we get a responce from over 1%, we are doing well.

Urgh, how crap is that ?

[Anonymous Coward]

Hey GORMless.

The 70's called.

They want their interface back.

Features of Gorm vs. Glade

[borgheron]

Compare Glade to Gorm feature-wise. Aside from the appearance of GNUStep, which admittedly stands to be improved, you will find that some of the features Gorm has, Glade blatantly lacks. One thing in particular is the ability to use custom palettes. Glade comes with a standard set of widgets, period... also Glade does not allow the modelling of non-gui objects, Gorm does.

So from that standpoint alone Gorm is compares favorably. I think you better step back from that "BMW" for a while and ask yourself what you're really getting.

GJC

gnome & kde obsoleted?

[dindi]

Well I don't think so, I used to be an FVWM, then afterstep user for a looooooong time before I switched to Wmaker ....

then I tried Gnome because I used redhat for a while and for somewhat reason I had a problem with Wmaker ...

since then I use Gnome, and while I am not happy with it I got used to it. On Xinerama it has a few problems, and the "remember window state" is pretty much broken in it.

I think the claim that it will take gnome and kde over is ridiculous for the masses, however it might be well possible for me....

I do not need the KDE and GNOME "christmas tree" bloat, but sometimes it makes life easier ....

For the masses ?
If I gave that or wmaker to my wife she would not be using linux, with gnome she is pretty happy

Actually, ASM is too high

[A nonymous Coward]

His point is that everything compiles down to, or at least ultimately runs as, machine language. By definition, not only CAN everything be done in machine language (or asm), but everything IS done in asm.

He's being sarcastic.

kernel?

[anomaly]

No, but it's theoretically possible. You'd have to have a perl compiler, of course. Perl has many asperations, but I'm not aware of the interpreter itself becoming an OS. :)

Dated, but on point....
http://plg.uwaterloo.ca/~glmclear/research/perlos/

Respectfully,
Anomaly

Fugly!

[Anonymous Coward]

As long as GNUStep is still ASS UGLY I don't think KDE and GNOME will be going away any time soon.

Gorm-less

[figlet]

I prefer to remain Gormless.

;-)

Jerk

[Somegeek]

Several dictionaries and half a million web sites say that it is a viable alternate spelling. Make that Cowardly Jerk. Get a life.

Since most people seem to have missed the joke...

[pschmied]

...and are involved in a flame war, why don't we start a constructive thread?

You GNUStep users out there: How do you configure your desktop? I've noticed that GWorkspace and Window Maker like to compete for how they manage virtual desktops.

Also, I'm not sure what application I should be using as a dock? Window Maker? Something written in GNUStep?

GNUStep seems like it's just about ready to start seeing some early adopter use, though I agree that the UI could use some spit and polish. How is work on Camaelon coming?

Also, does anyone know if Gorm is supporting Renaissance yet? It would be lovely to be able to target user interfaces for both Mac OS X and GNUStep. Maybe Gorm could be used as a drop-in replacement for Interface Builder when one wants to make a cross-platform app supporting GNUStep and OS X Cocoa?

-Peter

A Light Bulb Comes On..

[borroff]

...NOW I know what "gormless" means!

GNUstep deserves more attention

[Anonymous Coward]

I'm glad to see that GORM 1.0 has been released. IMO, the GNUstep project does not get the amount of attention from Linux users it deserves.

I use GNUstep on my Linux machines both at work and at home, and I have been very pleased with it. It's a very consistent desktop environment with a "clean" UI. For people who consider GNOME and KDE to be bloated and/or buggy, and don't care for "eye candy" that isn't all that functional, I'd recommend trying out GNUstep. I actually like the look of it; it has a cleaner and different look to it than GNOME/KDE, which seem more Windows-ish and Fisher-Price-like. And it's very simple to develop apps for it using GORM. I like the fact that it uses the NeXTSTEP API, which has been around a while, in contrast to the APIs of other desktop environments. Go to gnustep.org to see some of the apps that have been developed for it. Not as many as for GNOME/KDE, but hopefully that will change.

-1 Troll

[reed]

Wow, this story is one big troll. Which sucks, because (Open|Gnu|NeXT)Step is cool and Objective-c is even cooler.

Eat me: Re:Lighten up

[Anonymous Coward]

Dude, you're nuts. I would say almost every single thing that is brought to our "attention" is the result of PR whoring of one sort or another. I think if there were an announcement of a teleportation device today, some /.'rs would claim that this was obvious and no big deal, to move along, nothing to see, choose your retarded /. cliche de jour that puts forth the superiority of the commentor over the suplied news/posters. I'm almost positive someone will supply the same here.

The sad truth is every post here seems to follow dvorak's statistical line-up of responses (and I'm doing my part here). The sadder truth is that /.'ers are quick to discount everyone's work as "no big deal" with great regularity, with the selfish exception of their own work. Such is human nature I suppose, and such is the petty nature of people that post and criticize on message boards.

That being said, everyone here that is commenting on how valuable or not GORm is should consider themselves disqualified unless they used some real OOP/RAD setup. C++ not only disqualifies you, it may mess your ability to OOP well permanently. If you have some background in Smalltalk, (maybe eiffel blah), or Objective C, or a handful of other "real" OOP backgrounds then you might have a chance at "getting it." Like it or not, if you haven't, you probably do not and will not ever "get" OOP well. Those that use Java are more likely than not a mixed bag. Those coming from a traditional procedural background and moving to Java, you likely don't get it either. Which is somewhat sadly ironic in that the designers of Java were big fans of Objective-C and modeled quite a bit of Java on Objective-C.

