GNOME 2.30, End of the (2.x) Line

stovicek writes "GNOME 2.30 was originally intended to coincide with GNOME 3.0 — a massive cleanup and rethinking of the popular desktop. However, GNOME 3.0 is delayed for at least another release, which leaves GNOME 2.30 as most likely the last version in a series stretching back almost a decade. [...] 2.30 will probably be the final version of the 2.0 series. For those who were around for GNOME 2.0 back in 2000, the 2.30 release stands as evidence of how far GNOME in general and the free desktop in particular have come in the last decade in usability and design. If you do a search for images of early GNOME releases and compare the results with 2.30, you can have no doubt that, although GNOME sometimes tends to over-simplify, its improvements over the last decade remain unmistakable."

Comments

This PsOt is brawt to u by iAds

[Anonymous Coward]

TOY STORY 3!! NIGGERS!

Stop ignoring my posts.

[Singularity42]

I want replies.

Re:Stop ignoring my posts.

[lolwhat]

Is this some kind of reverse psychology? Well it will never work.

Uhmmmm

[Anonymous Coward]

I think like 99% of linux users use kde, so what's the point?

Re:Uhmmmm

[Jon.Laslow]

Because some of us don't like KDE4, and prefer the simplicity of Gnome.

Re:Uhmmmm

[bragr]

Trolling? I think so. If not: [Citation Needed] I very rarely see someone using KDE. In fact, its a bigger surprise, to me, to see someone using KDE than Linux itself.

Re:Uhmmmm

[Anonymous Coward]

It should not be a surprise to you that there are less KDE users than linux users. Because only a subset of linux users can use KDE.

Re:Uhmmmm

[Taco Cowboy]

I think like 99% of linux users use kde

Don't think so.

The latest figure is 98.9%

Re:Uhmmmm

[TiberiusMonkey]

Given the popularity of Ubuntu I find that hard to believe.

Re:Uhmmmm

[SteveFoerster]

Because no one who uses Linux uses the default install of Ubuntu... got it.

Re:Uhmmmm

[aztracker1]

Not to mention Fedora's default is Gnome too.

Re:Uhmmmm

[tomhudson]

GNOME is just jealous that they could be more popular if they just made it look and work more like KDE.

Seriously ... look at the difference between the fugliness that is Ubuntu (even with the new "blight" look), and the KDE variant. If they want to fix Ubuntu's visual problems once and for all, they should just do this. Because going from Halloween Orange-and-Black to "Rotting Eggplant" might be a change, but it's not much of an improvement

Re:Uhmmmm

[bragr]

You miss-understand me, I was implying that I see plentiful of Linux users around here, but that KDE is still so uncommon that it invokes surprise on my part.

Re:Uhmmmm

[Peach Rings]

There are a lot more things I don't like about KDE4. It tries to be all integrated, with a common notification daemon for example, so that status messages can appear with a consistent look in the corner of the screen. The problem is that virtually nothing supports it except for KDE apps that start with "K". If you want that sleek, consistent QT4 look, you're limited to a small subset of free software - there are a lot more GTK applications than QT applications. And I'd prefer to be able to use, for example, a different file manager. Without dolphin, you're unable to take advantage of KIO and whatever search index thing that KDE uses. KDE as a whole seems really tightly coupled - I regularly use gnome apps on my XFCE system without having the gnome libs installed. That's unheard of for KDE.

A particular barrier for me to use KDE is a decent web browser. I've used Konqueror for a few months and it's OK, but KHTML became intolerable. Arora (webkit powered) is good but incomplete. I have similar complaints about the usual KDE chat programs, music players, and Konsole.

Re:Uhmmmm

[Scarletdown]

KDE lost me as a user when they took away the ability to have a different background on each desktop (yeah, petty point, but it was still a really neat feature in my opinion.) That did cause me to take a closer look at KDE4 before I dropped it, and I found myself not liking the feel of that environment overall. Then I took a look at Gnome and tried it out briefly, but didn't care much for that either (plus, still no multiple backgrounds).

That was when I finally started taking a more in depth look at alternate desktop environments, which led to me settling on Enlightenment. It is definitely very different from what I have been used to (still have a lot to learn and unlearn), but the more I work with it, the more I like it. And the only annoyance thus far is the inability to have both an image as a desktop wallpaper and a different background color (either solid or a pleasant gradient) other than painful white if the image does not fill the screen.

I still use various Gnome and KDE based apps as required (K3B, Nautilus, GDM, and assorted others), but for the overall environment, I am converted over completely to Enlightenment.

Re:Uhmmmm

[Anonymous Coward]

You can run Firefox on KDE.

Re:Uhmmmm

[CopaceticOpus]

I'm really tired of the trade-off between simplicity and functionality. This trade-off should not be inherent to either windowing system. Rather, the variety of options presented to the user should be configurable. Each distribution should be able to decide how simple or how configurable they want to make their windowing environment when it is first installed.

A great windowing environment would be able to be made to look very similar to Gnome, KDE, Windows, Mac OS, etc. I don't just mean superficially similar - it should be configurable down to the menu options presented and the types of configuration options presented in dialogs. All of this presentation of options and behaviors should be managed in a special layer, much like CSS is used to configure the presentation of a website.

For example, Ubuntu might choose to ship with a very simple, user-friendly interface. In the system administration, this interface could be changed to a more configurable preset if the user so desires. It wouldn't be a matter of switching from Gnome to KDE. Visually, the change might be as dramatic as a switch from Gnome to KDE, but the basic windowing system would just be running with an alternate configuration.

Re:Uhmmmm

[celibate for life]

Do you really think Kubuntu looks good out of the box? I mean, I get it, it's not brown, but it's still ugly. Just a different kind of ugly: blue retard instead of poop brown. The fact is, if you want a tasteful desktop, you have to customize it yourself.

Re:Uhmmmm

[Peach Rings]

It looks like it's theoretically possible to build firefox with Qt widgets thanks to Nokia, but it's difficult and unstable.

And yes obviously you can just load both Qt and GTK libraries but it's ugly and memory-inefficient.

Re:Uhmmmm

[IMightB]

Correct. I need to get work done, not play around with the GUI all day.

Re:Uhmmmm

[celibate for life]

RAM is cheap today. Just throw a couple 2GB sticks and you won't notice.

Re:Uhmmmm

[bolthole]

"simplicity"? SIMPLICITY???!

"xfce" is simple.

GNOME, on the other hand, is now a more bloated pig than CDE ever was.. which is amusing, because one of the gnome1 boasts was that it was much lighter than CDE.

Re:Uhmmmm

[simcop2387]

and interestingly in kde 4.4 with firefox 3.6 it's even using the kde notification thing in the corner whenever it finds an update for things. I believe this is because its using a "standard" (don't know if it is or not) dbus thing to do the notifications so that both gnome and kde can use the same code.

Re:Uhmmmm

[Stormwatch]

"Simplicity" can mean different things.

Ask a regular user: a "simple" system hides any complexity; in this sense, Ubuntu is simple - everything is automated or set by GUI-based tools.

Ask a developer: a "simple" system is transparent; in this sense, Slackware is simple - there are few GUI-based tools to set the system.

Re:Uhmmmm

[tomhudson]

Or you can switch to a distro that wasn't put together by someone who is color-blind, aided by someone who is spatially challenged ... because like you said, it's STILL ugly.

My opensuse desktop and laptop both look gorgeous with the black Oxygen taskbar. Every once in a while, I change the wallpapers ... just 'cuz ...

Re:Uhmmmm

[Jurily]

I use KDE. 3.5.10 is still unbeatable when you want fast, clean and powerful. However, it's officially dead now, and the 4.x series sucks.

KDE4 is full of tiny little features that look like what you're used to, but do something completely different. Or they removed the most important button on the Konsole window, injected Amarok with featuritis, and k3b simply does not exist. (Wanna guess my three favorite applications?)

I'm starting to think Joel was right. The KDE project wasted two years trying to produce something "better" than what they already had. They failed miserably.

Re:Uhmmmm

[Anonymous Coward]

KDE lost me as a user when they took away the ability to have a different background on each desktop (yeah, petty point, but it was still a really neat feature in my opinion.)

When was this? KDE4.3 has that feature.

Re:Uhmmmm

[Abcd1234]

I'm really tired of the trade-off between simplicity and functionality. This trade-off should not be inherent to either windowing system. Rather, the variety of options presented to the user should be configurable. Each distribution should be able to decide how simple or how configurable they want to make their windowing environment when it is first installed.

Uhuh.

