Looking Ahead at GNOME 2

Able writes "This is a good article that will teach you how to use the new and improved libraries available with GNOME 2 so that you can write your own Nautilus view, and panel applets. It also provides you with the understanding to compile a few sample GTK+ 2 programs that will give you a good understanding of GTK+ 2's many improvements over GTK+ 1."

Comments

IBM

[brad2600]

anyone think it cool that this is hosted on the ibm site?

.brad

Re:IBM

[kitts]

anyone think it cool that this is hosted on the ibm site?

They're probably hoping this sort of thing will help lead to a return on their investment.

Higher numbers faster please..

[PopeAlien]

..Geez.. 2.0? Thats no good.. I mean look at windows for instance - Its way past 3.0, past 98.0.. I think its even past version 2000 now! How is linux ever going to competitive with such small version numbers?

Re:Higher numbers faster please..

[jasonv118]

Actually, it's eXtreme Programming. Seriously. And why did someone classify the version number guy as a troll, that should have been a funny.

Re:Higher numbers faster please..

[qweqwe]

No XP stands for the greek letters X (chi) and P (rho). In other words, Cairo.

Re:Higher numbers faster please..

[Antity]

Things like Linux or GNOME don't have to competite against "high" version numbers, since things like "98", "ME", "2000", "XP" aren't "version numbers". The real version number of windows 9x is still 4.x, the one of 2000 5.x, AFAIR.

Of course, this doesn't prevent RedHat/S.u.S.E. from using their own 7.x version numbers at the moment so they can competite eachother.. :-/

Re:Higher numbers faster please..

[Guy Harris]

No XP stands for the greek letters X (chi) and P (rho). In other words, Cairo.

Or "Christos", as in "Jesus wants you to run this operating system", or perhaps "Jesus Christ, what is this thing?"

Re:Higher numbers faster please..

[Guy Harris]

The real version number of windows 9x is still 4.x,

98 and Me aren't 5.x? I'm curious what GetVersion() and GetVersionEx() return on those platforms. (I don't have any Windows OT machines handy, although I suppose I'll eventually manage to get it running on VMWare at home - I've no desire to run Windows OT, of any version, for real; the Windows partition on my home machine is running NT 4.0.)

the one of 2000 5.x

W2K on my work desktop machine claims to be version "5.00.2195", i.e. 5.0 build 2195.

I think XP is really NT 5.1.

Of course, this doesn't prevent RedHat/S.u.S.E. from using their own 7.x version numbers at the moment

Hmm. At times it'd be nice if there were a Linux API to get the name and version number of the distribution, rather than that of the kernel, if for no other reason than to let you more easily or more automatically get that information from users when reporting bugs.

Re:Higher numbers faster please..

[Antity]

At times it'd be nice if there were a Linux API to get the name and version number of the distribution, rather than that of the kernel, if for no other reason than to let you more easily or more automatically get that information from users when reporting bugs.

Mostly useless. The only thing you could get would be the distribution release no. of the initial installation, maybe together with a counter about how many packages _might_ have been upgraded/changed (don't think about those geeks who still rpm -e (or their counterparts) and update from source.. eek!).

Things like "Uses FileSystem Standard x.y" and "BSDish/SysVish rc-s" according with libc version and gcc STL and binary API would be more interesting.

It's still ill to see how many packages are still offered to be "for" RedHat 7.0, 7.1, or this crap SuSE fs standard.

./config.guess now! :-}

Much better, but still behind KDE

[jasonv118]

Seems a lot nicer, but it's still behind KDE. I like GTK alot more than Qt though. I wish GTK/Gnome would get transparency (I think KDE is, right?) Finally we got anti-aliased text, (did that take long enough??) Will Abiword have anti-aliased text now?

Re:Much better, but still behind KDE

[hexix]

No, the transparent menu effect that Mosfet made for his liquid theme (and will now get added to kde 3.0 builtin themes I think) takes a snapshot of the area behind the menu and uses that image to make a pseudo-translucent effect.

Not a very good way of doing it in my opinion, and this becomes clear when you move through the different menus in a menubar, you can see the ghosts of the menu you had previously opened beneath your current menu.

If there is a better way of doing it, I'm not sure what it is, although I think you can do true translucent effects with Keith Packard's XRender extension.

But it doesn't seem that anyone is using that to do translucency so either its really hard to use or theres problems. I'd love to know which it is if anyone has some more information.

PS - No offense to Mosfet or anyone else using this code to do translucent menus, it's a very good idea and I do use it when I'm in KDE. I just think it'd be cool to have it without the little quirks like seeing things behind the menu that shouldn't be there.

Re:Much better, but still behind KDE

[Shanep]

Have you used Star Office 6 beta?

Awesome, aa fonts and all. It may as well be Word, Excel and Powerpoint for Unix.

Re:Much better, but still behind KDE

[ichimunki]

I think the key problem GNOME has is that there is no native web browser that isn't a complete monster. In my brief usability tests comparing Mozilla with Konqueror I concluded that Konq is lightyears ahead of Mozilla in terms of usability, customization, ease to compile, and integration with a desktop environment (note that this last is also a potential drawback since it depends on certain KDE i/o slaves to do its work-- and yes, it can run under any wm or destkop, but will be most efficient inside KDE).

I see the following two main advantages to gtk over Qt (and this is why I will not use Qt except as a user): gtk is written for C development. C is much more standard on Unix/GNU/Linux than C++. Perl is written in C, which means that adding wrappers for the gtk library is likely to be less problematic than for some other toolkits (although Perl has it's own excellent version of tk... this advantage is likely to extend to other languages written in C, Ruby being foremost in my mind).

gtk has the same LPGL terms for every instance of the toolkit. This means that no matter what type of development you are doing that you never have to worry about licensing, at least not the way you might with Qt.

It is too bad, however, that on top of these two fairly decent widget toolkits the desktop environments are vastly different. GNOME just ain't as easy to get up and running with as KDE-- and I mean in the "compile from source and install" sense. I've also noticed distinct efficiency differences between the two on low-end hardware (KDE coming out ahead).

What we're left with is a mess, but it's a good mess. We have real powers like IBM supporting gtk/GNOME, and we have other serious groups like the German government supporting Qt/KDE (witness the Aegypten project for secure communications). We get the best of both worlds, and fortunately we can do some mixing and matching as needed. So either way users win and developers seem to have two decent high-quality choices to choose from (forgetting about the ubiquitous Tk for a moment).

Re:Much better, but still behind KDE

[ZaMoose]

Galeon, baby. Galeon. Latest version is mui, mui cool (and fast as all get out!)

Re:Much better, but still behind KDE

[ichimunki]

Galeon depends on Mozilla to build. That is a serious problem, imho-- since it requires the building of the monster in the first place. And yes, I realize that even some geeks these days don't go to the extreme measure of compiling as much of their systems as possible from scratch. *grin* Maybe Galeon developers can figure out a way to lift the relevant shared library code from Mozilla so we can build Galeon without Mozilla.

I have liked Galeon when I've used it, but I didn't try it for some of my more important tasks that Konq handles quite well-- things like page-specific decisions on cookies and scripting, lying to online financial institutions about the user agent so that I can access my accounts, etc. Can it handle this stuff? And does it have tabs?

Re:Much better, but still behind KDE

[Menthos]

Galeon depends on Mozilla to build. That is a serious problem, imho-- since it requires the building of the monster in the first place. And yes, I realize that even some geeks these days don't go to the extreme measure of compiling as much of their systems as possible from scratch. *grin*

Nah, myself I always grab the fresh RPM:s from ftp.mozilla.org. :-)

Maybe Galeon developers can figure out a way to lift the relevant shared library code from Mozilla so we can build Galeon without Mozilla.

This really has to be done by Mozilla developers. I think Chris Blizzard has promised that he will work on splitting gtkmozembed libraries into a seperate RPM package, for example.