Anyway, now that I've strayed into flaimbait tangent, let me bring this back and try to establish this as gospel: All our news comes as hype and its the only thing that has a chance at getting our attention. I don't think that some "righteous" way even exists in having a major story hit that is devoid of hype anymore. That's more a function of who we are as people today--the media knows us well enough to deliver things in a way we'll accept. Even your "griping" and my "counter griping" at it being hype are part of the equation. This is so straightforward now that every average person knows that you just put stuff out with something that will agitate the flock of cackling chickens we've all become to get the desired response.

Re:Eat me: Re:Lighten up

[Anonymous Coward]

wanna run for president? we need sane people like you who know the score

AC is 08!

The Power of Choice

[blueZhift]

Hah! I like the idea of GORM and I hope that we'll see something similar for KDE and Gnome. The choice of desktop environment is one of the things I like about Linux, even if it touches off holy wars every now and then. In the end I think these wars will result in a better desktop experience not only for Linux users, but Windows users as well since Microsoft isn't above stealing good ideas for themselves. Oh, what about Mac users you say? The Mac desktop is already perfect, nothing needed there! ;-)

(uh that last bit was a joke...)

Looks nice, is a step in the right direction

[bensch128]

Gorm actually looks pretty neat.
Too bad it needs several years of work to catch up to
KDE/GNOME in terms of infurstructure work.

It really does illistrate the power of scripted interfaces through.
I want to see the same thing (but better!! :) coming from QtRuby/Korundum
since it sould be SOOOOO easy to do so.

Great demo,
Ben

Re:Looks nice, is a step in the right direction

[roard]

Gorm actually looks pretty neat. Too bad it needs several years of work to catch up to KDE/GNOME in terms of infurstructure work.

Actually, no, that's the point. Not that much is missing... the problem with GNUstep is that before implementing nearly the complete api (which is done) and having an InterfaceBuilder like (which is done by Gorm), it was rather difficult to use it and be it of any interest... Thus KDE and GNOME catched rather than GNUstep, in my opinion because they were able to incrementally build their desktop, and thus, propose something immediately of some use. Vicious circle of course for gnustep, as it then didn't attract many developers, etc.,etc.

It really does illistrate the power of scripted interfaces through.

Scripted interfaces ???? Gorm is not that, at all :) ... if you just saw the StepTalk demo, check the rest :) -- The StepTalk Palette is just a "plugin" for Gorm if you want...

I want to see the same thing (but better!! :) coming from QtRuby/Korundum since it sould be SOOOOO easy to do so.

That, I doubt it.. although Ruby is dynamic enough to build interesting things.. but the philosophy of OpenStep is rather different..

Re:Looks nice, is a step in the right direction

[bensch128]

Scripted interfaces ???? Gorm is not that, at all :) ... if you just saw the StepTalk demo, check the rest :) -- The StepTalk Palette is just a "plugin" for Gorm if you want...

It's scripted in the sense that the executable runs without any compilation step and because Objective-C is a dynamic language (interperted) at runtime I believe. Do you have any links for plugins for Gorm? It seems like the whole interface was based on the user writing ObjC for complex actions/methods.


That, I doubt it.. although Ruby is dynamic enough to build interesting things.. but the philosophy of OpenStep is rather different..

Oh, how does the philosophy matter at all? i didn't see anything in the demo which couldn't be done on a different platform and the right interface builder. I mean QT designer can do 50% of what Gorm does with the exception of the custom actions. (I admit that is a large shortcoming though). However the entire concept of writing actions in a tiny window will not scale for medium to large applications. So I hope Obj-C's import facilities are good. Just look at the hoops Flash developers go through.

Cheers,
Ben

Re:Looks nice, is a step in the right direction

[roard]

It's scripted in the sense that the executable runs without any compilation step and because Objective-C is a dynamic language (interperted) at runtime I believe.

Well... you're wrong. Objective-C is a compiled language, not an interpreted one.

Do you have any links for plugins for Gorm? It seems like the whole interface was based on the user writing ObjC for complex actions/methods.

Indeed, the basic idea is to use ObjC for complex actions/methods (considers though that ObjC is a strict superset of the ANSI C, so you could simply put your C code in an ObjC method without any modifications, and the next release of gcc will even allow you to mix ObjC and C++). There's no repository for Gorm palettes, but you can have a look on my blog to get the StepTalk palette.

Oh, how does the philosophy matter at all?

By the philosophy is different, I'm talking about the framework (GNUstep/OpenStep) and the language (ObjC) -- they are vastly different from Qt and C++. And frankly, I rather like Qt, programmed with it before jumping on the GNUstep ship; the signals/slots mechanism of Qt is actually quite inspired by OpenStep target/action. But ObjC and OpenStep are without a doubt much better suited for the job than Qt/C++, really (I would use C++ for a 3D engine, but ObjC for the gui -- the right tool for the right job, period). Partly because of ObjC functionalities and its dynamism (introspection, modify things at runtime, true polymorphism, replace a class by another, catch any messages, etc.) and partly because of the sheer design quality of OpenStep. It's an amazingly well designed framework. I may remind you that Cocoa on OSX is actually an OpenStep implementation..

i didn't see anything in the demo which couldn't be done on a different platform and the right interface builder. I mean QT designer can do 50% of what Gorm does with the exception of the custom actions. (I admit that is a large shortcoming though). However the entire concept of writing actions in a tiny window will not scale for medium to large applications. So I hope Obj-C's import facilities are good. Just look at the hoops Flash developers go through.

Ouch.. the point of Gorm is frankly NOT to write actions in a tiny window -- actually Gorm don't do it, in the demo it was done by the StepTalk palette. Gorm is an Object Modeller more than a GUI builder, in a sense. Its real job is to connect objects, graphical or non graphical. And you're working on the "real" objects. A .gorm file is simply the serialization of the object graph... So, no, you can't do that in any other UI builder, apart (obviously) from InterfaceBuilder on OPENSTEP^WMacOSX :-)

And yes, of course you usually import ObjC headers in Gorm, or create the class in Gorm itself. If you only saw the "StepTalk" demo, have a look to the Dataset demo, and if you want to see an example of ObjC code with Gorm, check the third demo.