And then people would bitch about bloat because supporting all those features, options, and workflows would required a fuckton more code.

So here's an idea: pick the environment that fits your needs. Gnome, KDE, XFCE, or heck, throw components together that fit your needs. But quit expecting these projects to be infinitely flexible, it's completely unreasonable.

Re:Uhmmmm

[celibate for life]

Well, I don't think that looks are a legitimate reason to switch distros. But I concede that I always thought that (open)Suse looks the most stylish.

Re:Uhmmmm

[Scarletdown]

It may have been up through 4.2. I am pretty sure I had given up on KDE before 4.3. And I remember at the time, a web search on KDE +"multiple wallpapers" brought up tons of similar complaints with no one being able to offer any solution.

No matter now. I am becoming more and more enamored with Enlightenment these days, and feel no need to switch back to either of the "Big Two" desktop environments. :)

Re:Uhmmmm

[EzInKy]

So here's an idea: pick the environment that fits your needs. Gnome, KDE, XFCE, or heck, throw components together that fit your needs. But quit expecting these projects to be infinitely flexible, it's completely unreasonable.

That was my solution when I saw I didn't like the direction KDE4 was going. Openbox with netwmpager, fbpanel, and feh allowed me to keep my environment free of garbage all over my desktop.

Re:Uhmmmm

[Rick17JJ]

With the latest versions of KDE, it is now once again possible to have a different background for each desktop. I use Kubuntu 9.10 with KDE 4.3.2 and use a separate background for each of my desktops. However, it is was not very obvious how to enable that feature.

The following article mentions how to do that. In the article, read the paragraphs just after the heading which says "Combine Virtual Desktops With Plasma Activities." Notice that the last sentence says "This will allow you to have a different wallpaper on each virtual desktop, as well as a completely separate set of widgets. "

Seven Great Tips To Make KDE 4.3 More Friendly

Kubuntu 9.10 actually uses KDE 4.3.2, but the instructions in the article refer to KDE 4.3. However, those instructions still worked fine for me. Having a separate wallpaper for each virtual desktop has always been one of my favorite features. I was very disappointed, when with the early versions of KDE 4.x, they took that ability away. I am glad that it is now once again possible to to have separate wallpaper for each virtual desktop when using KDE.

Re:Uhmmmm

[shovas]

I use KDE. 3.5.10 is still unbeatable when you want fast, clean and powerful.

100% on the same wavelength here. Wrote a blog praising the benefits of CentOS 5 w/KDE 3.5.

I spent the better part of a year trying out new distros when they all started switching to KDE 4 and nothing seemed to click. So I'm crossing my fingers and hoping the distros polish KDE4 and get it back to the level of 3.5.

I have been eyeing XFCE, though, it looks like it might do. Gnome just does not work for me for a productive environment.

Re:Uhmmmm

[Anonymous Coward]

It would be easier, and ultimately more satisfying, to upgrade to Slackware 7.1 and FVWM2 than to stay on the DE hamster wheel. (And, for the pipsqueaks here, I remember running gnome 1.4 on my Linux-PPC install on a gossamer G4. Fast, solid and easy to use and configure. Downhill from there, but I'm glad mono-man has steady employment now).

Re:Uhmmmm

[Anonymous Coward]

So is your mom, but you still love her. What's your point?

Re:Uhmmmm

[CopaceticOpus]

It's interesting to me that my idea received one positive and one negative mod, but your negative, dismissive response quickly shot to +5 insightful. We could do with a little open-mindedness.

Depending on the implementation, my idea could result in a bloated mess. I readily admit that. But it's really easy to criticize an idea and call it impossible. When browsers were first released, it would have been thought impossible for them to do anything close to what they do today, and yet here we are. Javascript performance has made great leaps forward because Google was willing to question if those leaps were possible.

I can see two factors which could help my idea succeed. One is the continuing increase in computer power. The other is the chance to actually simplify and optimize the core windowing system code. For it to be sufficiently flexible, it would need to be carefully written. For example, when displaying a tree showing directory contents, the developers wouldn't be thinking about the specifics of the graphics, colors, and spacing. They'd be writing to a general case, which could be made very simple and efficient. Then another piece of code would apply a style to this element. CSS has already shown that styles can be applied with great efficiency.

Re:Uhmmmm

[phantomfive]

I think you didn't get modded up because you stepped into an old flamewar that people already have strong feelings about. Everyone has their own idea of what the perfect GUI should be, so unless you have extremely good writing skills and really help people to see what is so great about your idea, they are going to just stick with what they already think is best. FWIW I think your idea is a good one, but the implementation is extremely important. KDE used to be extremely customizable (maybe it still is, but I use Windowmaker because focus-on-mouseover rocks, although no one really agrees with me on that) but the implementation was horrible, so it was just a mess.

Re:Uhmmmm

[nine-times]

I agree. I'm about to say something that almost everyone here will think is completely crazy, but I'm going to say it anyway: some of these DEs and Linux distros should focus less on being infinitely flexible and configurable, and more on coming up with one single configuration that works.

Not that I don't appreciate the flexibility and configurability, but having all these different options ought to mean that at least one option is consistant, standard, and controlled. give me a distro that only supports one DE, but make sure that DE's experience is really smooth. Take away all the different themes, and give me one single theme that's extremely polished. Maybe even don't try to support every possible kind of hardware, but certify some set of hardware and support that hardware really well. Go ahead and make some choices for me, just so long as those choices are really good choices.

A lot of people would say that sort of mindset is antithetical to the open source movement, but I don't think it is. Leave it open source, and let other people make any changes they'd like.

Anyway, I it's part of the reason for the success of OSX. While the geeks are all complaining about the lack of configurability, everyone else is happy with how well crafted the defaults are. I think Gnome operates along the same line (but not to the same degree), and while that earns it the wrath of a lot of geeks, it's the reason why so many Linux distros use Gnome as the default environment.

Re:Uhmmmm

[PeterBrett]

The problem is that virtually nothing supports it except for KDE apps that start with "K".

Actually, that's not true. With the latest Fedora, for example, Firefox uses KDE notifications under KDE, and GNOME notifications under GNOME. The integration's spreading really, really quickly, now that the DBUS API has been fixed down for a release cycle.

Re:Uhmmmm

[AndGodSed]

if you mean "focus follows mouse" by "focus follows mouseover" then I heartily agree with you. In our office ALL the linux users use this feature. Gnome and evil supports it, and everytime we need to work on an OS that doesn't support it we get very frustrated.

Re:Uhmmmm

[muckracer]

> And then people would bitch about bloat because supporting all those features, options,
> and workflows would required a fuckton more code.

Perhaps they could use that new-fangled programming thingy that makes additional features modular add-on's....

Re:Uhmmmm

[AndGodSed]

Thanks for posting that link, that was the single most insightful ANYTHING I have read on programming this whole year.

Great read.

Re:Uhmmmm

[wrook]

Simplicity is a bad word in this context, as you point out. There are two concepts that it points to which are always getting confused.

Ease of learning: Something that is easy to learn lacks conceptual complexity. It is easy to understand quickly and it is easy to remember after being exposed to it only a few times. The downside to ease of learning is that you usually can't do things that are conceptually complex.

Ease of use: Something that is easy to use lacks operational complexity. It is easy, fast and convenient to do what you want to do, no matter how complex. The downside to ease of use is that the ability to do complex things quickly usually requires an interface that is difficult to learn.

You can optimize for either of these kinds of simplicity (ease). Your interface can be unnecessarily complex in either or both categories. Good software chooses the least complex solution, while still optimizing for the type of simplicity that is required.

Re:Uhmmmm

[Abcd1234]

Depending on the implementation, my idea could result in a bloated mess. I readily admit that. But it's really easy to criticize an idea and call it impossible

I never said it was impossible, I said it was completely unreasonable.

They'd be writing to a general case, which could be made very simple and efficient.

See, statements like this demonstrate that you're not a developer.

Writing code for the general case is *harder*. It makes code more *complicated and slow*. Why? Because ever piece of added flexibility means more lines of code to implement that flexibility. Hell, just look at the massive complexity of the GtkTreeView widget to see what can happen... great, it uses an MVC architecture to make it possible to display a whole range of different tree and list styles, allowing a single control to be used for multiple purposes. But the result is extremely complex.

They'd be writing to a general case, which could be made very simple and efficient. Then another piece of code would apply a style to this element. CSS has already shown that styles can be applied with great efficiency.

Hint: CSS affects presentation, not behaviour. And every GUI toolkit out there *already* makes it possible to customize presentation, they're called themes.