I have liked Galeon when I've used it, but I didn't try it for some of my more important tasks that Konq handles quite well--

That's a shame, you should have. :-)
I have heard many people that like Galeon better than Konqueror, but that's of course a matter of taste. It certainly is a worthy competitor to Konqueror.

things like page-specific decisions on cookies and scripting,

Cookies yes. Scripting is a global setting but easily accessible

lying to online financial institutions about the user agent so that I can access my accounts, etc.

Yes. See the Galeon FAQ.

Can it handle this stuff? And does it have tabs?

Certainly. What did you expect? :-)

Re:Much better, but still behind KDE

[Tralfamadorian]

I used to build parts of my system from scratch, and Mozilla did take a long time to build, but then I learned the convenience of binary builds... still, I wonder how long it would take to compile mozilla on my Athlon XP 1600 :)

Galeon can do cookies and images on a per-site basis, though I don't know about using a false user agent string. WRT tabs, Galeon was the first browser I saw to have tabs (though that doesn't necessarily mean that it was the first, just the first that I saw :))

Galeon is by *far* my favorite browser on any platform.

Re:Much better, but still behind KDE

[ichimunki]

Awesome post-- especially the link to the FAQ. I will definitely keep trying to compile gnome-libs and gnome-base. Right now I'm stuck on some BS in gnome-libs where it's telling me I don't have the right db compatibility option available even though I know full well I do because I just compiled BerkeleyDB with that exact option enabled. Back to the terminal window I guess. :)

Re:Much better, but still behind KDE

[Burgundy Advocate]

Longer than that. Remember Eterm?

Re:Much better, but still behind KDE

[Burgundy Advocate]

It's more truer transparency.

Bull. It's the exact same idea. Eterm just takes a snapshot of the desktop, whereas KDE takes a snapshot of the app.

Re:maybe GTK is not the correct choice for the fut

[jasonv118]

Aaaah, so you're one of those old UNIX guys who refuses to leave his beloved Motif... when I put Linux on my brother's computer he refused to use Netscape because it is so ugly.

Worried Gnome User.....

[CDWert]

I am a Gnome user, and athough I am NOT a sky is falling person, KDE seems to be making much more usefull strides, I am also concerned about the Ximian fork, (even though I use it) How long till XImian hack up all the libs to work for their effort and how compatible will it be ?

I have thought about switching to KDE for no other reason than they seem to have a much better, much more focused direction.

Does it seem to anyone else latley Gnome is becoming a throw in everything and if the kitchen sink dosent work its OK, or is it just me.

Admittedly Gnome 2 has some nice stuff but how much will be functional by first release ?

Re:Worried Gnome User.....

[dagoalieman]

KDE makes big strides, I'll grant you.

For me, even more interesting, but quietly mentioned, was the Accessibility Tool Kit (ATK). If you start pushing Linux out there to handicapped people, OLDER people, who need help hearing, seeing, what have you, they can help provide a major market push for linux. Obviously the needs of these people vary greatly, so customized solutions are a must.

In the same vein, it seems that KDE is for the people who want the solutions given to them, and Gnome is for the people who want to build their own customized solution (IE what they want.) That explains a little bit of the difference in the attitudes. Capt. Obvious does point out to me that the "build your own solution" approach is a general platform for Linux, KDE included, but I think Gnome really goes after that more.

I hope most of it will be functional, but like you, I fear that we're going to see a whole lot more of "build your own" than in the past from Gnome. It's not just you...

Re:Worried Gnome User.....

[ed__]

Ximian isn't a fork of gnome. They just package it for the end user.

Re:Worried Gnome User.....

[VFVTHUNTER]

I am a Gnome user, and athough I am NOT a sky is falling person, KDE seems to be making much more usefull strides, I am also concerned about the Ximian fork, (even though I use it) How long till XImian hack up all the libs to work for their effort and how compatible will it be ?

Ximian is still open source. When THAT changes, worry. For now, don't.

Their stuff is pretty compatible, tho to get things like Evolution working you have to have the latest versions of some things (gal, gtkhtml, I think bonobo)...

I have thought about switching to KDE for no other reason than they seem to have a much better, much more focused direction.

I use both Gnome and KDE. The only thing that irritates me about the holy war between them is that one would expect that, given all the hype, someone from one camp would have made that camp compatible with the other camp as a sort of reverse-psychology love-flame. They don't work perfectly together, but I digress.

Gnome seems faster. It also does not have the annoying habit of attaching a blinking app icon to my mouse cursor when an app is doing something that takes a long time...but Gnome apps don't to seem to have the lag that the KDE ones do. KPoker roxxx tho :)

Does it seem to anyone else latley Gnome is becoming a throw in everything and if the kitchen sink dosent work its OK, or is it just me.

It's just you.

Admittedly Gnome 2 has some nice stuff but how much will be functional by first release ?

The idea of a first release is that the code is stable and fully functional.

Re:Worried Gnome User.....

[hexix]

Well it sounds to me like you're creating your own worries. I have trouble understanding why you'd be a "worried gnome user" because something else might become or is better. Why not just switch then instead of worrying?

With that said, I don't think you should worry about Ximian, much of their work gets added into the vanilla gnome.

My personal opinion of GNOME is that a lot of work is going into two sections right now: great applications (evolution, gnumeric, galeon, abiword), and whole new libraries for gnome 2.0. I think once gnome 2.0 stabilizes many people will be eager to take advantage of the new features and you'll see the desktop itself get many cool new features.

This point is brought up constantly but people seem to prefer to ignore it. When KDE 2 was being worked on many critics were saying KDE must be dying because they weren't seeing the work. You need to understand that such big code changes happen a lot more smoothly without non-programmers trying to use it and complaining about this and that or sending questions on how to get it working. You're just going to have to be patient, or you could just use KDE2 for a while.

Re:Worried Gnome User.....

[Tyndareos]

I am a Gnome user, and athough I am NOT a sky is falling person, KDE seems to be making much more usefull strides, I am also concerned about the Ximian fork, (even though I use it) How long till XImian hack up all the libs to work for their effort and how compatible will it be ?

I have thought about switching to KDE for no other reason than they seem to have a much better, much more focused direction.

I am what you could call a KDE user, ethousiastically even developing my own KDE app, but still I don't feel that Gnome is losing out to KDE. KDE might seem to have some advantage at this time, but Gnome is certainly strong in the application department. Off course I can't be sure, but I definitely hope that Gnome 2.0 will rock and maybe recapture a bit of the current kde users. If there's one thing that helps to get motivated then it's solid competition and I don't like Microsoft very much in this role of 'evil' competition. I'd rather see two projects with the same idealogy compete and I think that most KDE developers certainly wouldn't want Gnome 'out of the way' either.

Okay I'm just rambling now ... so I'll quit.

Re:Worried Gnome User.....

[Skeezix]

I am a Gnome user, and athough I am NOT a sky is falling person, KDE seems to be making much more usefull strides

Which strides is KDE making that are more useful than the ones GNOME is making? I'm curious.

I am also concerned about the Ximian fork, (even though I use it) How long till XImian hack up all the libs to work for their effort and how compatible will it be ?

Ximian does not produce a "fork" of GNOME. Ximian packages a "distribution" of GNOME and makes it easy to download. They tweak some minor things such as artwork, splash screens, etc, but it's not a fork of GNOME. I don't think you understand Ximian's relationship to GNOME. I suggest you spend some time on irc.gnome.org in #gnome and spend some time getting to know folks better.

Does it seem to anyone else latley Gnome is becoming a throw in everything and if the kitchen sink dosent work its OK, or is it just me.

That is not at all how it works. We're very particular about what we put in the release. I suggest you spend some time reading the archives of mailing lists such as desktop-devel. Much work has gone into making GNOME 2 more usable, accessible, functional, and a better development platform while keeping it solid.

Admittedly Gnome 2 has some nice stuff but how much will be functional by first release

We won't release if it's not functional :)

-jamin

Re:Worried Gnome User.....

[rasjani]

• I am also concerned about the Ximian fork, (even though I use it) How long till XImian hack up all the libs to work for their effort and how compatible will it be ?