Re:Looks nice, is a step in the right direction

[bensch128]

Gorm is an Object Modeller more than a GUI builder, in a sense. Its real job is to connect objects, graphical or non graphical. And you're working on the "real" objects. A .gorm file is simply the serialization of the object graph... So, no, you can't do that in any other UI builder, apart (obviously) from InterfaceBuilder on OPENSTEP^WMacOSX :-)

this is not so obvious from the demo. And it is very difficult to tell what is connected to what because the edges are hidden. I suppose you want to emulate CoreData from Apple (or UML) but there the connections between objects are visible.

That being said, I'll have to look into it and ObjC when I have the time.

Aesthetics

[eGabriel]

I don't understand why people say this is ugly. The lines are straight, the fonts readable, the interface usable. I don't understand most of the time why people think such and such a UI is ugly. I understand that for these people, aesthetics are a real issue, but their decision-making process seems totally arbitrary to me. Usually it seems that "ugly" just means not in line with the current artistic trend; what is ugly today may have been beautiful 10 years ago to even the same people.

It's sort of like people who loved synth pop in the eighties and now say it sounds cheesy. If it was good then, why is it cheesy now?

I've even heard people call Nethack ugly. How can it be ugly? It's just a bunch of text.

Anyway, it sort of saddens me that aesthetics are so heavily weighted over more important concerns, but I guess I have to accept it.

Just three words:

[Anonymous Coward]

Smoke my cock.

Eye candy???

[Urusai]

That *step interface looked hot in the early 90s, but now we have skinnable GUIs instead of Motif and the like. Changing colors doesn't count as "themable", BTW.

Re:Eye candy???

[thre5her]

It is themeable.

Ass.

Re:Eye candy???

[RAMMS EIN]

Alright, now I believe it. So far I had only seen posts claiming that GNUStep is themeable, cause, look, we've got Camaelon, and it's got this theme that is a little different from the standard look. No evidence that any other themes were available.

So, is the theme engine actually part of the base, or at least easy to install? What themes are available, and where can I see screenshots? How do I make my own themes?

Much as I like the arrangement and operation of the GNUStep UI, I don't like the looks of it, and I won't take the effort to install it all and get it up and running only to find out I can't change the theme.

So, why all the jokes?

[Overly Critical Guy]

I'll never get Slashdotters' obsession with KDE and GNOME. Both projects absolutely suck. Their APIs are a joke. GNUStep/OpenStep, on the other hand, is a true object-oriented environment, and it really does make Gnome and KDE obsolete.

But, as with their defense of the ancient X11 protocol, Slashdotters fear change and will defend what they've been using no matter how inferior it is. Hence, all the sarcasm we see here. If you guys would actually take a look at it (OpenStep is the basis for the incredible OS X development system), you'd realize it really does wipe the floor with GTK/QT.

Re:So, why all the jokes?

[zxSpectrum]

I'll never get Slashdotters' obsession with KDE and GNOME. Both projects absolutely suck. Their APIs are a joke. GNUStep/OpenStep, on the other hand, is a true object-oriented environment, and it really does make Gnome and KDE obsolete.

I'll never get someone's obsession to shove hideously ugly GUIs down users' throats. This looks like something I used when my Unix terminal was still using a black-and-white screen. This looks like something I used when my Unix desktop was totally useless.

Gnome and KDE can at least be made to look and feel nice. That is all I need to know. I don't care about any "elegant" solutions under the hood.

Re:So, why all the jokes?

[abigor]

Your post is correct, except for the bit about KDE. It really is a nice object system they have going there (not sure what you mean by a "true" object-oriented environment - are there false ones?), if you ever bother to sit down and try it out. Qt's signals/slots mechanism is intuitive and "magical". The main issues in KDE are usability (which is a big one), multimedia, and dealing with the X11 heritage, the final one being a problem faced by all Linux desktops in general.

Re:So, why all the jokes?

[Breakfast Pants]

The fact that is GPL instead of LGPL isn't necessarily bad, but it does ensure that Gnome will never go away.

Re:So, why all the jokes?

[PlacidPundit]

GNUstep is themeable and can be made to look like anything you want it to look like. But it's primarily focused toward developers. The development framework is where the interest is for me.

Re:So, why all the jokes?

[alienw]

Yeah, Gnustep is object oriented. So is every other GUI environment designed after 1989 or so, including GTK and Qt. Unless you are a fan of obscure niche languages like ObjectiveC, you'll probably like GTK and Qt a lot more. But yeah, Gnustep is great if you want to go back in time and experience the Linux desktop like it was in 1996.

And hey, you can yell "X11 sucks" on top of your lungs all you want. The problem is, until you actually _make_ something that's better, nobody will listen to you. And last I checked, Gnustep ran on top of X11 anyhow.

Re:So, why all the jokes?

[Anonymous Coward]

"Obscure"?

Apple has sold more Unix-based machines than any company in history. All based on Objective C.

Re:So, why all the jokes?

[alienw]

Apple has what, 5% of the market share for personal computers? Of the apps on OS X, probably 70% are written in something other than Obj-C. A language with a market share significantly less than 5% is by definition a niche language.

Re:So, why all the jokes?

[TheRaven64]

I'll never get someone's obsession to shove hideously ugly GUIs down users' throats

You mean like this[1]? Not only is it visually attractive, it's been carefully designed for usability as well - for example only controls that can be manipulated with the mouse have a gradient. The default theme for GNUstep is still the original NeXT look which, while dated in appearance, is very usable (although possibly because the developers put more effort into the feel than the look). Many of the GNUstep developers are NeXT refugees, and so the default theme is unlikely to change, although Étoilé - a GNUstep-based desktop environment - uses the Nesedah theme.