What you're talking about is customizing *behaviour*, and that must be coded into the widget/framework/tool from the get-go, and every piece of customized behaviour is more lines of code and more bloat.

As for the other AC suggesting plugins, once again, clearly someone doesn't understand how software is written. In order to build plugins for customizing behaviour, you have to implement the software so that there are entrypoints for those plugins to hook into. That means every piece of behaviour you might possibly want to configure will need to be specially designed so it can be hooked into and overridden. That's highly non-trivial. And you'd have to do this *for every single application*.

In short, it just makes no sense. The best solution, IMHO, is to go down the path we're already going down: standardize desktop protocols and APIs so that people can slap together a window manager, panel, file manager, etc, that suit their preferences. Fortunately, EWMH already exists, and Freedesktop is working hard to standardize other desktop-related APIs.

Re:Uhmmmm

[jadrian]

I regularly use gnome apps on my XFCE system without having the gnome libs installed. That's unheard of for KDE.

If they don't use gnome libs why should they count as gnome apps?

Re:Uhmmmm

[Vahokif]

You know that WebKit is based on KHTML right?

Re:Uhmmmm

[PoprocksCk]

Yes, my understanding is that it's a FreeDesktop.org standard (at least a proposed one?)

See this YouTube video for more details: http://youtube.com/watch?v=YjnCXKQ3MUc

Re:Uhmmmm

[laurelraven]

It's also a lot more advanced than KHTML from what I've seen.

Re:Uhmmmm

[snadrus]

Actually, opinionated choices are very much the nature of open source. Ubuntu follows this (especially with Quickly, one of its programmer tools), but a better example is PERL, the Pattern Extraction and Reporting Language. It's applied everywhere (in spite of | because) it was designed for such 1 specific purpose.

Sometimes I wonder if we should give up on the DE entirely and have an Eclipse-like arrangement of interrelated pieces that target one specific purpose.

Re:Uhmmmm

[Risen888]

1. The notification system on KDE is a FreeDesktop.org standard. Anyone can choose to implement it. Firefox already does.

2. The Qt-Curve theme will allow your GTK2 apps to share the same look and feel as native Qt apps. I'm typing this on Firefox right now, it fits in with the rest of my desktop quite nicely.

3. w/r/t your comment on Dolphin and KIO, it sounds like you're saying "Gosh it's wonderful how integrated KDE is, but I really wish it would integrate with [whatever non-KDE app you want]." Well... no. That's not how it works.

Re:Uhmmmm

[Risen888]

I don't think that's true. I think there are several DEs that work and are polished and work well and are ready for John Q. Public to use and enjoy. Really, just pick one, it doesn't matter, they're all great. But if people like the GP insist on not using them the way they're meant to be used, and try to mix and match shit and expect it all to work exactly the same, well, they'll end up pissing and moaning because that's not how it works.

Re:Uhmmmm

[Peach Rings]

No, I'm saying that KIO is a mistake because it requires Dolphin instead of being transparent.

Re:Uhmmmm

[Risen888]

KIO requires Dolphin? Um... no. Really, no.

First for the first time!

[skywatcher2501]

Yayyyy! ;D

Re:First for the first time!

[skywatcher2501]

OK after this moment of glory let's RTFA...

Re:First for the first time!

[lolwhat]

Fail.

Re:First for the first time!

[confused one]

Damn dude. Sorry. I accidentally stole your moment of glory. : (

Re:First for the first time!

[sayfawa]

Good thing you didn't waste it.

I got first post once, but I didn't even say "first post" or any other misspelled incarnation of it, as I assumed (incorrectly) that someone else would have gotten it by the time my comment went up.

But there it was, at the top of the pile.

Now, when I hear someone say they have no regrets in life, I can only sigh and sadly look down at my feet.

Re:First for the first time!

[Anonymous Coward]

Epic fail.

Re:First for the first time!

[Anonymous Coward]

It's more challenging to be the... ... LAST POST!

Re:First for the first time!

[mqduck]

I never had any desire to get a first post, but I noticed once that I was reading a story without one, so I gave it a shot. "First post!" I said, and I was right!

Won me some negative karma of course, but whatever. I'm glad I did it, so one day I can tell my grandchildren that I once got a First Post on Slashdot, who I suppose will somehow manage to care even less than my friends do now.

Re:First for the first time!

[mqduck]

By the way, first person to say that my having successfully made a First Post guarantees that I won't have any grandchildren gets hit.

Re:First for the first time!

[AndGodSed]

I wonder if there is a first post achievement? I recall getting some first posts on stories a few times, and like you I never "frist prost" ed, but a first post achievement would be pretty darn cool.

early gnome

[confused one]

I remember gnome when it first arrived on the scene. I seem to recall testing it around late 1998 on a workstation. Definitely not the same as current stuff.

Re:early gnome

[Anonymous Coward]

Definitely. Early gnome was definitely snappier to use. I migrated away from Gnome in the 2.28 release due to lag in Alt-Tab'ing. Moved over to fluxbox-based environment. Couldn't be happier.

Re:early gnome

[Anonymous Coward]

In many ways it was better; people don't use desktops, they use applications. I suspect that many of the early users stopped using Gnome after suffering 2.0. I switched to blackbox, flux and even enlightenment for a period. Finally I settled on XFCE4 and though heavyweight for my needs (terminal, web browser and graphics apps) I can at least configure it to get the hell out of my way.

Some of the Gnome libs are useful but GConf, GStreamer and the other interdependent crud are simply an annoyance.

Re:early gnome

[Homburg]

GConf, GStreamer and the other interdependent crud are simply an annoyance.

Unless you want to write apps that store a configuration, or handle sound or video.

Re:early gnome

[jonadab]

Not the same, that's for sure.

Personally I liked Gnome 1.x a good deal better than I like the 2.x series.

Except for gnome-terminal. The newer versions of gnome-terminal are better.

But everything else is worse. More dependencies that shouldn't be necessary, worse performance, more emphasis on completely pointless features like the ability to use the file manager as a web browser (WHY would I EVER want that?) but fewer *useful* features (like, the ability to have an always-on-top panel of a particular size in a particular position, which was great for stuff like having a clock just to the left of where the minimize button was on maximized windows), more gratuitous bug-the-user annoyances (like dialog boxes asking you stupid questions and/or unasked-for windows popping up voluntarily every time you connect a USB device or insert a disc), more undesirably arcane Windows-esque stuff (like gconf), more effort required to get the theme the way you like it, and some things you just plain *can't* do, or I have not figured out how (like, changing the icons on the built-in feature buttons on the panel for things like logging out; in 1.x this was as easy as changing the icon on an app launcher).

If Gnome 1.4 were compatible with modern software (both directions: modern versions of the software it requires, like libraries, and, going the other way, modern versions of applications), I'd still be using it. It was good. I have no idea why they decided to screw it up so much. Gnome 2.x comes across as a bad sequel or a poor remake. It is inferior in nearly every respect.

I can't say I'm very excited at the prospect of Gnome 3.0. What features are they going to take away now, the foot menu and the ability to have a clock on the panel? And what are they going to add? A useless 3D "walk through" filesystem animation like in Jurassic Park, which activates automatically every time a filesystem is mounted? Fixed-size desktop-bound "gadgets", like in Windows Seven, which are strictly inferior to panel applets in every way? Take your time, guys, take your time. I'm in no hurry to upgrade.

Re:early gnome

[Grishnakh]

Storing a configuration for an application is a good thing. Doing it with a binary-only registry, a la Windows, is not.

As for gstreamer, Linux audio is really a big mess right now unfortunately, and gstreamer is just one cog in that machine. The whole thing, from ALSA to gstreamer to PulseAudio to Phonon to everything else needs to be completely re-thought and re-architected.

Re:early gnome

[Anonymous Coward]

HelsinkiSyndrome(TM)

Re:early gnome

[DiegoBravo]

> more emphasis on completely pointless features like the ability to use the file manager as a web browser (WHY would I EVER want that?)

Yes, for some reason that kind of "resource namespace abstraction" seems to be some kind of Holy Grail for a lot of developers who assume users will love it; the same thing happens with Kde's Konqueror, year after year, despite 99.99% of the web sites needing something heavy like Firefox et al. and 99.99% of the users desperate due to slow/unresponsive basic file browsing (as provided by Nautilus.)

I understand the font issues for GTK 2, but as the parent, hope this time Gnome 3 wouldn't need a hardware upgrade.