I've understood that actually, ximian gnome is not a fork when it comes to the code, its just packaging and compilation of different versions of programs thats supposed to work well together. I might be wrong thou.

Basicly ximian works on many OSS projects and actually contributes back to them, not just making "their own fork" ... For example, read release notes of of Gnumeric, quite a few ximian dudes are warmly thanked there.

Anyway, even thou the codebase is same, that doesnt mean all gnome distributions will work with each other. So, if you are using ximian, will stick with it or change it totally if the times comes ...

(and yeah, i dont claim these are the absolute facts)

Re:Worried Gnome User.....

[Skeezix]

I do understand what a fork is. Ximian GNOME uses the same source for its builds as you'd use if you grabbed the tarballs. That is not a fork. The plop in some new images and splash screens and give a nice installer. Any changes they make to the source code go back to gnome cvs. It's not a fork. It's a distribution of GNOME.

Re:Worried Gnome User.....

[Skeezix]

You don't have to purchase it and it's not instead of a Linux distribution, it's in addition to. The advantages include:

- a better tested set of packages. Ximian makes sure they all work together. If you download updates and build them on your own, well...you're on your own.

- Automatic dependancy resolution with Red Carpet.

- A cross-platform GNOME distribution that is consistent. Red Hat, Mandrake, Debian, etc. all package GNOME slightly differently, include different artwork, include different versions of the software and update at different times. Ximian provides one distribution of GNOME across something like 11 different Linux/UNIX platforms.

Re:Worried Gnome User.....

[Lumpy]

I switched to KDE recently.. mostly because of nautlis. It's forced upon you and if you follow their instructions to remove it, it dont work... nautlis still loads as the "desktop manager" and still steals about 30% of the entire computer's performance. KDE is a bit snappier, If you set it to the "crappy computer mode" it's dang fast and I'm happy with it. KDE though fails miserably at the install and upgrade level. as does Gnome. Ximian is the ONLY desktop system that has a decent install system and the others need to stop all development right now and work only on an install system to catch up.

Basically, if they remove nautlis from Gnome... gnome would be great. but for now it's the reaswon I am staying away.

Re:Worried Gnome User.....

[Skeezix]

The dependencies of Mandrake or Redhat or any other distro conflict with Ximian. If I want to install Ximian I am better off uninstalling all my Redhat packages that Ximian duplicates.

If there are conflicts with your distribution, you should report them on bugzilla.ximian.com. I have never had problems with installing Ximian GNOME on top of Red Carpet. (I use Red Hat).

I thought this was the job of distros, to test packages for their systems.

It is the jobs of distributions, however, most distributions do not release updates for GNOME all that often and often do not test well, with a few exceptions.

Do not take this personal, but Ximian has no solid business plan and is most likely to fail.

I don't take that opinion personally. For one thing, I don't work for Ximian.

Re:Worried Gnome User.....

[msaavedra]

It's forced upon you and if you follow their instructions to remove it, it dont work... nautlis still loads as the "desktop manager" and still steals about 30% of the entire computer's performance.

I agree that nautilus is too slow to use on most systems, but you do not have to use it to use Gnome. Go to the preferences dialog in a Nautilus window, select the "Windows and Desktop" section, then uncheck the "Use Nautilus to draw the desktop" box. Nautilus will no longer load when you start Gnome. Simple, wasn't it? If for some reason this doesn't work on your system, you should submit a bug report.

Re:Worried Gnome User.....

[Lumpy]

I did, I have to rm nautlis and symlink it with xfm to make it go away completely.

Re:Worried Gnome User.....

[VFVTHUNTER]

Thanks for the tip...since you too are running both Gnome and KDE, any tips on how to get them to happily co-exist?

Re:Worried Gnome User.....

[d_i_r_t_y]

In the same vein, it seems that KDE is for the people who want the solutions given to them, and Gnome is for the people who want to build their own customized solution (IE what they want.) That explains a little bit of the difference in the attitudes. Capt. Obvious does point out to me that the "build your own solution" approach is a general platform for Linux, KDE included, but I think Gnome really goes after that more.

how'd you figure that? kde is equally or more configurable in total than gnome -- from widget styles to xml-rpc and soap scripting, not to mention the number of options for the rather amazing konqueror. i think you must be refering to playing with the gnome panel and pixmapping widgets. the kde windows manager does rather suck, i must admit.

Re:maybe GTK is not the correct choice for the fut

[Frothy Walrus]

it's not that i love Motif. it's that i don't think GTK has done anything except make it look better, and that's not a lot of progress for ten years.

Re:maybe GTK is not the correct choice for the fut

[zulux]

i don't think GTK has done anything except make it look better,

The coolist things about GTK are #1 - It's Free (Speach) and #2 It's cross platform (Unix, Max and Win32)

Sick of Fantasy Movies

[Anonymous Coward]

Damm, first Harry Potter, then Lord of the Rings, now Gnome Two, what next: The Company of Orcs? An American Elf in London?

Sheesh...

My advice

[kitts]

This might sound like trolling, but I'd wait a bit until the GTK libraries are settled before beginning to seriously develop for Gnome. One of the big problems with GTK from version 1 to 2 was how certain widgets went through two or three different revisions, namely that very attractive treeview widget. With bonobo architecture on the way, stuff might change even more.

Re:My advice

[jrb]

GTK+ and the rest of the GNOME libraries are currently API frozen. Now is a great time to start using them.

Re:My advice

[kitts]

If this is the case, I'm glad to hear it. It wasn't all that bad working with GTK+. It would leave my only knock being trying to create custom widgets in a non-OO language, but since Gnome being C-based is part of its original conceit, it's not really a productive complaint.

A question though. Is it normal for a library set to surpass the 1.0 version but still continue changing APIs?

Re:My advice

[bunhed]

Exactly! I gave up on gtk before v1.0. It was and still is a an unfocused pile of spaghetti code. There has never been a thought toward backward compatibility and for that matter why the hell would somebody write and oop style api in a non-oop langauge? It boggles my mind that it even exists. But then again, KDE is ugly and gnome is pretty. :)

Re:My advice

[GauteL]

Actually... AFAIK the API has been frozen for quite some time. The only changes that have been accepted since 1. December is accessibility-stuff and very small and critical changes.

I think it should be pretty safe to start developing for Gnome2 now.

Re:maybe GTK is not the correct choice for the fut

[Frothy Walrus]

The coolist things about GTK are #1 - It's Free (Speach) and #2 It's cross platform (Unix, Max and Win32)

yeah, those are the coolest things about it. i prefer software whose coolest features are features, not licensing and 8-way compatibility.

call me a troll i guess...

look ahead to GNOME 2....

[hyperstation]

...and see all of the KDE developers looking back at you!

Yay!

[Burgundy Advocate]

Hey kids! Now it's time to pretend that everyone here on Slashdot actually develops programs!

"Yeah, last night I was really tring to get the object-oriented cobol bindings to gtk+ working but then in a fluke there was this gcc bug that caused my userspace code to go wonky and install the wrong x colormap which recursed until the system locked up. It was righteous."

I don't know where open source would be without the fine users of Slashdot and all the wonderful programs they develop.

Re:Yay!

[jasonv118]

Um... I think most Slashdot users do develop. I know I definitely do.

Re:Yay!

[Shabazz]

Thanks for giving us something to go on. Just because you program doesn't justify your statement. I don't develop anymore. Doesn't qualify me to assume that that means that most slashdotters don't develop, however. Looking through past polls makes me think the number of IT workers is high. Don't know how many people develop though.

Using "ClearType" with XFree86, GNOME, KDE

[AirLace]

While they are working on basic AntiAliasing, you should try out ClearType [microsoft.com]. It improves font quality up to 800% on LCD screens. The antialiasing level is also adjustable... One caveat: you need WinXP [microsoft.com] to run it...

Rubbish. XFree86 has supported what Microsoft calls "ClearType" for over a year.

The method, called sub-pixel rendering, is designed to work with LCD panels. This is why Microsoft are pushing for its use on laptops and palmtop devices. On standard CRTs, it holds no advantage over standard greyscale anti-aliasing.