[1] The window borders in this screenshot are provided by WindowMaker, not GNUstep, and are not themed. A native GNUstep WM exists, but was not used here.

Re:So, why all the jokes?

[Overly Critical Guy]

Uh, people have already posted screenshots of the other themes available in this discussion.

Not only are you 100% wrong about the visual look, but you basically admit that all that matters to you as far as a desktop is its prettiness. No wonder Linux on the desktop never seems to happen.

Re:So, why all the jokes?

[Overly Critical Guy]

Obscure niche language? Objective-C was the basis for Java. It actually supports C fully, unlike C++, and it supports run-time dynamic typing.

How is Gnustep a backwards step in time? Because it doesn't have 20,000 sidebars and tabs like KDE? GTK and QT absolutely suck and feel like 1998 all over again. GTK and QT have nothing on Gnustep's object-orientedness. In fact, for you to actually claim the C-based GTK is "object-oriented" is hilarious.

You haven't even used Gnustep. You're just assuming things that you've read in other comments. A typical Slashbot.

Re:So, why all the jokes?

[Overly Critical Guy]

Uh, most OS X apps are written in Cocoa. If you're talking USAGE, then most are Carbon, but that's because those big apps are carryovers from the pre-OS X days, like Photoshop and Dreamweaver.

Market share is determined by units sold in a year. MARKET SHARE IS NOT USERBASE. It's important for me to scream it since users here seem to repeat the same falsehoods time after time. For what is supposed to be a "tech news" site visited by nerds, there sure is a lot of non-intellectual thinking every day.

Basically, you took a market share figure (OS X has 5%...though it has 15% installbase), decided that the market share had something to do with the userbase of a language (it's at least 15% going by your logic of OS X = Objective-C), and now you magically have a "niche language."

I love Slashbot logic.

Re:So, why all the jokes?

[Anonymous Coward]

Hey, dumbass, a language doesn't have a "market share." Market share is percentage of units sold in a year, not userbase. IDC says Macs make up 15% of the world's computers. Going by your crazy-assed logic, Objective-C is at least 15% of all used languages.

God, some people are so rabidly anti-Apple that they make themselves look like idiots in the process. Congrats.

P.S. Linux's share on the desktop is even smaller than Mac. Are GTK and QT niche APIs? Idiot.

Re:So, why all the jokes?

[alienw]

I'll ignore the childish insults. Let's look at open-source applications. On Freshmeat.net, there are approximately 39,000 projects listed. Out of them, about 7000 use C, 4000 use C++, another 4000 use Java, and a whopping 312 use Objective C. How is that not a niche language?

Re:So, why all the jokes?

[alienw]

A niche language is one which few people know. I looked at the stats for open-source projects (a good indicator of programming language popularity) on freshmeat.net. Guess what: Objective C accounts for less than 1% of those projects. Even Tcl has about 3x the number of projects. How is that not a niche language?

Re:So, why all the jokes?

[ucahg]

Very good, you've successfully proved that Objective C is used in less than 1% of all projects that are open source AND that are listed on Freshmeat AND that have correctly listed their language.

Get back to me after you take a statistics course.

Re:So, why all the jokes?

[alienw]

Obscure niche language? Objective-C was the basis for Java. It actually supports C fully, unlike C++, and it supports run-time dynamic typing.

Yeah, and Smalltalk was the basis for C++. How is Smalltalk not an obscure niche language? Besides, you could equally well argue that C++ was the basis for Java. Java borrowed heavily from existing languages.

How is Gnustep a backwards step in time?

It attempts to emulate a 12-year-old architecture, for one. Besides, have you seen the screenshots? They look like a UNIX workstation, circa 1993. More importantly, having many crappy desktop environments fractures the Linux platform way too much. I don't want to maintain 3 different desktop environments on my machine just to use a couple of apps written for each one.

GTK and QT absolutely suck and feel like 1998 all over again.

1998 is better than 1993. Besides, how are GTK or QT actually different from Gnustep? I mean, apart from not forcing you to adopt some weird language...

In fact, for you to actually claim the C-based GTK is "object-oriented" is hilarious.

You can write object-oriented code in any language. You do realize that C++ was initially a preprocessor for C, right? If you want to use C++, gtkmm does the job just fine and gives you an object-oriented API, C++-style.

Re:So, why all the jokes?

[Anonymous Coward]

> Uh, most OS X apps are written in Cocoa.

If you are counting crappy shareware apps with no users, then RealBASIC is a lot more popular than Cocoa.

Face it, even from Apple, very few OS X apps are written using Objective-C.

GNOME...not fun to play with?

[RLiegh]

Are you seriously going to tell us that Objective C/OpenStep IS?

Parallel Topic: Win32 targets

[hey!]

Not exactly on topic, but a parallel one. GnuStep is interesting, but sadly many of us can't afford to spend time researching GUI frameworks that only work on Unix. What we really need is something QT, GTK and WXWindows provide: access to the majority of GUI workstations in the world. I really like the idea of cross platform, but if I have to choose the most important desktop platform to my earning my daily bread, it's Windows.

I understand that I can download the Gnustep framework, tweak and recompile some programs so they place nice in windows; but even that level of effort is a deal breaker for me given all the other things I'd like to be looking into like WxPython. What would motivate me to do this is some actual applications that run acceptably in Windows, that would show that this is ready to play on the systems that create my paycheck.

Re:Parallel Topic: Win32 targets

[tarzeau]

Actually there is people doing just that. Check these: GNUstep and Gorm on Windows. See the screenshots of train data exchange software and train simulation software.

Crack Mods

[coolGuyZak]

It's not -1 Flamebait. +5 Funny FTW. ::shoot me now::

Lacking gorm

[dacarr]

Based on this little blurb, I think I shall remain Gorm-less.