Re:early gnome

[Homburg]

GConf isn't a binary-only registry, it's XML files stored in a directory structure. More importantly, it's a library that provides a convenient way to update and monitor the information in these files.

Linux audio is a bit of a mess, but the mess is due to there being lots of different ways to access the sound hardware (OSS, ALSA, PulseAudio, Jack, whatver else there may be). GStreamer doesn't really contribute to that mess, as it's at a higher level; if you standardized on, say, pulseaudio, you'ld still want something like GStreamer to handle file formats and codecs.

Re:early gnome

[Anonymous Coward]

If Gnome 1.4 were compatible with modern software (both directions: modern versions of the software it requires, like libraries, and, going the other way, modern versions of applications), I'd still be using it. It was good. I have no idea why they decided to screw it up so much.

Maybe it's a case of the second system effect or maybe Miguel was already getting Swiss bank account deposits from Microsoft. :-)

Re:early gnome

[oakgrove]

Poke around in ~/.gconf and you'll find that the configuration files aren't binary at all as they're actually just simple xml files that can easily be edited by hand if you want to go that route.

Re:early gnome

[steveha]

the ability to use the file manager as a web browser (WHY would I EVER want that?)

Wait, does GNOME even have that? I just tried it in Ubuntu 10.04 Beta 1, and it doesn't work.

On the other hand, KDE has had that for years; Konqueror. (Google it; the top hit says "Konqueror - Konqueror - Web Browser, File Manager - and more!")

I agree with you that I am fine with the file manager and the web browser being two different tools. I guess I don't care if they are merged, but I don't view it as a feature.

As for most of the rest of your complaints, I can't feel much sympathy. When I plug in a USB hard drive, I like that a notification pops up. There is plumbing for you to control what happens; right there in the popup dialog you can choose what you want it to do, and then choose "always do this". At a deeper level, there are config files that let you absolutely control what happens when you plug things in. As for GConf, I like it: it takes every good thing about the Windows Registry and ditches the bad stuff. (Good: a daemon manages the settings and avoids race conditions when multiple processes want to update the settings; also good, the settings are stored in XML text files. Bad stuff in Windows: an opaque binary database format that is prone to corruption... yuck.)

I can't say I'm very excited at the prospect of Gnome 3.0.

Nor am I. As far as I am concerned, Gnome 3.0 should look pretty much exactly like Gnome 2.30. They will be removing all deprecated technologies; things like Bonobo will be completely gone. That's worth bumping the major version number. But I don't want an exciting new paradigm. (I am not happy about the direction Ubuntu is taking, either.)

steveha

Re:early gnome

[broken_chaos]

Oh man, nostalgia. I loved Gnome 1.4. Particularly gmc (the old file manager). Simple, clean, and fast -- Nautilus was so terrible initially that I made several efforts atreplacing it with gmc. I don't remember if I was ever sucessful, as it was years ago.

Sure, it got better over several releases (I remember ~2.6-2.8 beginning to be usable again, I believe), but I never have liked new-Gnome as much as old-Gnome -- though XFCE is a somewhat reasonable 'replacement' for it these days.

Re:early gnome

[BrokenHalo]

...seems to be some kind of Holy Grail for a lot of developers who assume users will love it...

Unfortunately, too much of Gnome development has been driven by a craniorectal chasing of holy grails and other mythological phenomena. Which is not to say I'm knocking Gnome particularly: I have been a user of it since about 1997, when it was very rough around the edges. KDE was then a bit ahead of Gnome in terms of stability and features, but I never got to like it much, and forays into experimenting with that and other desktop environments always led to me returning to Gnome.

Version 2.0, when it came out was a mixed blessing: the eye-candy was prettier than 1.x, but it came with a swath of features that were broken or removed outright "for our own good" (read someone doesn't think it's trendy to work that way any more). And it was an absolute bastard to build, with its labyrinthine freight of circular dependencies. The Dropline Gnome distribution for Slackware saved my sanity with a series of well-maintained builds. Nowadays I use a mongrel setup with Compiz-Fusion with Gnome, along with a few screenlets to help productivity along.

Re:early gnome

[Anonymous Coward]

Don't know about the rest of the stuff (haven't even tried using Nautilus as a web browser, haven't used old versions of GNOME so no idea how the performance has changed), but in my nice and shiny Ubuntu installation, the only notification for most USB devices is a little unintrusive bubble in a corner of the screen that goes transparent and can be clicked through when the mouse hovers over it. Also, that little window popping up every time you insert a storage device (action selector for a CD, file browser for a flash drive, etc.) is actually very convenient (although that's kind of subjective)

Re:early gnome

[Draek]

I hate using Apple zealot drivel to defend a Linux project but... maybe it just isn't for you? Gnome with the 2.0 version decided to cater more to new users (read: Windows rejects) and leave the 'hardcore Unix nerds' that had been its main market until then to more niche WMs, and I for one can't blame them for it.

Every one of us wants everything we need and nothing we don't, and Gnome 1.x was getting particularly problematic at the latter with everything and the kitchen sink thrown in somewhere, and each one of those features was a necessity to at least one user while being unnecessary bloat to at least one another. Remember, KDE was considered the *simpler* desktop of the two back then.

So, try one of the smaller desktops. Perhaps WindowMaker is for you, perhaps it's Openbox, perhaps you're like me and it's IceWM, or perhaps you're just weird and twisted and you'd prefer Ratpoison with a bunch of terminals running Emacs ;) There's no shame in using a "non-mainstream" desktop, and by taking a bit of time to find the perfect one you can increase your comfort in your day-to-day usage a whole lot. One thing is clear though: if you aren't a normal "luser", your frustrations with Gnome are only gonna get worse from now on.

Re:early gnome

[Grishnakh]

Ok, maybe I was wrong about the binary bit, but XML is anything but "easily" edited by hand, at least if it's the packed variety that doesn't have any carriage returns. Ever try editing an ODF document by hand?

Are the config files separate for each application? With KDE, all the config files and all other related files are in a separate directory, under ~/.kde/share/apps/. This makes it easy to find and edit stuff by hand.

Re:early gnome

[oakgrove]

Actually, I tend to agree with your opinion on editing XML by hand. Not nearly as fun as just a plain flat file. The XML files for each application in .gconf are organized in directories, one for each application. So, if you go into the totem directory for example, there will be one XML file in there to fiddle with that surprisingly enough is laid out well enough with carriage returns, etc. The only real problem there is there don't seem to be any comments to go off of so you have to know exactly what to do.

Re:early gnome

[Anonymous Coward]

I'l bite...
1)
You have never been able to use the webbrowser in the nautilus...

2)
Think of dependencies as planned libraries that are designed to be shared among applications. A dependency isn't necessarily a bad thing, it might just be proper software engineering.
But people can see it in the package system, what they can't see is the code.. so nobody complains about that.

3)
A world with no gconf (note it ins't mean for application state) would be worse. You just haven't experienced it. Yes the implimentation is kinda of dated. But that stuff is windows-esque isn't (Which in implementation/scope certainly isn't) is a argument. MS does make good software,and some crap as well just like everyone else.

There are many more things that are completely wrong, so either you really are clueless or just trolling...

Re:early gnome

[muckracer]

> you'd prefer Ratpoison with a bunch of terminals running Emacs ;)

Emacs IS my desktop, you insensitive clod!

Re:early gnome

[moonbender]

As long as it's related to file management, the resource namespace abstraction seems sensible. I like being able to access FTP using the file manager and it's useful to do the same with all kinds of file management protocols, ie SFTP, WebDAV, etc. HTML rendering shouldn't be part of it, though, and apparently it isn't, in Gnome.

Re:early gnome

[jandersen]

I'm glad to see that I am not the only one taking issue with GNOME.

It is sad, really; in the beginning it really looked very promising - I liked a lot of the philosophy, but then it became mired in this Windows aping and dumbing down that I find so condescending. Now I use KDE - definitely better in terms of practical usability. I don't have much use for intrusive gimmicks and missing functionality. Would I return to GNOME one day? I'm not sure - it is a question of trust, I think. It is so easy to break and so very hard to build up again.

Re:early gnome

[X0563511]

Isn't there an XML "version" of tidy? That would clean up the formatting. Put that with an editor that does proper highlighting and you should be good.

If you can edit HTML, you can edit XML. Just be more careful.

Re:early gnome

[CarpetShark]

Personally I liked Gnome 1.x a good deal better than I like the 2.x series.

Except for gnome-terminal. The newer versions of gnome-terminal are better.