A single pixel of an LCD screen is actually composed of three "sub-pixels": one red, one green, and one blue (R-G-B). Taken together this sub-pixel triplet makes up what we've traditionally thought of as a single pixel. This means that an LCD screen boasting a horizontal resolution of 800 whole pixels is actually composed of 800 red, 800 green, and 800 blue sub-pixels interleaved together (R-G-B-R-G-B-R-G-B ...) to form a linear array of 2400 single-color sub-pixels. That's where I guess you got your 800% from.

"ClearType" can be enabled in XFree86 versions 4.01 and greater by modifying /etc/X11/XftConfig. Just append the following line:


match edit rgba = rgb;


An in-depth look into sub-pixel rendering support in XFree86 is available here.

Re:Using "ClearType" with XFree86, GNOME, KDE

[Com2Kid]

This is the difference between Linux and Windows.

Microsoft makes their cleartype technology enablable with a big large red button shouting "PUSH ME DARN YOU PUSH ME!".

Linux puts it in a text file.

While having the text file option is nice and all, it would be better if the friendlier installation packages said at install time "heya, we noticed that you have an LCD screen, any chance you want to enable LCD font antialiasing? It'll make your text alot easier to read!"

For the boxed distros, shove it on the back of the box in a yellow jaggy oval. Bright yellow.

Also include it in a settings->display style applet. Make it obvious. Make it easy. Make it so that people KNOW its there.

Linux has alot of unused features, unused because few people /KNOW/ about them.

Hell, it could be able to cure cancer and theres a chance the lot of us wouldn't know what string to put on what line to enable it. :)

Re:Using "ClearType" with XFree86, GNOME, KDE

[slamb]

This is the difference between Linux and Windows. Microsoft makes their cleartype technology enablable with a big large red button shouting "PUSH ME DARN YOU PUSH ME!". Linux puts it in a text file.

The parent post was a bit deceiving

The code to do this on Linux very new. It's part of the Xrender extension to XFree86 which was introduced in 4.1. It requires toolkit support. Qt has it. Gtk does not have it (though 2.0 will).

Editing that text file won't do much for you if you aren't using Qt. And if you are using Qt, you probably have it enabled by default...there's a checkbox to use it, though the XftConfig settings also affect it. In the future, KDE will probably have a fancy graphical configurator that will do everything the text file has. These things take time. I'm sure in Microsoftland a lot of Control Panel options started out life as something you could only change with regedit. You're just seeing the process with Linux instead of getting the final product.

Re:Using "ClearType" with XFree86, GNOME, KDE

[Com2Kid]

Concentrate the effort on first producing those features that look/sound good.

I know it sounds nasty, but a balance can be found.

If Linux wishs to pull ahead, then a balance has to be found.

Concentrate on the features that people really want. Or at least that people think that they really want.

The opensource development community has the capability to not only /match/ microsoft feature for feature but also to /surpass/ them.

*COUGH* USB2.0 *COUGH*

Re:Using "ClearType" with XFree86, GNOME, KDE

[slamb]

If Linux wishs to pull ahead, then a balance has to be found.

Linux is an operating system. It doesn't wish anything. And the open-source development community isn't the borg. Developers are individuals. If this feature is important to you, why don't you do something? Do the work yourself or, if you don't know how, pay someone else to do it.

Someone will surely get around to it eventually. But when that feaure becomes important to them. Once again, developers are individuals. Many are not striving for Linux dominance in any given market; they are simply making software they want or need.

Concentrate on the features that people really want. Or at least that people think that they really want.

Absolutely. The developers concentrate on the features they want. Oh, is that not what you meant? Well, too bad.

Re:Using "ClearType" with XFree86, GNOME, KDE

[Com2Kid]

Alot of people seem to want for Linux to gain a large user base.

I am simply stating what is needed if Linux is to do such.

A larger user base would mean more hardware support and more software support from other companies.

It also has its down sides of course, but this is not the place for that debate.

If features are to be added then they should be added in some sort organized way. Sure alot of new features may look good, but do they help contribute to the overall usability of the OS/GUI/ETC?

Even more, are they the most efficent contribution? Could time have been spent implementing features that add even more to the OS then the ones that were implemented? Too late now, water under the bridge and all of that, but prehaps a list of these hypothetical "high gain" features should be compiled for prioritized implementation.

Re:Using "ClearType" with XFree86, GNOME, KDE

[Pussy Is Money]

Bah, if you're worried about developer efficiency, then Linux is not for you. Stop using it. It will never do what you want it to.

Re:Using "ClearType" with XFree86, GNOME, KDE

[slamb]

Alot of people seem to want for Linux to gain a large user base. I am simply stating what is needed if Linux is to do such.

I think you are wrong. It's not necessary for each developer (or any developer) to have the goal of a larger Linux userbase for that to happen. I assert that Linux developers have never had that goal, yet obviously the Linux userbase continues to grow.

Each developer pushes for the features that are important to him or her and making them happen. Collectively, all those features result in a pretty good product. As a consequence, Linux gains a larger userbase. It doesn't have to be their primary goal for it to happen.

You will never get people to do the "most efficient contribution" or the one that "contributes to the overall usability of the OS/GUI/ETC" because you will never get people to agree on what that is.

It's not a bad system. If you really want something, you make it happen. If you don't, you whine on Slashdot. If someone else eventually finds that same feature useful, you can take advantage of it as well.

Re:Using "ClearType" with XFree86, GNOME, KDE

[sumengen]

I think you are wrong and mistaken.
Cleartype on CRT monitors looks really nice, much nicer than antialiasing. I am comparing winXP box at home to win2000 box at work.

Cleartype makes a huge difference on LCD on my labtop, but it makes also enough difference on CRT's.

Re:Real Features

[diamondc]

blah blah blah. that's why there's so many Linux distros, to suit every kind of need. If you want multi-language distro use Suse or Mandrake since those are more international than RedHat or Slackware. Choosing a language is one of the first choices when I install Debian.

Oh, and to get my Creative Webcam 3 installed in Linux took at the most 2 modprobe commands and compiling gqcam to take pictures. While I could never get it working in Windows 2000 (this was a year ago though)

Re:Debian

[diamondc]

apt-cache search libgtk1.3

for unstable at least =)

Re:Real Features

[Com2Kid]

"Choosing a language is one of the first choices when I install Debian. "

When you install?

How. . . . quaint.

Multiple language packs easily and seemlessly working together in all applications. Installed on the fly as the user encounters new text, small downloads perferable.

"Oh, and to get my Creative Webcam 3 installed in Linux took at the most 2 modprobe commands and compiling gqcam to take pictures. While I could never get it working in Windows 2000 (this was a year ago though)"

That is one slight issue, a shitpot load of patchs not on the windows update site (hidden within the bowls on microsofts webpage, sometimes not even linked too from within it, Google is your friend.) need to be installed to get great USB support. By default it is only decent.

It doesn't help that most webcams insist on using their odd arsed data tranfer format instead of .

Re:Real Features

[Com2Kid]

Uh, make that "bowels", not "bowls". ^_^

Re:Real Features

[diamondc]

I am not sure how to switch language sets from one to another (never had to), but i'm sure its a matter of setting the LANG variable or LC_TYPE and then generating the locales.

Misleading article summary

[Anonymous Coward]

This is a good article that will teach you how to use the new and improved libraries available with GNOME 2 so that you can write your own Nautilus view, and panel applets. It also provides you with the understanding to compile a few sample GTK+ 2 programs that will give you a good understanding of GTK+ 2's many improvements over GTK+ 1.

Not really -- that's what the series intends to cover, but this particular article only covers an introduction and overview of Gnome 2. For those familiar with the pace of previous IBM Developerworks Gnome articles, you know that it will take awhile before this series gets finished (assuming it does).

Re:Real Features

[Com2Kid]

That is precisly what I am talking about. It should require no user intervention at all. :(

In fact Asian Charecter support in Windows is also easy, just a single EXE file to be downloaded and run. Just select the desired language and spell things out phonicaly in romanized letters, watch the desired charecter appear.