Re:You're close

[OrangeTide]

Go to "Window Settings" then "Keyboard" then change pgup, pgdw, home and end behaviors. The default is to control the scroll region, people who use Vi don't mind since there are better keystrokes in Vi for moving the cursor around.

X11 apps shouldn't show up in the dock since they are not fully managed. But an icon box would have been nice. But it's trivial enough to run fvwm2.

Window selection for Terminal.app is a lot better than gnome though, cmd-1 to cmd-9 lets you quickly flip between terms.

Also I'm not sure how you can stand that GNOME and KDE junk on Linux. Xfce is much nicer. GNUStep is nice too. GNUStep's interface is not anything like MacOSX, it just has compatible programming API. But ultimately it all boils down to your personal preference. Use what you like and don't get angry if someone else has different preferences. (with comments like "Fuck Mac OS X." I wonder why you are so full of rage over something as unimportant as a person's choice of GUI)

Re:You're close

[Anonymous Coward]

> Window selection for Terminal.app is a lot better than gnome though, cmd-1
> cmd-9 lets you quickly flip between terms.

Note that alt-1 through alt-9 allows you to flip between tabs in any given GNOME terminal. Same goes for any app with tabs: GAIM, Epiphany, Firefox (on *NIX, anyway), etc.

WindowMaker

[KidSock]

This is good considering WindowMaker is going downhill in a hand basket. With each release they add things specifically to compete with features from GNOME and KDE. The new antialiasing font rendering looks so bad I had to downgrade to the previous version and add an exclude= to my yum.conf. I'll guess I'll be looking at AfterStep soon considering GNOME and KDE are Windows 95 but slower.

Re:WindowMaker

[Anonymous Coward]

I agree that WindowMaker's antialiasing isn't the greatest, but with the Bitstream Vera Sans font it's not too bad. And it's easy to disable antialiasing if you don't want it. I still think WindowMaker is the best window manager around.

I beg to differ

[biendamon]

The look and feel of the desktop are extremely important. They aren't the only factor a user should take into account, but to discount them completely is a mistake.

If the default desktop is an eyesore, keeps its menus and options in strange places, and has a lot of confusing buttons that don't explain what they do or what they're for, it doesn't matter how powerful the environment is.

Let me say it again: It doesn't matter how powerful the environment is.

Because most users will balk at the environment I've just described. Heck, *I'd* balk at it. I want a usable and intuitive interface as much as the next user. I would no more use this desktop than I would drive a car with strange and cryptic controls, no matter how powerful and efficient the engine is.

Re:I beg to differ

[Anonymous Coward]

So you agree that nothing about that screenshot says anything bad about the desktop environment in general? The screenshot is flat and simple, but it is not an eyesore.

Re:I beg to differ

[Archibald Buttle]

Graphics can be fairly easily changed. As has been pointed out elsewhere there's various skins for GNUStep that make it look quite pretty.

The default/original look and feel of GNUstep was closely modeled after NeXTstep. It's not pretty to our modern eyes, since we're now used to high-colour anti-aliased widgets all over the shop. We're not used to seeing a menu stack at the top left, and not used to seeing an essentially greyscale UI. Saying that GNUstep keeps its menus and options in strange places is only a reflection of what you're used to. Why is it less valid to put the main menu of an application in a stack at the top left than across the top of the screen, or the top of a window? Had you no conditioning about where a menu should be it wouldn't be so confusing. It's actually incredibly easy to get used to.

(Cue the "why do I have to get used to it" type whining here. Blah. I use Macs, Windows PCs, and a few varieties of Linux regularly - I have never had any problems adapting to the idiosyncracies of the various GUIs I come across. Yes, my favourite is the prettiest, which also happens to be the most powerful IMHO.)

I'm not discounting the importance of the appearance of a GUI completely, however you have to bear in mind that appearance is mostly a marketting issue, not a technology one.

Re:You're close

[Tomchu]

Sounds like someone is feeling insecure about their poor operating system choice. Is OS X making you/Linux feel in adequate? That's a shame.

Hey GORM...

[LarryTheGeek]

1994 called, they'd like their user interface back. Seriously though, that's butt ugly.

vs glade?

[Compenguin]

So how does Gorm differ from glade/libglade besides being for GnuSTEP instead of GNOME?

Re:vs glade?

[stivi]

Well ... not exact explanation, but you can find some answers to your question here and here.

The Window looks familiar...

[krewemaynard]

...not totally unlike XFCE4.

Much nicer screenshot, too. The icons also look great. Thanks for posting them. I was afraid that slashdot editors and the Gorm folks were trying to resurrect 1995 after I looked at that old screenshot...yeesh.

Given Turing equivalence

[Julian Morrison]

...it's still possible to define a useful meaning for "what you can do in language X can't be done in language Y". Namely: the fastest way to implement the features of X in Y is to use Y to write an emulation of X.

Thus it it's fair to say eg: "you can't do Haskell stuff in C, without rewriting a Haskell interpreter in C".

The brick and "mortar" icon on their site

[Cleetus Freem]

Ugh. The "mortar" on that icon makes it look as if Peter North mistook it for one of his co-stars.

DnD

[spitzak]

These problems could be addressed in Linux/X11. The way to do it is to redefine middle-mouse-click as "drop the most recent selection". This just happens to match what xterm/etc do and that many people mistakenly call the cut & paste X function. Because of this X programs are much more likely to be able to get this enhancement because they tend not to have use the middle mouse for anything else.

Because you can rearrange, open/close, and otherwise manipulate windows between the selection and the "drop", this addresses all your concerns about DnD.

What I suggest is that every toolkit or system that implements drag & drop be written so that middle-mouse-click acts exactly as though you dragged from the most recently selected thing and dropped it. For text this is not very hard because it would use the "PRIMARY" selection. For other data this may require programs that select items that can be dragged to also update the primary selection.