I tend to agree -- gnome 1.x was much cooler in some ways. Gnome-terminal still sucks compared to rxvt or konsole sadly. One thing I was REALLY glad to see in 2.x was the pervasive unicode support from pango, and the decent anti-aliased font support. That stuff, at least, deserves some serious credit.

Oh, and thank god GNOME was easily installable when enlightenment never moved on, and KDE 3.x basically imploded in the name of 4.x "improvement".

Re:early gnome

[IceFreak2000]

FYI, tidy itself can do the job; just use

tidy -xml

Failing that, the excellent xmlstarlet is a great tool to have around - it's useful for formatting, querying or even editing XML documents straight from the command line

Re:early gnome

[jonadab]

> > the ability to use the file manager as a web browser
> Wait, does GNOME even have that?

It was at one time purported to be one of the major "selling points" of Nautilus (over more traditional file managers like gmc), but upon closer inspection it appears to have been removed at some point since, probably because nobody was using it.

I never noticed its removal, probably because I never used it either.

So that particular complaint is obsolete.

> On the other hand, KDE has had that for years

Yeah, I know about Konqueror: "What do you mean you want to upgrade your web browser without upgrading your entire desktop environment? Modern websites you say? Bah! And don't even think about reporting any bugs in the CSS handling if you're running the latest bleeding-edge dev build. What do you mean you don't want to run bleeding-edge dev builds of your whole desktop environment? All the core devs are doing it, so you should too!"

As a web developer, I am not particularly fond of Konqueror.

> As for most of the rest of your complaints, I can't feel much sympathy.
> When I plug in a USB hard drive, I like that a notification pops up.

Don't get me wrong: I don't care what happens by default. I just want the ability to turn it off.

> There is plumbing for you to control what happens; right there in the popup
> dialog you can choose what you want it to do, and then choose "always do this".

Nice theory, but in practice it doesn't work. You still get bothered again next time.

> things like Bonobo will be completely gone

Can't say I'll be sorry to see that go.

(What does bonobo even *do*, besides complicate the dependency tree?)

Re:early gnome

[jonadab]

> Nautilus was so terrible initially that I made several efforts at replacing it with gmc.

I used to just chmod ugo-x `which nautilus`

Over the years, as hardware has gotten better and other software has increased its requirements, Nautilus no longer seems like such an extremely terrible waste of system resources. I mean, it still seems kind of silly to leave the file manager running 100% of the time just to draw wallpaper and desktop icons, but whatever. With all the other software I'm running, it's a drop in the bucket these days.

I'm still not really fond of it, though. Truth be told, I do most of my file management on the command line. It's just easier. All the handy shortcuts (like tab completion and wildcards) don't work so well in a GUI file manager.

Re:early gnome

[jonadab]

> So, try one of the smaller desktops.

Actually, I'm leaning toward a custom session consisting of sawfish and gnome-panel. Most of the actually *useful* features of Gnome can be had just by running the panel (and individual apps launched as needed, though the terminal is the only Gnome-branded app I use very much) in conjunction with another window manager. I'm not sure I need the rest of Gnome. And unlike most of Gnome 1.x, sawfish has been maintained and still works just fine with modern stuff.

Setting up a custom desktop session is surprisingly easy. You only have to edit a couple of files...

Re:early gnome

[PoprocksCk]

Oh man, nostalgia. I loved Gnome 1.4. Particularly gmc (the old file manager.

Not to be a "Nitpicky Sue", but wasn't 1.4 the first release to include Nautilus?

Re:early gnome

[Anonymous Coward]

100% agree.

I remember when a old computer that perfomed poorly on windows, run fast and smoothly on linux/gnome-1.4

Now gnome it is slow, fat and ugly....(without mentioning cut & paste ideas from macosx and windows).

Re:early gnome

[jonadab]

> One thing I was REALLY glad to see in 2.x was the pervasive unicode support

Oh, was that new in 2.x?

At the time, I wasn't doing anything that required Unicode, so I didn't notice any changes there. All the foreign languages I'd studied up to that point had small alphabets or abjads and so were easily supportable using nothing but ASCII characters and transliterative fonts.

But yeah, I'm using Unicode now (studying Japanese, presumably because I'm some kind of glutton for pain or something), so that's good.

> and the decent anti-aliased font support.

That's Gnome's doing? I was under the impression that was an X.org improvement. Some spiffy new rendering extension called xftt, or something like that.

Re:early gnome

[X0563511]

Oooh, that one slipped under my eyes. Nice!

Thanks!

Re:early gnome

[Anonymous Coward]

GLib also supports keyfiles which are the correct choice unless you _need_ hierarchical configuration data. In fact GConf probably abstracts the GLib path constants and XML parser. What is needed is not a library but a simple LSB appendum to the effect that all user configuration data should live in ~/.rc/$appname or ~/.config/$appname

Admit it, there's zero point to GConf, it's a non-solution to a non-problem!

GStreamer does nothing because I use mplayer and VLC. There's nothing I love more than having to compile my own apps because so many packages link against GStreamer by default. Which morons are assuming that users of desktops other than Gnome want something as useless as GStreamer installed?

Re:early gnome

[steveha]

Nice theory, but in practice it doesn't work. You still get bothered again next time.

You sound like an advanced hacker type. Google search for "Linux USB hotplug scripts" and read up. You can control what happens when you plug a device in, and you can choke off the behavior you don't like before GNOME is ever notified that something was plugged in. (I haven't bothered to research this in detail because, as I said, I'm okay with the default behavior.)

What does bonobo even *do*, besides complicate the dependency tree?

Bonobo was a way for GNOME applications to expose functionality and data to each other. It is very similar to Microsoft's OLE, or KDE's KParts. The replacement for Bonobo is that the whole world is using DBUS now. (DBUS isn't just for the kernel to publish notifications to user space; user space processes can also use it to talk to each other.)

http://library.gnome.org/devel/platform-overview/stable/bonobo-corba.html.en
http://library.gnome.org/devel/platform-overview/stable/dbus.html.en

steveha

Re:early gnome

[atamido]

also good, the settings are stored in XML text files. Bad stuff in Windows: an opaque binary database format that is prone to corruption... yuck.)

People really need to stop mentioning "prone to corruption" in regards to the Windows Registry. That was true in the Windows 9x days, but I know I haven't seen a corrupted registry in years. I suspect I haven't seen one since 2000/XP (excluding the 9x based ME).

There are also benefits to a binary format, such as speed. Storing hundreds of thousands of settings in XML isn't exactly ideal. If it were, SQLite would use it. The Windows Registry also allows advanced per-key permissions, only some of which can be worked around using file permissions (although admittedly I've only seen complex permissions used to work around really poorly designed programs in Windows).

Still, there are areas where it falls short. If you're going to make a complex database like system for settings, you might as well abstract it so that a user's registry settings can be accessed from anywhere on the network using something like SQL commands. It would also make sense to use a common storage format. Microsoft has had decades to work on good database designs, and they have some really good ones. But instead of reusing their good designs, they reinvent the wheel every time they want to store a little data. (Mozilla realized this for themselves, and switched from their bizarre in-house-developed Mork to SQLite.)

Still, if you need to fix a bizarre issue without a GUI, it's nice to be able to open an XML file in your editor of choice and poke around for the problem.

Re:early gnome

[Anonymous Coward]

> the ability to use the file manager as a web browser (WHY would I EVER want that?)

Why wouldn't you? This is a genuine question.

Re:early gnome

[CarpetShark]

[Anti-aliased fonts are] Gnome's doing?

I believe it spanned a lot of the graphics stack (so yes, involved X drivers, X itself, XRender, etc.). GTK 1.x attempted anti-aliasing with hacks, but it was fundamentally limited and caused artifacts. For 2.x, this was redone (largely thanks to pango and freetype, IIRC).

Re:early gnome

[Interoperable]

No you get off my lawn!

Seriously though, Gnome is considered a heavyweight desktop these days. If you don't like the direction it's taking, try XFCE or one of the true lightweights like Fluxbox.

thats right folks

[chibiace]

this could be the last good release if it goes bad. i didnt like some of their ideas for 3 but im sure they will try to please us.

Re:thats right folks

[Steve Max]

IMHO, 1.4 was the last good release (as in "I'd use it over anything that was available at the time"). Right now, KDE3.5, KDE4.4, OS X and even Windows 7 offer more than Gnome. I jumped from Gnome to KDE right after the train-wreck that was Gnome 2.0; I've tried Gnome every couple of releases after that and it definitely improved a lot since 2.0, but everything else also improved a lot. Maybe 3.0 will be better (and I will try it, despite the current version of Gnome-shell on Debian Sid being awful), but I don't have much hope.

isnt gnome smth like dead technology??