More popular programs like this are needed on Linux, and need to be installed by default. A standard of some sort needs to be set that allows for all programs to easily use these features. Imagine every Gnome/KDE/ program having easy to implement naturalization.

Or a Hiragana flashcard program, or a Kanji flashcard program, or both intermixed with each other with no issues at all. The english letters/word choices being shown at the bottom of the screen.

Now imagine the Kanji charecters also being antialiased. Along with everything else.

All the user would know is that he/she is improving their Japanese skills on a very nice looking display and that it was even easier to get running that the Windows equivilent.

You know

[Microlith]

GTK/GNOME, QT/KDE, and GTK/Ximian and other similar packages all have their advantages, but you know...

Taking all their advantages, sorting out the code until only the best remains, and having linux users standardize on one backend (not frontend) would be so great, but it shocks me how much shit one would catch for it.

Anyone ever thought about just combining all the various backends into one, unified, STANDARD (that's what linux users like to parade about), backend API and let the users use any frontend without library problems?

Just a thought...

Re:You know

[Jagasian]

Yeah, I am putting a Honda Civic backend into my F50. Oh wait, the backends for both cars are so different that trying to use the same backend for both cars is just silly. You ever wonder why the auto industry doesn't put the same backend into every car?

Yes it could be done, two wrappers around the same backend resulting in QT and GTK. However, such a task would require so much additional coding that it would negate any benefits that it was supposed to have.

I hate to say this, but what will happen, is one of the desktop environments will die out. Its the nature of Open Source Software. Things can't stay splintered forever. Eventually natural forces (user's needs, technical needs, development needs, market needs, etc) will cause the gap between both desktop environments to widen. However, the initial competition that exists when there is still more than one option helps the end user get a better desktop environment sooner.

Now this has nothing to do with other ultra-feather weight desktop environments, which will compete amongst eachother for the bare bones performance niche.

Back to the dying out of one of Gnome or KDE... well, you can give any prediction you want, but the productive thing would be to contribute to the development of the desktop you like the best. Contribute by using the desktop and reporting bugs. Contribute by writing code. Contribute by making art (icons, themes, sounds, wallpaper, etc). Contribute by educating others about the desktop of your choice.

Finally, my point is that the dream of unifying Gnome and KDE is silly. One of them will kill the other, and thats a good thing because it won't happen until one is orders of magnitude better than the other meaning that natural community and technical forces will choose the best desktop for us.

GNOME is unorganized

[Anonymous Coward]

GNOME team is out of focus. Its development is very very slow. Not much stuff has changed. The differences will be very few.

Like said in a post, it's too little to consider a 10 years work. GNOME is always difficult to install from source. Hundreds of dependencies and packages.

Would be better if they did arrange the packaging in a better way:

gnome2-gui.tar.bz2 (4MB)
gnome2-extra.tar.bz2 (4MB)
gnome2-libs.tar.bz2 (5MB)
gnome2-core.tar.bz2 (9MB)
gnome2-applications.tar.bz2 (9MB)
gnome2-addons.tar.bz2 (3MB)

Hmmm... Looks much more organized. Lots of packages merged into "gui" like gtk+, glib, pango, etc.

In "extra" we have the stuff like esound, audiofile, etc.

And so on... forget the hassle to download all the 60 tarballs. Just download one single tarball, untar/ungzip it and start compiling!

But yet, GNOME as it is, is unorganized for real.

Re:GNOME is unorganized

[Skeezix]

GNOME team is out of focus. Its development is very very slow. Not much stuff has changed. The differences will be very few.

Actually we are pretty focused though of course there is always room for improvement. Much has changed in the GNOME 2 platform. If you'd read the article you'd gleam a smidgeon of of the vast work that has gone into making Gtk+2 better. That's just one aspect of GNOME. GNOME 2 requires porting to a new platform and as such, is taking time. Many of the user-visible improvements will be visible in subsequent releases, though I personally think GNOME 2 is quite exciting from a user's perspective.

We release the way we do for several reasons. The individual packages are just that, individual pieces of a platform. For users who have slow modem connections this is a godsend. Also many people do not want to get the whole platform. They just want small pieces. There are other reasons as well which have been hashed out several times in the past. I'd be happy to talk more about it offline if you want...

Re:GNOME is unorganized

[Anonymous Coward]

The problem is, GNOME is designed to maximize sharing of code. Take libxslt, for instance. You can use it on it's own on Unix or Windows. You can use it in a pure GTK+ application. You can use it in a GNOME application. You can also use it in KDE (as a matter of fact, KDE 3.0 includes it precisely because it's not tied to GNOME).

You wouldn't have that flexibility if libxslt were packaged into libgnome. Even if GTK+ programmers and KDE 3.0 programmers agreed to depending on libgnome, you'd still lose the Windows port since libgnome would be much more difficult to port to Windows than libxslt.

Re:GNOME is unorganized

[stefanlasiewski]

The individual packages are just that, individual pieces of a platform. For users who have slow modem connections this is a godsend. Also many people do not want to get the whole platform.

There is another group of people who want a version of Gnome which is easy to install that is not Ximian Gnome.

KDE is easy to install: You download several RPMs or several tar.gz files & unpackage them. Same thing with Xwindows.

Gnome, OTOH, can a real pain to download & install.

Granted, I haven't tried this since the 1.2.0 days, but I remember that it took me hours to find a working ftp mirror which had all of the right packages, and I was never able to statisfy all RPM dependancies from a single FTP site.

That's why so many people rely on Ximian to provide the Gnome: They have great packaging, and Ximian is simple to install.

Re:GNOME is unorganized

[Sloppy]

GNOME is always difficult to install from source. Hundreds of dependencies and packages.

Would be better if they did arrange the packaging in a better way

You fool! That's no way to sell Linux distributions!

Re:GNOME is unorganized

[oldmanmtn]

I guess I still don't understand Slashdot moderation. How can something be both a '4' and a 'troll'?

Honestly, s/he has a very good point. The current Gnome oranization is probably relatively easy for the Gnome developers, since each "project" has its own tarball, but it's a real pain in the ass for the end users.

In theory, you should be able to just download the tarballs that make up the latest 'stable' build, and build that. The last time I tried this (for 1.2, I think), one of the packages needed a newer version of another than was in the 'stable' distribution. The newer version of that package needed a newer version of another one. The whole thing snowballed until I found myself trying to get bonobo and the new GConf stuff shoehorned into gnome 1.2. I gave it up and went back to fvwm2.

I wish I could remember the specifics, because I know this sounds a bit unlikely, but it was 6 months ago. I expect I'll get a lot of "It worked for me, so you must be a dumbass" responses. By simple /. demographics, most of those will be from Linux users. I tried this on Solaris 8, which probably gets a lot less testing than Linux. (Not that that should be an excuse. If you claim to support a platform, the damn thing should at least build successfully.)

Re:GNOME is unorganized

[johnnyb]

Actually, the packaging makes sense as far as the flexibility goes.

glib is useful on its own. Without anything else. it gets its own package.

Gtk+ is the widget set. It gets its own package (you can write applications just to Gtk+ if you want).

CORBA stuff is independent. They get their own packages.

Some things are likely to install even in distributions without GNOME. They get their own package as separate downloads.

Each application needs to be independently installable.

If you want something easy and organized, why not just either a) skip the direct download, and wait for your distribution, or b) use Red Carpet for download, and let it resolve the dependencies for you.

If the complexity of compiling bothers you, don't compile.

Re:GNOME is unorganized

[Chuck Chunder]

I'm not so sure that big packages would help (they'd only really be useful for the initial installation, for incremental upgrades you'd need to find the individual packages that had changed since your last one, especially if you are on a modem).

What would be nice though would be if the latest sections of the GNOME ftp site would be well maintained, free of multiple entries and kept up to date or culled (xchat 1.2?).