I'm still on fvwm and tkdesk

[srobert]

I'm still using fvwm with tkdesk. (Old machine). They work just fine for my apps, Thunderbird, Firefox, OpenOffice, xterm with Bitchx, Links or FTP, running in it, and GnuCash.
Well I'm off to saddle up my dinosaur.

Re:Since most people seem to have missed the joke.

[idiotnot]

[i]You GNUStep users out there: How do you configure your desktop? I've noticed that GWorkspace and Window Maker like to compete for how they manage virtual desktops.

Also, I'm not sure what application I should be using as a dock? Window Maker? Something written in GNUStep?[/i]

I've taken lately to disabling the WM dock and clip, and letting GWorkspace/GDesktop handle it all. Some instability, still, with GDesktop, but it works pretty well most of the time.

(the command line switches I don't remember offhand....they're in the wmaker man page)

Openstep?

[MemoryDragon]

Well the UI is nice but rather minimalistic, even NeXT has moved on, they have taken over Apple :-)

Re:Openstep?

[tarzeau]

And they have lost some important things: see here for some features. Another thing that I find annoying, stuff are slow, even on a G5 dual cpu machine. I run OPENSTEP on a Pentium 233 mhz machine, with 64mb memory, and it's rock solid, and damn fast. Everything, see here for some nice apps.

Turing Complete...

[irritating environme]

Ever heard of it people? Probably not. Sigh.

BS: Re:Graphical Object Relationship Modeller

[Anonymous Coward]

Everything that can be programmed can be written in assembler

I call this BS ! Everything that can be programmed can be written by switching single bits in memory ! That's the way Real Programmers do it !

Windows -- Bizarro Edition?

[Ahnteis]

I use drag and drop a lot when using Mac OS X simply because it works so well. Dragging files to the trash, dragging files to applications to open them, dragging images off web pages to save them etc..

Just because Windows (and therefore Linux, as sadly the linux desktops have heavily copied windows as opposed to OS X) can't do drag and drop effectively

Odd, every one of the examples you mentioned works in MY Windows. Have you screwed yours up somehow or are you using some bizarre version? Perhaps you meant to bring up some of the actual problems with windows drag and drop (there are one or two, but it's hardly "broken")?

Re:Windows -- Bizarro Edition?

[2nd Post!]

Try dragging a file from the Explorer to a minimized application, like a Word document or something.

You can't, because Windows has to first restore it, and then you have to drag the file to the newly restored application window.

Can you drag links from a webpage to a browser window?

Re:Windows -- Bizarro Edition?

[Anonymous Writer]

Odd, every one of the examples you mentioned works in MY Windows.

I don't get why the previous poster used those examples because they're the same function-wise, but there is a difference in the GUI implementations. The thing about the Windows drag and drop I remember (it's been a while) is that many times it simply changes the cursor to indicate you're dragging something. For example, when you're dragging a picture off a web page onto the desktop, or when you are dragging highlighted text. In OS X, you actually have a transparent image of what you are dragging, like the picture or highlighted text. I actually think this has more to do with how Microsoft would integrate features from other software producers, but simply change the look and feel to make sure it doesn't infringe on copyright, rather than lack of functionality. Like I said, it's been a while so I'm not sure if they have changed this.

BS. And again, BS.

[l3v1]

release of Gorm 1.0, the Interface Builder for the GNUstep project, and with its release, comes the obsolesence of the GNOME and KDE projects

First, please educate yourself on kdevelop3. Then, you can come around and talk about superiority of something else, but then again, talk won't be enough, prove.

Let's cut through some of the crud here...

[Art Tatum]

First, the way the interface looks is irrelevant. A GNUstep theme engine is available here. There's a nice theme in progress called Nesedah (mockup and screenshot of IRC client along with OS X comparison shot)

Second, why is this such a big deal? Don't QT, Visual Basic, and Delphi provide the same RAD approach? No. I've used all of those tools and they just don't stack up. QT is about as good as you're going to get out of a static compile-time-oriented C++ approach. But it's not as simple or direct as a runtime-oriented OO solution like Smalltalk or Objective-C. This is the power of Cocoa/OpenStep/GNUstep.

Delphi, .NET, and QT GUI designers focus on generating code. This is cumbersome and brittle. But Apple/NeXT's Interface Builder and GNUstep's Gorm take a different approach. They actually instantiate objects, set state, create inter-object connnections, and then persist the in-memory objects to disk. When your application is loaded, these objects are unarchived and your application connects to them. This prevents the OO-mocking approach of subclassing a Window class just to create your own instance--something that always makes me laugh but is ubiquitous in the Windows world and has been blindly copied by KDE and GNOME.

Finally, the poster is not a native speaker of English and clearly was not able to convey the sense of humor intended.

Re:Let's cut through some of the crud here...

[Arandir]

This prevents the OO-mocking approach of subclassing a Window class just to create your own instance--something that always makes me laugh but is ubiquitous in the Windows world and has been blindly copied by KDE and GNOME.

Okay, your object persistance model of development is interesting. But your last point is dead wrong. You don't have to subclass a class to instantiate it. That's stupid. No wonder you hate C++ if you think you have to do that.

I don't have to subclass a Button class to create a button. I merely declare one. Just like <gasp> Smalltalk and <gasp> Objective C, I only need to subclass when I need to extend the base class in certain ways.

Re:Let's cut through some of the crud here...

[Art Tatum]

But your last point is dead wrong. You don't have to subclass a class to instantiate it. That's stupid. No wonder you hate C++ if you think you have to do that.