[Anonymous Coward]

bonobo is dead... they are using spinoff of kde tech (dbus, khtml, etc)..

Re:isnt gnome smth like dead technology??

[DarkKnightRadick]

I remember using GNOME in the late 90s, and if bonobo is dead than it's a good thing. That was a nightmare to mess with.

Re:isnt gnome smth like dead technology??

[Anonymous Coward]

Bonobo, Miguel's OLE knock off.

That guy just can't get enough Ballmer cock.

Re:isnt gnome smth like dead technology??

[DarkKnightRadick]

Like I said, good riddance to bad rubbish.

To be honest I really don't care anymore. I've been using Windowmaker for about 5 years now.

Re:isnt gnome smth like dead technology??

[Homburg]

dbus was actually initially developed for GNOME; other GNOME-originated libraries that are still in use include GVFS, gconf, and gstreamer. And GTK, though of course not originally developed from GNOME, is now developed in parallel with GNOME (much of the stuff in the depracted GNOME libs is no longer needed because GTK itself provides solutions for the same problems). So I think GNOME is still a pretty vibrant development platform.

Sounds like a KDE-type cleanup

[slimjim8094]

If this is done properly, I think it'll be good for GNOME. From where I sit, they sound like they're shooting for a major architecture redesign. In other words, this 2.30 release is analogous to the 3.5 releases of KDE.

And I think starting largely from scratch will be a net benefit. I've never personally used GNOME (though I've recommended it to others) and I've found it to be technologically lacking compared to KDE (KParts and KIOSlaves are awesome, and while there are GNOME counterparts they aren't as used).

One thing I think GNOME does very well is their HIG - probably the best outside of Apple. The new release is very simple - dump a lot of legacy code and keep the HIG. Maybe drop the old-fashioned look too.

Though my fantasy is to see them use Qt.

Re:Sounds like a KDE-type cleanup

[Anonymous Coward]

Gnome using Qt? My bet is that pigs evolve into flying creatures long before that happens, even though Qt is far supperior technically. Back when they started, the Gnome guys didn't have a lot of choice. Qt had its own license so it was either this not quite free stuff (in the old millenium) or gtk+, which was LGPL from the start. They took the later and now they're stuck with it for better or worse.

Well, if Gnome ever does switch, maybe I'll seriously give it a try then, though.

Re:Sounds like a KDE-type cleanup

[Mr Thinly Sliced]

Though my fantasy is to see them use Qt.

I take it you've never compiled any KDE applications .-)

I'm kinda joking, but the compilation time for complex kde stuff is humongous. It's one of the few things I quite like about Gnome stuff - it's pure C and is real rapid to build.

Hehe, yes, I'm a gentoo ricer.

Re:Sounds like a KDE-type cleanup

[cheesybagel]

It's a matter of taste. Personally I hate Qt's slot mechanism. And Moc. IMO the problem with GNOME is not GTK+, it's Mono.

Re:Sounds like a KDE-type cleanup

[neiras]

KIOSlaves are awesome, and while there are GNOME counterparts they aren't as used.

One neat thing about GVFS, the GNOME abstraction, is that part of it wraps FUSE filesystem modules. Any application, not just GNOME applications, can use filesystems mounted with GNOME's 'connect to server' feature, for instance. I think it's more desirable to write a FUSE module than a KDE-specific KIOSlave.

GNOME sometimes comes across as a hodgepodge of bindings and semi-coherent libraries, but there has been a great deal of work to consolidate and even eliminate core libraries, tighten up coding standards, get rid of deprecated symbols in GTK+ and GLib... At least they're trying to get things right, right up and down the stack.

GNOME 3 will be a big shift. I can't say I'm crazy about the new shell, and the Task Pooper scares the shit out of me (ha ha).

They'd have to screw it up really badly to make me go back to KDE. Even then, I'd go to 3.x.

Re:Sounds like a KDE-type cleanup

[MrHanky]

The KDE type cleanup is what they did for 2.0, which was what made Linus Torvalds say "fuck this shit, I'm switching to KDE" (and, incidentally, what made him say "fuck this shit, I'm switching to Gnome" after trying KDE4). It pissed off a lot of other users as well. Of course, Gnome 2.0 was a bit more stable and less bug-ridden than KDE4, but on the other hand it had almost no features you'd expect from a computer (which was supposedly 'good for you', according to the HIG apologists, pretty much like the absence of multi-tasking on the iPad until yesterday), and took several years before it was as useful as 1.4 (the last version I used).

I forget. Did I have a point with all this? Oh, yes, the cleanup: it sucked the last time, and I hope they manage it better now, or they will probably hear it until the next time some huge project mismanages a major revision. On the other hand, maybe a botched Gnome3 release will help KDE get the recognition it deserves again.

Re:Sounds like a KDE-type cleanup

[zach_the_lizard]

I much prefer working with Qt on my little projects, so I don't mind the longer compile times so long as they save me some dev time. Except when I run Gentoo. Then it matters a lot.

Re:Sounds like a KDE-type cleanup

[mrmeval]

"Over stupidify" defines guhnome but it's not as bad as KayDeeE

Re:Sounds like a KDE-type cleanup

[slimjim8094]

I love signals and slots. They require a bit of a different way of thinking, and a semi-proprietary compiler (it's open source but still).

It's really the first time since Visual Basic where I think a language has really 'gotten' event-driven programming. Everything else has you writing your own event loops to switch on a message type. Signals/slots let you use a single statement as a patchboard. It's the reason they can have an example where a slider changes a text box in one line of code.

Is it different? Sure. It's slightly different than straight C++, but not by much. It definitely demands a new way of thinking about how to program graphical applications. But if you can manage it, I think it's far superior.

Though I'd also agree that Mono's crapulence is Gnome's biggest problem. I don't want the whole damn framework for some note-app

Re:Sounds like a KDE-type cleanup

[Mad Merlin]

It's a matter of taste. Personally I hate Qt's slot mechanism. And Moc. IMO the problem with GNOME is not GTK+, it's Mono.

I'd say it's Mono, to a lesser extent GTK+, and to a greater extent the fetish of removing any and all features.

Re:Sounds like a KDE-type cleanup

[MrHanky]

Looks like it over-stupidified your comment as well. Tough luck!

Re:Sounds like a KDE-type cleanup

[Anonymous Coward]

I love signals and slots. They require a bit of a different way of thinking, and a semi-proprietary compiler

I'm not a C++ programmer and not very familiar with Qt, but is MOC really needed when there's something like this available?

Re:Sounds like a KDE-type cleanup

[Rozine]

Any application, not just GNOME applications, can use filesystems mounted with GNOME's 'connect to server' feature, for instance. I think it's more desirable to write a FUSE module than a KDE-specific KIOSlave.

You say that, but if you look at the apps on the ground, most, well, don't. For instance, I can mount my network drive using GVFS all I want, but I still can't watch anything on it, since VLC has no clue about it. A nice idea, but it would be preferable if they integrated it into the rest of Linux too. That said, I use Gnome every day.

Re:Sounds like a KDE-type cleanup

[nxtw]

I'm kinda joking, but the compilation time for complex kde stuff is humongous. It's one of the few things I quite like about Gnome stuff - it's pure C and is real rapid to build.

Is that the fault of KDE/Qt or is it a result of a slow C++ compiler?

Re:Sounds like a KDE-type cleanup

[d-Orb]

Though my fantasyThough my fantasy is to see them use Qt.

Well, you already have KDE for that don't you? :-) What I really wish they sussed out once and for all, is freedesktop.org, so you can use whichever desktop you want with whatever tools you want, and it all works. They already co-operate a bit, but I'm not entirely sure how deep it goes in many ways.../p

Re:Sounds like a KDE-type cleanup

[Anonymous Coward]

When I read your comments I think lavender peril!

Re:Sounds like a KDE-type cleanup

[bmo]

Did you read the "Credit where credit is due: QT" paragraphs?

I think you should.

Without Qt's moc, she would have never been inspired to make her own version of moc.

--
BMO

Re:Sounds like a KDE-type cleanup

[Homburg]

is MOC really needed

No, basically. But back in the early nineties, when Qt was first developed, the various C++ features that make these pure-C++ signal and slot libraries usable weren't widely available.

Re:Sounds like a KDE-type cleanup

[Homburg]

I can mount my network drive using GVFS all I want, but I still can't watch anything on it, since VLC has no clue about it.