Re:GNOME is unorganized

[be-fan]

Actually, with modem connections, smaller files are a pain in the ass. The thing is that most of the packages are *required* dependencies. Instead of just starting a big transfer and leaving it on overnight, you have to deal with starting many different file transfers, which is quite a risk on many modem connections.

Re:GNOME is unorganized

[Taurine]

Bang on! I followed both GNOME and KDE for a while, but eventually I gave in on GNOME for a number of reasons, foremost of which being the complexity of a build from source - first identify which thirty or so tarballs you need for the base system, then which order to build them in.

From what I have read, getting it installed on package-based Linux distributions isn't much easier. Maybe what started out as disorganisation became a great business opportunity for Ximian - providing a simpler way to get GNOME installed? Meanwhile, many people are just happy building KDE.

And the problem is?

[Otter]

Worried Gnome User.....I am a Gnome user, and athough I am NOT a sky is falling person, KDE seems to be making much more usefull strides..

What's the problem, exactly? Gnome will get better, KDE will get better, you can use as much of either or both as fits your needs and at worst, you can go on using the current versions.

It's not clear to me where there's a problem. What's the worst that'll happen -- you might be tempted to change desktops to something that works better for you? You can even keep using your GTK themes.

By the way, VFVTHUNTER, you can turn off the launch feedback indicator on the cursor. I'm on a Mac right now, but it's in a pretty obvious place in KControl.

Re:And the problem is?

[Anonymous Coward]

It is not that great that there are two different competing desktop systems. We have limited resources and spend too many of them duplicating each others work. If you write a program, there is no easy way to have a dock app for each competing window manager. It would be nice for some simple standards in this area. It would not be too difficult to cooperate a little.

Re:And the problem is?

[msaavedra]

there is no easy way to have a dock app for each competing window manager. It would be nice for some simple standards in this area

That is simply untrue. There has been a joint WM Spec for quite some time. Not just for Gnome and KDE, but for other WMs such as Enlightenment and IceWM as well. I personally use IceWM, along with the Gnome panel (with pager and task manager applets). Everything interoperates fine.

Re:Real Features

[phaze3000]

Generic HID drivers have been in the 2.4.x tree for quite a while.

I am currently typing this with a USB keyboard, and will click the 'submit' button with a USB mouse, both of which are plugged into a USB hub on my monitor. If you plug more than one mouse into the computer each gets its own /dev/input/mouseX device.

As for installing font packs, yes it can be that easy. On my system I'd just have to type 'apt-get install kde-i18n-ja', but if you really must have a point and click interface to it, you can install RedCarpet which will make things similarly easy.

look ahead to KDE

[Pope Slackman]

...and see Microsoft looking back at you!

C-X C-S

Is it really cross platform?

[stefanlasiewski]

>#2 It's cross platform (Unix, Max and Win32)

Is it really cross platform?

I keep hearing this rumor of GTK running on Win32, but all I can find is Gimp on Win32 and these, which look more like geek-hobby-project then a stable product that a company can rely on (Note, I'm *not* trying to insult Tor at all, but in his words "I work on this project in my spare time ...don't hold your breath")

Can someone please provide some real links for *real* 'cross platform' GTK projects ?

Re:Is it really cross platform?

[zulux]

http://www.gimp.org/~tml/gimp/win32/new-downloads. html

is a good starting point. GTK makes a good attempt at cross platform - I've had better luck with QT myself though. YMMV.

Re:Is it really cross platform?

[zulux]

The only benifit that I can see for GTK vs QT is that GTK is LGPL and QT is GPL.

Re:Is it really cross platform?

[asincero]

> QT is QPL, by the way

And also GPL.

- Arcadio

Re:Is it really cross platform?

[broody]

You want corporate stable now? Try WxWindows. Granted the IDE choices are limited more or less to WxDesigner and it's not cross platform GTK but it's close.

Re:Is it really cross platform?

[Ghyl]

My understanding is that the Windows version of QT doesn't have a GPL licence (QPL only). Why Trolltech doesn't want to release a GPLed version, I don't understand, since you can't do Windows versions of GPLed software, like you can with GTK (eg The Gimp). Can anyone enlighten me?

Re:Is it really cross platform?

[zulux]

People were refusing to use KDE due to Trolltech's licencing restrictions on QT - and even a few KDE developers started to create a free version of QT. Trooltech then decided to GPL their Unix version of QT and kept the QPL for the Windows in order to keep the revenue. Basically, Trolltech wants money.

Re:Is it really cross platform?

[harvalen]

and kept the QPL for the Windows in order to keep the revenue. Basically, Trolltech wants money.

QT for windows isn't released under QPL.. Version 2.3.0 is released under a non-commercial license (which isn't open source), this and the rest of the versions are only released under their commercial license.
Hopefully they'll release QT 3 under a nice license for windows too, but atm I have no need for the windows support and just loves QT-X11 =)

Re:Is it really cross platform?

[bzzzt]

Mozilla _does_ use GTK for low-level drawing. Reason why it's so slow is the system on top of it (rendering, XUL etc). Besides that, did you profile both toolkits for your CPU/memory conclusions?

Horizontal Scroll Bar for new Text Widget?

[mr_don't]

I don't know if i missed it or not, but did the article mention anything about using a horizontal scroll bar for the new text widget? I hate those stupid "wrap arrows", and I was hoping the new text widget would replace them! The screenshot shown doesn't show either a wrap arrow nor a horizontal scroll bar... Anyone closer to the project know?

Re:Horizontal Scroll Bar for new Text Widget?

[Havoc Pennington]

It has a horizontal scrollbar, yes.

Re:Horizontal Scroll Bar for new Text Widget?

[tempfile]

Lo and behold, Gtk 2 won't need a hscroll because it will have word wrapping (finally!). The developerworks articles on Gtk 2 have a few screenshots.

Re:Real Features

[djcdplaya]

three letters to make unix easier:
B S D

We should be taking a lead from BEos.

[foqn1bo]

Before you automatically hate me, I am not necessarily bashing KDE + GNOME. However, I am a former BEos user and I would love to see a window manager/system/full-on GUI that maintains an ideal balance between beauty, features, simplicity and small size. For those of you who never used BEos, its GUI was so incredibly fast an efficient it made even the most hardcore Windows/Mac/UNIX user drool uncontrollably. That, and it was really nice to look at, and packed with desktop features that even KDE in its advanced state has yet to implement entirely to my liking (e.g., 'drag and drop to the extreme'). GNOME looks nice and is definitely useable, but it's sluggish on my 400MHZ PII, and I've seen the Beos Tracker fly on 166MHZ pentiums. KDE is too much like Windows, and has also gotten much slower since v 2.0. On the sparse side of things, Blackbox is quick and nice looking but low on features, and pretty much everything else I've seen is outdated and ugly looking. Has anyone ever considered a project using the resources of OpenTracker?

Good to hear news from Gnome

[p24t]

...something we don't get often enough. I remember when I first loaded Linux, Gnome was the greatest. It was easy to use, stable (well, relatively), and it looked nice. The GTK+ helped get some great programs available.

But since then, Gnome has remained Gnome. No new, fantastical releases. No big news. KDE has taken the spotlight.

And between the two, I now prefer KDE. I mean, KDE 2.2.2 is fabulous. The desktop icons are easy to work with, the panel is just sweet. So much easier to use than Gnome. Konqueror is come so far. The first time I loaded Konqueror, I thought it was useless. I actually preferred Netscape (blech). Now whenever I'm in KDE, I use Konqueror almost as much as I use Mozilla. Especially when I don't want any cookies being passed without my knowledge. I know people that use Koffice exclusively.

Don't get me wrong, I am -so- glad to see some big achievements from the Gnome camp. One of the biggest and greatest attractions to Linux is choice. Choice to use whatever you want, and to have as many options as you can handle.

I'm not a developer, I wish I was, but don't we all? I didn't read the code in that article too closely, so I may have missed some great points. But GTK+ 2 looks like it has a promising future. As long as they keep going on it, making it easier to use (and install), maybe they'll surpass KDE again.