First, I don't hate C++. It's great, particularly for mathematics and data structures. But QT Designer, Delphi, and the .NET tools I've used all subclass the Form class for each form you create. Then, you add control objects as "new" members of this class. There's really no need for this in most cases, since you rarely add new functionality. And object composition, as you suggest, would be a much better way to go. I'll never know why this has become such a common way of doing it.

Re:Let's cut through some of the crud here...

[Arandir]

Speaking from Qt Designer experience. The "Form" class needs to be subclassed, because you are EXTENDING it. You ARE adding functionality. You can't make a calculator app without at least thirteen buttons and a view window. You could make a generic collection class widget, but *somewhere* something has to deal with those fifteen collected widgets. Something has to deal with the events/signals from the buttons and something has to deal with updating the view. Something has to connect the buttons to the viewport. I don't know what handles this stuff in GORM, but something MUST be handling it. Why not simply do it in the collecting widget? Thus, the need to extend the Form class.

This is not an onerous way of doing things. It's quite simple actually. In fact, it's trivial. And lo and behold! The Form class the only damned thing you need to subclass!

Re:Let's cut through some of the crud here...

[Art Tatum]

Speaking from Qt Designer experience. The "Form" class needs to be subclassed, because you are EXTENDING it. You ARE adding functionality. You can't make a calculator app without at least thirteen buttons and a view window. You could make a generic collection class widget, but *somewhere* something has to deal with those fifteen collected widgets. Something has to deal with the events/signals from the buttons and something has to deal with updating the view. Something has to connect the buttons to the viewport. I don't know what handles this stuff in GORM, but something MUST be handling it.

In GNUstep/Cocoa, you have a Controller that connects to your object archive and talks to the window and controls. Also, the window is connected to the controls and subviews through a Responder chain. It's a more "correct" way of doing things, from the perspective of object-oriented design and MVC because you really shouldn't be extending your Form class (or Window class, taxonomy depends on the framework <grin>) unless you're actually going to change the fundamentals of the way windows work.

I should say, however, that QT is miles ahead of Delphi, and maybe even .NET. If I were going to develop applications with this approach, QT would probably be the way to go.

Re:Let's cut through some of the crud here...

[Arandir]

It's a more "correct" way of doing things

I'm sorry, I must have missed that announcement. I guess if you're going to saddle yourself with a runtime environment, you might as well have the runtime do all the work. But since C++ doesn't have a runtime environment I have to make the application do the work of the application.

Instead of using Widgets (or Forms or Windows), a framework could use a Container instead. In Qt terms, it could be a layout manager with a pure generic list of widgets residing in it. You would never need to subclass anything, just "drop" your widgets in the container. But what's the point? What practical benefit do you gain?

When I compare the "correctness" of GNUstep to the "practicality" of KDE, I'm going to put my money on the practicality.

Re:Let's cut through some of the crud here...

[Art Tatum]

When I compare the "correctness" of GNUstep to the "practicality" of KDE, I'm going to put my money on the practicality.

Well, far be it from me to tell you what to use, but I have found the OpenStep way is faster and more easily maintainable because of these decisions.

OMGWTF HOW DAR U!!!111

[BumpyCarrot]

lololololol "beter than GNOME" yeh woteva!!!11

[/n00b]

Re:You're close

[Anonymous Coward]

Sounds like someone is feeling insecure about their poor operating system choice. Is GNU/Linux making you/Mac OS X feel in adequate? That's a shame.

Re:You're close

[gentlemen_loser]

Fuck Mac OS X. I have a Mac and still prefer Gnome or KDE (I have used both extensively and like both). OS X is a sorry excuse for a UNIX.
Good - don't use it then. Sell your Mac on ebay and buy a new Linux box. However, while you are trolling - go take a long look at modern Linux desktops, then go look at OS X. You'll probably be supprised to notice that alot of the "good" functionality in KDE and GNOME was swiped directly from OS X.

It doesn't even have a working native terminal program (one which allows PGUP/PGDN), and Fink's repository is very poorly maintained. Running X apps doesn't work all that great (why don't they show up in the Dock?). Maybe with a huge amount of effort (and buying a few shareware apps), I could get OS X to work how I want it.
That argument is the EXACT argument used by Windows users against Linux in the early days. Why would you do it? To perhaps show your support to a commerical company that IS actually innovating. Oh, and they don't show up in the dock because they were written using a different API and are connecting to a different window SERVER!

But why would I bother when I can have Linux working out-of-box the way I want it? Why would a want a non-free OS when a free one works just as good if not better?
Again - show some support for a company that is innovating.

There are a few apps that are really great on OS X, but that is all I use it for. Linux is my primary OS. Just because all the pussy, wannabe users fled to OS X, doesn't mean that OS X is better. It's better than Windows, but that's about all that can be said for OS X.
I'll call it like I see it here. You must be an l33t h@x0r right??? Go back to high school and do your homework. OS X has a unique and interesting background in NextStep - which is what both AfterStep and Windowmaker (two of my favorite window managers) are open source clones of. I find it ammusing that someone with your mentality is extolling the virtues of Linux and in the same breath praising KDE and Gnome rather than a more lightweight window manager. Your views are an uninformed mishmash of "popular" thoughts without ONE of your own.

This is special. I WISH I had modpoints today. Can someone tell me why my post (see above) is modded as flamebait while this parent is modded as insightful? PLEASE - read the two posts and tell me why. I'd love to know.

API Compatibility

[RAMMS EIN]

FYI: GNUStep and Mac OS X both implement the OpenStep API, and GNUStep are implementing more and more of OS X's extensions. There is a large degree of API compatibility between the systems, and I've personally written software that ran on both systems, without changes to the source code.

As another poster pointed out, there is a theme engine, and it can be used to make GNUStep look very slick indeed.

Having said that, I do agree with you that the project has had a tendency to overhype things. Things are coming along nicely, though, and I could see GNUStep becoming a viable and popular desktop.