Yes, you can, that's what the parent means about the FUSE integration. GVFS mounts show up as directories accessible by all applications at ~/.gvfs . VLC doesn't need to know anything about GVFS to access these files.

Re:Sounds like a KDE-type cleanup

[Anonymous Coward]

Or the result of C++ being fundamentally difficult on the compiler?

Re:Sounds like a KDE-type cleanup

[nxtw]

Or the result of C++ being fundamentally difficult on the compiler?

Some valid C++ code indeed requires more work to compile (if it can be compiled at all), but Qt? It does not use features like exceptions or templates...

And GCC C++ is known to not be the fastest C++ compiler...

Re:Sounds like a KDE-type cleanup

[AchilleTalon]

Okay, if I understand the way you view things, you would use Gnome if it would have been written to be KDE and the next guy using Gnome would use KDE if it has been written to behave as Gnome. Nice!

Each time KDE or Gnome appear in a discussion, we inevitably refrain the same old things about how nice one is compare to the other. Or, if the guy is polite like you, he says he doesn't use it, but he recommends it to others.

The point in the original discussion is about how Gnome has changed over a decade compare to... itself.

Thanks for reading.

Re:Sounds like a KDE-type cleanup

[chgros]

I love signals and slots. They require a bit of a different way of thinking, and a semi-proprietary compiler (it's open source but still).
I'm not against the concept, but, really, stringizing function names? So if you type the name wrong you only get an error at run-time. Bye-bye type safety!

Re:Sounds like a KDE-type cleanup

[Anonymous Coward]

The only part that of the official gnome desktop that uses mono is tomboy... for that single use it's kinda of bloated (boot/login time hurts). But there are plenty of alternatives. Even a direct port in c++ (gnote).

Re:Sounds like a KDE-type cleanup

[Anonymous Coward]

I remove mono after a new install of Fedora. No problems.

Re:Sounds like a KDE-type cleanup

[lolocaust]

Qt is easily the best toolkit and api I've ever worked with. I've been saying for a while that Gnome 3 should ditch gtk and switch to Qt, and not just so that apps will look/feel better in another environment.

Re:Sounds like a KDE-type cleanup

[DrXym]

GNOME has sometimes lost its way with usability and I have my concerns about Mono and GNOME Shell in 3.0. But I have no doubt that it would trounce KDE in usability tests if the two were put head to head with a range of users. Even power users. It's just a more simple, cleaner, task oriented and better designed experience.

KDE 4 just has far too much complexity going on, not just compared to GNOME, but to Windows and OS X too. In particular the settings dialogs are WAY too complex with advanced, esoteric and simple settings all mixed together. Often you'll open one settings dialog to discover it contains a tree of subsettings each with their own panes of tabs and even secondary dialogs with even more settings. I thought KDE 4 was going to fix all this but it's as bad as ever. Konqueror (for example) has SIX "Configure..." menu items under its settings menus that each lead to dialogs with pages and pages of more options. Even Netscape Navigator was never this bad, and the likes of Firefox, Chrome (which also uses WebKit) demonstrate there is no excuse for this either.

Sometimes less is more. Even if KDE advocates like having all those settings, why can't they shove most of the esoteric & advanced ones out into some kind of power tool to keep things simple? There is no excuse the way it is and I suspect GNOME's success is testament to that fact.

Re:Sounds like a KDE-type cleanup

[dr.newton]

F-Spot uses Mono. I'm not sure if it's part of the "official gnome desktop", but that is one excellent application, and worth installing Mono for.

Personally, I have my doubts about Mono in terms of providing compatibility for .NET apps. However, I really couldn't care less what the app uses on the backend as long as it works well, even if it adds a couple hundred megabytes to the install size. My Gentoo days are over. ;)

Re:Sounds like a KDE-type cleanup

[lilo_booter]

Agreed and it works quite well... except... 1) why put the damn things in a hidden directory?, 2) would be nice if there was a way to ensure they remounted on a restart automatically and 3) the memory management is shockingly bad (at least with the smb implementation - typically see it shoot to 3gig over a couple of hours of listening to mp3 and looking at pictures). Yeah, OK, should report the latter one as a bug, but not got round to it yet. Basically, it's a nice idea, but the implementation makes it slightly less than the useful.. shame really - back to hacking /etc/fstab.. sigh..

Re:Sounds like a KDE-type cleanup

[snadrus]

Agreed! I revisited KDE @ 4.3 and was stunned that they hadn't dropped KIOSlaves for something GVFS-like.

I can edit a file in a zip on a samba share all remotely via command-line file tools or by right-click in Nautilus. It can even be a huge file over a slow connection as I only pull down what my editor needs.

Gnome Desktop

[Nerdfest]

You can try it out without any harm. It's ... interesting. It is actually quite usable and does give a lot of features that will be quite nice for people that use multiple desktops. I tend to use a single desktop and gnome-do to provide quick access, so I find it gets in the way a bit. There's a few things I don't like about it that I can't change, like the panel at the top of the screen. I prefer it at the side instead, as my netbook and laptops both have decent horizontal resolution but crappy vertical. Perhaps they'll fix that.

I haven't tried 2.3, but I'm sure it will be minimal incremental improvement over the previous version. Nice to see that they're trying new things ... as long as they leave the damn window controls where they belong.

Re:Gnome Desktop

[Anonymous Coward]

ummm.... you can put the panel where ever you want.

Re:Gnome Desktop

[Nerdfest]

Not in 3.0 from what I've seen. Likewise with the sidebar ... it's currently locked in position as well.

Re:Gnome Desktop

[bragr]

The wonders of click and drag. I usually keep it at the top and the windows taskbar at the bottom. I'd get confused otherwise because if I am using one of them, I have the other open in a VM or remote session.

Re:Gnome Desktop

[Anonymous Coward]

2.30 and 3.0 (using gnome-shell) are still quite different, right?
I'm not sure about the plans for 3.x, but before that the panels should not be locked.

Re:Gnome Desktop

[Nerdfest]

I should have been more specific in the parent post ... I was referring to Gnome Shell.

Re:Gnome Desktop

[drinkypoo]

Not in 3.0 from what I've seen.

You can lock the panels to prevent movement in 2.x, so probably there's either a GUI or gconf setting to permit you to tamper with its location. I hope. I kill off the panel and use AWN so I don't care too much.

Likewise with the sidebar ... it's currently locked in position as well.

Dunno about that one.

Re:Gnome Desktop

[Anonymous Coward]

I particularly like how they have "improved" GNOME 2.30 by using the same icons for text files and unknown file types. They have significantly re-done the gnome icon theme. Text files used to preview the file contents, but that no longer functions with the gnome default icon theme.

Also, out of the five buttons in Firefox, back, forward, reload and home are now transparent/light grey, and although the stop button is red, it's usually is greyed out, so they're all grey.

Perhaps that could simplify and improve things further by using the same dull grey icon for all file types, and use some dull grey "generic button" icon for any remaining buttons that somehow haven't had their icons removed yet.

I haven't tried gnome-shell yet, but somehow I'm not really looking forward to it. Every new version seems to be aimed at pissing off existing users, to improve things for people who haven't used it before.

pffft... fvwm is 2.4.x already...

[Anonymous Coward]

2.30 is totally overrated... 3.0 would be buzz...

Oh good!

[bmo]

Can we kill off Gnome 3.0 too? How about getting rid of GTK+?

We especially need to get rid of Mono. Funny how Gnome is less free than KDE now.

Admiral Ackbar says "It's a trap!"

--
BMO

Re:Oh good!

[Anonymous Coward]

Your official freetard party line bleatings are tired.

www.tmrepository.com

Re:Oh good!

[shutdown -p now]

LGPL is less free than GPL, really?

Re:Oh good!

[Anonymous Coward]

I resemble that remark.

Powered by Django

Heh. Freetard win.

Re:Oh good!

[bmo]

It's called the "Lesser" GPL for a reason.

Also, unless you get mono exclusively from Novell, you are (potentially) infringing on Microsoft intellectual property. And Gnome has been adopting mono like it doesn't matter.

Yes, Gnome is less free now. Gnome fans totally miss the irony.

--
BMO

Re:Oh good!

[jjohnson]

TM Repository is officially the saddest, most pathetic website I've ever seen: A tiny community of people who get together just to snark at Linux propaganda. It's like setting up a site to mock the CPUSA.

Re:Oh good!

[Anonymous Coward]

DjangoIsFreetardTechnology(TM)

Re:Oh good!