Or maybe everyone will just go back to the CLI.

Personally, I use AfterStep. Dunno why, I just like it. Talk about hard to configure, though.

p24t

Pathetic

[njdj]

As a developer, I found the article sad. It's like going back at least 5 years, to the days when people used to use C. Somebody ought to tell these guys that the rest of the world has migrated to C++.

Re:Pathetic

[tempfile]

Oh yes? Such like Windows' fantastic C api that is *way* more cluttered than Gnome 2's due to MS's strict backwards compatibility. Windows developers know the pains of dealing with "versioned" function names, *Ex elements etc.

If you want a C++ interface to Gnome, go ahead, help developing a wrapper library.

Re:Pathetic

[emerson]

Oh God. They have?!? Quick, somebody, TELL LINUS!

Re:Pathetic

[21mhz]

Somebody ought to tell these guys that the rest of the world has migrated to C++.

Oh really? Did the rest of the world settle on the ABI then?

Some people have no damn sense of humor..

[PopeAlien]

Extreme Programming? Really? I always thought it had something to do with making your machine run as fast as an XT machine, and starting to look a bit more like Xwindows, without the flexibility. I didn't know that the 'P' stood for 'programming'.. Well I guess we'll just have to take their word for it, since you can't actually look at the code..

Hmm.. I guess I better start putting disclaimers and lame emoticons all over my posts to make sure no one takes them to seriously. ;) :-0 +8()

Re:Real Features

[Pussy Is Money]

How did we end up with all these people who really shouldn't be using Linux, but nevertheless feel they need to dictate where Linux should go?

Linux is not for you. Use Windows. Be happy. Stop whining.

I hope this will be great.

[tempfile]

Finally, a consistent version of Gnome. I've been using Gnome (1.2 and 1.4) for about a year now, and was even driven to KDE for some time by Gnome 1's extreme ugliness in terms of consistency. Several different CORBA interfaces for the same thing. Ancient, slow-ass Gtk+. A file manager (gmc) that comes with its own collection of file type bindings, independent from the rest of Gnome.

As it looks, Gnome 2 will finally offer a consistent and well-defined set of mechanisms, enabling reliable inter-application operation and efficient working.

Now, if the Gtk+ team has sorted out Gtk 1's horrible slowness, Gnome 2 looks like something I could fall in love with.

Share and enjoy, but remember

[alext]

Whoop-de-doo! At last, I can write GNOME apps and make them available for:

1. My distro, to share with all those other SuSE 7.3 on AMD K6-III users

2. Other geeks who carry big enough machines around to compile with

Excuse me while I just schedule a rewrite of my business application for all those GNOME enthusiasts out there waiting to build it!

Of course, the idea of GNOME 2 including a wacky and untried bit of rocket science like a VIRTUAL MACHINE to provide for sensible code distribution would wayyyy too much to deal with. REAL CODERS have no time for toys like IBM's Java SWT for GTK, Miguel's .NET VM clone and 'Parrot' - REAL APPS will only ever be written in C. To all those non-geeks with their wimpy embedded Linux boxes listen: get a clue (preferably a shelf-full) and upgrade, fer chrissakes!

Re:maybe GTK is not the correct choice for the fut

[Doomdark]

Well;

• Motif implementations in general aren't free/Free (lesstif does exist, though)
• Athena looks butt ugly, and doesn't have half of useful UI components people need.

So yeah, gee, if you don't mind using UIs that use properietary and/or ugly and/or incomplete widget sets, there isn't all that much benefit?

Additionally, GTK (and QT) widgets do have 'real' usability improvements too; creating internationalized consistently functioning accessible (via keyboard or one of the special devices blind users etc use) UIs is much easier, and as a result developers are more likely to create better UIs. UIs that are much more accessible to 'minority' user groups (foreigners, people with disabilities).

Interesting....

[EXTomar]

...how things like...

- Kernels(Linx, BSD, etc)
- Apache
- Most of the GNU Tools
- Most implementations of X Windows

...are all writen in C. And because it is "old" makes C "bad" why njdj? C++ and other OO languages aren't a "magic bullet" by any stretch of the imagination.

C and C++ and any other language binding have thier uses in the right places but to claim a language is bad because it is simply "old" is stupid.

Re:Interesting....

[Sentry21]

While I can't speak for kernel development, which is easily beyond any programming skill I've ever had, I can say a few comments about other things you cite examples of.

Apache was written in C, either because the programmers were UNIX programmers who are hardcore into C and refuse to use C++, because they had C source to work from, because there were no good C++ compilers out there, or any combination or lack of the above. Same with X-window implementations.

These both would be nice if implemented properly in C++, because the object-orientation features of C++ make a lot of things clearer and easier, and in a lot of cases, mind-numbingly less complex.

The GNU tools (I assume you're referring to things like wget, fileutils, binutils, and so on) are, 99% of the time, pointless to write in C++, because you wouldn't use object-orientation on such a small/limited scale (wget deals with one file at a time, why bother objectifying?).

I do, however, point you to other large projects that DO use C++ - KDE, Mozilla, AtheOS, just to name a few.

Large projects that deal with objects - buttons, windows, controls, lists, etc. - are great when implemented in C++ (if done properly), because it makes the code easier to deal with, less complex, more reusable, and on and on.

C++ isn't for everything, but for something like a graphical user interface, it would sure be nice.

--Dan

Re:Interesting....

[njdj]

And because it is "old" makes C "bad" why njdj?
...
but to claim a language is bad because it is simply "old" is stupid

If you'd actually read my comment before replying to it, you'd have realized that I did not say C was bad. In fact C is pretty good for some things. But for large projects that are best developed using object-oriented techniques, C++ is much better. If you dig into GNOME, (and if you're a developer) you'll see that it's using a lot of object-oriented ideas, and expresses them clumsily because C just doesn't have the support for them. Back in 1996, there may have been reasons to go this way. But in 2002 there aren't.

Re:Interesting....

[cout]

According to http://httpd.apache.org/ABOUT_APACHE.html, apache was based on NCSA httpd, which I presume was written in C.

C++ is nice for large projects, but its lack of a standard ABI makes dynamically loading modules a little interesting. This usually isn't an issue for GUIs, but for things like plugins, it would be nice to simply be able to inherit from a base class with a bunch of pure virtual functions, and have their implementation be the implementation of the plugin. Too bad I have to settle for a C-style hack if I want to be sure my code will work between different compilers.

Re:Interesting....

[johnnyb]

C++ ADDS complexity. It does not reduce it. The language is the biggest frankenstein language I've ever seen. There are several reasons not to use C++:

1) It adds complexity

2) No standard ABI (this is a very big point for libraries). This means if Solaris' GNOME was compiled with Sun's C++ compiler, you couldn't use gcc to build GNOME applications.

3) There is a larger pool of skilled C programmers. C++ programmers generally do not know enough of the strange nuances of C++ to adequately implement libraries.

If there were another OO language with a standard ABI, then there might be a case. Maybe Objective C. However, C++ is definitely a bad implementation choice because of the complexity it adds.

Re:Interesting....

[elflord]

1) It adds complexity

Depends. If you're going to write object oriented code, it's a hell of a lot simpler to write it using well documented, standardised semantics than trying to write C++ code in C. In the case of GNOME, complaints about it "adding complexity" are moot -- GTK/GNOME add complexity by using obscure idioms to emulate what C++ supports using well-defined semantics.

2) No standard ABI (this is a very big point for libraries).

The KDE people have been working on a library that generates C wrappers for C++ applications.

3) There is a larger pool of skilled C programmers.

If they're really skilled, they shouldn't have much trouble learning C++.

If there were another OO language with a standard ABI, then there might be a case. Maybe Objective C.

Anything but C++, huh ? A standard ABI would be nice, but it's not a be-all-end-all.

Re:Interesting....

[elflord]

C++ is nice for large projects, but its lack of a standard ABI makes dynamically loading modules a little interesting.

Not that interesting. It's about as hard as saying extern "C"

able to inherit from a base class with a bunch of pure virtual functions, and have their implementation be the implementation of the plugin.