Re:You're close

[justsomebody]

Window selection for Terminal.app is a lot better than gnome though, cmd-1 to cmd-9 lets you quickly flip between terms.

Well, that's a new one. Someone saying Terminal.app is better than anything. Terminal.app is the... worst... terminal... emulation... ever...
cmd+1..9 would mean anything if terminals would be as tabs and one could actualy know which is 1 and so on. Personaly, on Linux I always open at least five terminals in one fullscreen window, each for its basic task and I can always select whichever I need (alt+1..9 or Ctrl+PgUp/Dn or mouse).

Also I'm not sure how you can stand that GNOME and KDE junk on Linux.

Don't know for original poster, but I can tell you much easier than with OSX (and yes I do have G5 with Tiger, sucks just as previous with Panther).

GNUstep licensing question?

[argent]

Question...

It looks like GNUstep's dependent on some GPL libraries. Would this make apps linked for GNUstep GPL, or would the fact that GNUstep's implementing a non-GPL interface (NeXT/OpenStep or Cocoa) change that?

If GNUstep is to replace KDE or Gnome (which I would dearly love to see happen) the latter interpretation would need to be true.

Re:GNUstep licensing question?

[Anonymous Coward]

GNUstep is released under the LGPL ( see http://www.gnu.org/copyleft/lesser.html ), that means it has no effect on the applications based on it.

regards, Lars

Re:GNUstep licensing question?

[borgheron]

The GNUstep libraries are LGPL. So, no, it wouldn't make your app GPL to link with them.

Re:GNUstep licensing question?

[argent]

I thought it was dependent on GPL libraries, which would make anything linked with it GPL, but it seems that the libraries that I'm thinking of are LGPL now.

<emilylatella>nevermind</>

Mod Parent Up

[mad.frog]

Parent is AC, but makes excellent points.

PLZ port GWorkspace 2 OSX kthxbye!

[argent]

No, really...

I'm SO goddamn tired of Finder and I SO miss NeXT's file manager...

Re:You're close

[OrangeTide]

I use alt-1 to alt-9 to flip windows in my text editor and text irc client. I'd rather use a key that isn't used over telnet/ssh.

Also multiplexing using 'screen' in an xterm seems to have more advantages than people using tabs in GNOME Terminal.

Re:You're close

[user9918277462]

You'll probably be supprised to notice that alot of the "good" functionality in KDE ...

Riiight. 'Nuff said.

With looks like that? Really?

[tux0010]

Looks like windowmaker to me. With looks like that, how do they expect people who are into eye-candy (and these are the people who choose gnome/kde anyway) to like it?

Stable API on Mac OS X 10.4

[mbessey]

The message has been a bit unclear, and I think some people are misinterpreting what they've been told. What I remember from WWDC is:

1. The Kernel API's (KPI, whatever) are now fixed, and you shouldn't need to revise your drivers for future releases of the OS. This was explicitly NOT guaranteed in previous versions of Mac OS X.

2. At one or the other of the Cocoa sessions, there was a kind of vague statement that could have been interpreted as "the core Cocoa classes are not going to change in incompatible ways, going forward". The thrust of that was, once again, that you should not need to revise your application code to remain compatible with future Mac OS versions. In a practical sense, I guess this would mean APIs aren't being deprecated anymore.

Obviously, Apple is still going to be implementing new functionality (exposed via new APIs), and they'll fix bugs, which may break some apps. But there was a definite commitment to improved backward compatibility.

-Mark

Re:Stable API on Mac OS X 10.4

[SteeldrivingJon]

"Obviously, Apple is still going to be implementing new functionality (exposed via new APIs), and they'll fix bugs, which may break some apps. But there was a definite commitment to improved backward compatibility."

Actually, it might have been an effort to reassure developers who are already worried
about the PPC -> Intel port.

If the APIs are stable, then developers don't have to worry about
handling the port, supporting 2 architectures, *and* dealing with
some future API shift which could be a major fork between
PPC and Intel software.

Not that simple. GNUStep vs. KDE,GNOME,E17

[CarpetShark]

Well, actually, they all do very similar things, and provide similar tools. The differences would be in technologies supported, and in ease of programming. From what I've gathered, GNUStep is very object-oriented, which means that it will be easy to develop applications quickly for, without them growing so confusing as they get bigger. Qt and KDE are both similarly object-oriented, and I think it shows in how quickly they develop and evolve. I know of one game, oolite (an Elite clone), which was quickly ported from OS X to Linux using GNUStep. It did still have issues though, so it wasn't a simple re-compile. Doing that with GTK would be much harder, though, and I don't think of GTK as Object-Oriented at all. It does have extensions to support OOP, but IMHO, that should have been a major design principle, right from the beginning. I think GTK would have developed faster, and been more usable and integrated, if it was. I'm not sure if GNUStep fully supports m18n, but GNOME and KDE do, and probably E17 too. E17 is the only one to currently support ACLs and EAs in a usable way.

Re:Not that simple. GNUStep vs. KDE,GNOME,E17

[Archibald Buttle]

GNUstep has a fairly complete implementation of CoreFoundation which includes a whole host of m18n support functionality.

Oolite had a few problems recompiling on Linux (and Windows) with GNUstep because it used a few features of the Mac OS X Cocoa libraries that haven't been completely implemented in GNUstep, especially in relation to OpenGL integration.

Re:Not that simple. GNUStep vs. KDE,GNOME,E17

[CarpetShark]

Ahh, that's good to know. Thanks. I wonder if oolite folks fix GNUStep to port it then, or if they just hacked their app...

Doesn't anyone else Drag - Alt + Tab - Drop?

[mikefe]

Am I the only one who drags from one full (or mostly full screen) window -- then press alt + tab to change to the desired target windows -- and then drop?

I do this all of the time on windows to attach files to an email and can't do in gnome 2.10. :(