But this is quite easy to do. To make this work, you need some variant of the prototype pattern, since you presumably don't know about all the derived classes at compile time. The function that one extracts from a dlopen()ed module should just create register a prototype by inserting it into a static member of the base class (eg: a list or map). To use the prototype pattern, you're going to have to create and register a prototype regardless, so I don't see how the language is making life more difficult.

Re:Interesting....

[johnnyb]

1) It adds complexity

Depends. If you're going to write object oriented code, it's a hell of a lot simpler to write it using well documented, standardised semantics than trying to write C++ code in C. In the case of GNOME, complaints about it "adding complexity" are moot -- GTK/GNOME add complexity by using obscure idioms to emulate what C++ supports using well-defined semantics.

***

To say the semantics of C++ is well-defined are true, but the semantics are not really that logical.

***

2) No standard ABI (this is a very big point for libraries).

The KDE people have been working on a library that generates C wrappers for C++ applications.

***

You are missing the point. The C wrappers are fine, but they don't help C++ people using a different compiler.

***

3) There is a larger pool of skilled C programmers.

If they're really skilled, they shouldn't have much trouble learning C++.

***

But why? C++ adds little but complexity.

***

Anything but C++, huh ? A standard ABI would be nice, but it's not a be-all-end-all.

***

Yes, anything but C++. I teach C++, so I know it quite well, and all of it's magic templating, virtual base class features. And I can tell you that trying to write useful C++ libraries is just plain a bug-ridden, pain-in-the-butt nightmare. Especially if you are using libraries from multiple developers. You have to wonder if pass-by-value might invoke the copy constructor. You have to make sure that all of the templates support the features you want, just right. And then, there's function overloading, which is just a plain bad idea (it's basically setting you up for calling the wrong version of a function accidentally). C++ has a lot of features. It even has one feature (the ability to pass templates as template parameters) that no other language has. However, the way that the features are put together is so wretched, that I can't imagine a worse way to do it.

And on top of that, there is no standard ABI.

mod parent down -1

[Penguinoflight]

Not only is this offtopic, it's also lame. Gnome still performs better than kde, and even if it didn't most people would say it doesn't matter.

Re:Real Features

[Com2Kid]

More users == more HW support (or at least cooperation) and more games.

Now I don't know about you, but more games. . . . ^_^

Not to mention productivity applications. It'd be sweet to see, say, 3DSM or Rhino3d ported to Linux.

Of course somebody could just write a proffesional level 3d modeler for Linux, but. . . . That would take an entire team of people working their butts off for many years in a row for a product that would be used by a small percent of a decent precent of the computer using community.

The main features I have listed are just mainly things that would cut through the MS FUD and /LOOK/ good. Hell most of them aren't even all that complicated. The main idea is to make it /appear/ that features which have existed for along time in Linux are really New Nifty Things.

Yes it itself is FUD, but it at least it would encourage people to switch to an OS that doesn't suck.

Hell I'm -stuck- on windows dispite moral issues with the way MS conducts business. I am stuck on it because I do 3d modeling and animation and that comes first before any ethical issues about where my software comes from. Sorry to say it, but most of the world is that way too.

Sure I could switch to Linux. Sure it'd be more stable and more robust and likely a good deal quicker.

But hell.

WTF would I do with it? I don't type very many documents, I don't do lots of networking stuff, umm. . . . . heh.

Rhino3d 3DSM and even some (yuck) Bryce.

Those programs _MAY_ come to Linux, but only if the Linux user base increases by enough to make it worth the developers while. If the Linux community does go about (and succeed in) attracting those developers to their platform, they will have gained a very valuable ally.

Now excuse me I have to go start up a product spanning across Rhino3d, Photoshop (yah yah I know, but damnit, it makes stuff easy to do, /REALLY/ easy. Great UI, and when I am creating something, the UI matters. check out my other rants to see me go pro-CLI. :) ) and Bryce.

Call me when those apps are ported.

GTK+ 2.0 is already outdated

[cout]

Do you really think people will know that a floppy disk icon is supposed to mean "save as"? When I see a floppy disk icon, I usually think "put this file somewhere that I will never want to access it again because it probably will be exposed to an insignificant magnetic field and will have all its bits scrambled into tiny pieces like eggs with little bits of chewy spam."

Re:GTK+ 2.0 is already outdated

[drewness]

Without question, floppies suck, but looking at Word 2000 on my laptop running Win2k I see a floppy disk on the toolbar, and when I mouse over it I get a tooltip that says "save". It may (or may not) be a dumb icon, but it has precedent, and current usage by the company that dominates desktop word processing.

Try fluxbox

[cout]

Blackbox development seems to have stopped since some time in 1999. A hacker known as fluxgen has picked up the project, and is doing some really cool things with it. None of that drag and drop silliness (which doesn't belong in a window manager), but lots of other useful features that blackbox is missing. He's also cleaning up the code, too -- not that blackbox's code was ugly, but it was written in a "C++ is a better C" form, which just isn't a good development model for large projects.

http://fluxbox.sourceforge.net

Re:Real Features

[Pussy Is Money]

Sorry to attack you. But what should I care that you are stuck on Windows? Most of us saw that coming a decade ago. And what exactly is your beef with Windows? It seems to run the apps you need and has the GUI you like.

You might want a "modeller" and "Bryce" for your work. That's great. But I prefer to simply use the computer to realize an idea, whether that's processing text or synthesizing sound or manipulating graphics. Then I just write a little program in shell or Perl or C to do just what I need. For that you don't need a team of people, just good ideas.

Enjoy your stay in all those big apps.

OT: /. moderation

[kubrick]

I guess I still don't understand Slashdot moderation. How can something be both a '4' and a 'troll'?

Modded up to 5 by other users, but the most recent moderator chose 'Troll', reducing the score to 4 and ensuring that word would be placed next to the score.

GTK--

[drewness]

Well, if you prefer C++ you could always use GTK--.

Re:Real Features

[Com2Kid]

Complex 3d scenes + animations != suitable for perl scripts. :)

Sure some people can nativly write POVRay code to render any scene that they can possibly imagine.

Some people can also effortlessly play Baroque music after hearing it just one time.

I can not do either.

I do not like GUIs for most things, and in fact one of the things that I love so much about Rhino3D (as do many other users after they make the initial 'adjustment' ) is that it has a CLI integrated into it for entering commands into.

But the fact is that in order to do my job I _NEED_ a 4 viewport 3D modeler with an Autocad derived interface.

I'd rather do without Bryce truth be told, numerious free alternatives have attempted to pop up, and Terragen was coming along nicely as Pay if Your Commerical Ware, but it has seemed to stop development (I haven't checked it out for 4 or 5 months, but it was inactive for over two years).

Until some nice highly developed tools come along the fact is that alot of people are going to keep on giving MS their money.

I am _NOT_ happy with Windows. I recently lost a bit of work (yah I save often, but still time between saves is creative time when you are doing artwork, and creativity is hard to redo even when you do have a recent save) dispite the fact that I am using a Win2K machine with "Everything Patched to Hell(tm)".

The lab I am in right now has in the last 4 years ungone the following upgrade proccess:

Win95>Win98>WinME>Win2KPro.

That is alot of money for Microsoft right there. Huge lab.

In fact with the most recent upgrade to Win2K Pro all of the applications but for IE, Netscape, Microsoft Visual Studio, and MS Office, have been taken off of the system.

I'm still trying to figure out why they are running Windows myself. . . . musta cost a fortune to upgrade the lab. . .

It would have been a perfect situation for Linux to step in and say "Hey, educational institution, you wanna save a /ton/ of money? Here ya go! Take this CDR I burned for you, install the software on all of your computers. Enjoy."

Ok so it would have had to been a decision on the part of the lab maintainers themselves, but still, the idea is that if Linux just made a bit of an effort to show how appealing it is by putting just a few features up front, and mabye one or two in big bold print, then alot of money could have been redirected from Microsoft's coffers to someplace else.