[Anonymous Coward]

Sorry sir, but clearly you have never read the LGPL. The "Lesser" part refers to less rights being retained by the granter. The license gives MORE freedom to the users.

Re:Oh good!

[aztracker1]

FUD, plain and simple... interesting coming from a *nix supporter though.

Re:Oh good!

[bmo]

Oh look, it's a BSD troll.

How cute.

*pats on head*

Now run along!

GIT!

--
BMO

Re:Oh good!

[Mathiasdm]

Wait, what? The LGPL is less free because it's called that way? The LGPL gives me, as a developer, MORE freedom. It's all a matter of perspective.

Re:Oh good!

[bmo]

What's fud about it?

Novell signed a contract with Microsoft. It indemnified people who got mono from Novell from liability. It doesn't cover third parties.

You go find the clause that covers third parties and get back to me.

--
BMO

Re:Oh good!

[Anonymous Coward]

Your site is a close second.

Re:Oh good!

[Timothy Brownawell]

It's called the "Lesser" GPL for a reason.

Yes, it has less restrictions. (ie, more freedom to do different things)

Also, unless you get mono exclusively from Novell, you are (potentially) infringing on Microsoft intellectual property.

/me rolleyes

Remember a while back they were claiming to have some triple-digit number of patents that the Linux kernel infringes on? Remember IBM warning off that mainframe emulator a couple days ago? I'm not convinced that mono infringes significantly more or stronger potentially-hostile patents than any other similarly complex piece of software.

And Gnome has been adopting mono like it doesn't matter.

I thought they only used it for a couple of trivial/perhiperal things?

Yes, Gnome is less free now. Gnome fans totally miss the irony.

Yes, true freedom is an OS that will refuse to run anything except locally-compiled programs, with a compiler which will only compile code that has the GPL licensing headers.

Re:Oh good!

[MrHanky]

Qt has been LGPL for a few years now, and KDE has always been part LGPL (like WebKit, a derivative of the old khtml).

Re:Oh good!

[bmo]

Remember a while back they were claiming to have some triple-digit number of patents that the Linux kernel infringes on?

Yes, how can one forget?

IBM and TurboHercules debacle

Yes. TurboHercules is suing IBM in a SCO-esque lawsuit. IBM is supposed to take it lying down?

I'm not convinced that mono infringes significantly more or stronger potentially-hostile patents than any other similarly complex piece of software.

It was enough to convince John Dragoon and Ron Hovesepian.

Given the choice between technology that is potentially more infringing than the other, which one would you pick?

I thought they only used it for a couple of trivial/perhiperal things?

For now. Perhaps you've forgotten Miguel's rantings about how it should be used throughout Gnome? Perhaps you forgot that Miguel works at Novell and that Novell has SuSE with which they can "differentiate their product" with mono (depending on how it's used) in a yet to be determined crucial application. This is not tinfoil. This is how companies work.

snippage of insult

Yeah, whatever. Perhaps you forgot that Qt is LGPL?

--
BMO

Re:Oh good!

[bmo]

Tmrepository is entirely unfunny and a lame ripoff of Adequacy.org.

Get some better writers.

--
BMO

Re:Oh good!

[Ant P.]

Kill GTK+ off? Maybe when Qt has more than one decent theme that doesn't require installing half of KDE, or when Qt themes don't require a compiler to create. Until then, get lost.

Re:Oh good!

[stinerman]

That may be true. Frankly, I don't give a rat's ass because Microsoft will never, ever go after an insignificant, individual end user like me for patient infringement.

If you're thinking of GNOME in a business setting or are are distributing Mono, you may want to think twice. However, for as long as Mono exists under a free license, I'm happy to use it.

Re:Oh good!

[moosesocks]

Oh, for Christ's sake. Mono is safe. Microsoft made a legally-binding promise not to sue. No judge would even hear the case if they attempted to bring a suit against <I honestly have no idea who they'd sue> unless the GNOME people managed to infringe upon something else in Microsoft's portfolio.

We have much bigger fish to fry. Squabbling about Theora and Mono isn't a productive use of your time, no matter how valid your arguments might be. The standards have been set.

Re:Oh good!

[Homburg]

You go find the clause that covers third parties and get back to me.

Here you go.

Re:Oh good!

[shutdown -p now]

Fine, so then LGPL is less free then LGPL?

Re:Oh good!

[Grishnakh]

I think the idea is that X + Mono is less free than X.

It still hasn't been proven that Mono isn't a patent trap.

Re:Oh good!

[Grishnakh]

The problem is that they're trying to push Gnome/Mono into all the Linux distros, and Linux is increasingly being used by businesses and governments. If there is a patent trap in Mono (which exists regardless of its license), then that means that MS can then sue all those businesses and governments for patent infringement. Of course, the real idea probably isn't for MS to get money from people from patent suits, but basically to scare everyone away from Linux for once and for all, and only use "safe" MS software.

The reason that the Linux software you now enjoy is as usable and functional as it is is because many businesses and governments have been investing in it, working with it, and using it. If it were some project only used and developed by hobbyists at home, like ReactOS, it wouldn't be good for anything but playing around. Instead, we have an OS and thousands of apps that are all free, and we can use to do just about anything you can do in a Windows environment (and sometimes much more). The only thing we can't do is run some Windows-specific apps, but that's becoming less and less of a concern as more companies make Linux versions of their apps, and as more alternative apps become available or mature (e.g. OpenOffice).

So these issues which seem to affect only the larger players may not seem to affect you personally, but in reality they do.

Re:Oh good!

[shutdown -p now]

It still hasn't been proven that Mono isn't a patent trap.

It hasn't been proven that Qt isn't a patent trap, either. Thing is, to prove that, you have to go through the entire patent database, one by one, and see if any of them are applicable. In fact, you can be almost certain that both Qt and KDE are covered by at least a few, though their validity may be questionable (but then the same is also true for any MS patents that supposedly apply to Mono).

Anyway, this turns out to be irrelevant, since, as someone else pointed out in this thread, Gnome does not depend on Mono in this release.

Re:Oh good!

[bmo]

The contract that Novell signed is not the same as the "promise" that Microsoft published later. There is a difference. Novell is indemnified, and so are Novell's customers and people (e.g., developers) who get mono directly from Novell instead of a third party.

Come on, Microsoft does not like standards and interoperability. They are already undermining their own OOXML by using the proposed standard (the one that passed ecma, but not iso) that was rejected instead of the one that the ISO actually approved.

I didn't fall off the kielbasa wagon yesterday.

Call it tinfoil. I don't care. You're naive.

--
BMO

Re:Oh good!

[bmo]

Yes, but with Qt, you don't have a rabid Microsoft fan bent on directly implementing Microsoft technology on Linux. Trolltech does not have Miguel. Novell does. Judging from Miguel's actions and his words, I think we have something to worry about. I don't think that Miguel is some sort of Manchurian Candidate, but he is driven by his admiration of everything Microsoft.

--
BMO

Re:Oh good!

[siride]

GTK+ themes require a compiler to create as well. Unless you count the color tweaking that you can do with the gtkrc files (something you can do easily through the GUI with KDE and isn't considered a theme). Of course, I don't see why a couple of megs of KDE libs is really a problem unless you are using a ten year old computer (but then it wouldn't be fast enough to use GNOME anyways).

The toolkit/DE zealots really amaze me sometimes.

Re:Oh good!

[bmo]

The MCP only covers the ecma parts.

Anything mono that is not ecma is not covered.

The Novell situation is a whole different kettle of fish.

Since Mono comprises a lot more than just the parts covered by the ECMA standards, De Icaza also announced that Mono will be split in half. "In the next few months we will be working towards splitting the jumbo Mono source code that includes ECMA + A lot more into two separate source code distributions," he explains, "One will be ECMA, the other will contain our implementation of ASP.NET, ADO.NET, Winforms and others."

http://www.osnews.com/story/21784/C_CLI_Under_Community_Promise_Mono_Split_in_Half

Would you want to touch that with a 10 foot pole?

Re:Oh good!

[anshulajain]

GNOME will happily run without Mono. Heck, even in Opensuse you can perform a "Mono-sectomy", remove F-Spot, Banshee, gnome-do, Monsoon (whatever that crap is) etc and still have a perfectly well functioning system.

Re:Oh good!

[Abcd1234]

The contract that Novell signed is not the same as the "promise" that Microsoft published later. There is a difference.

Yes, there is a difference. In one case, Microsoft can't sue because of a contract. In the other, they can't sue because of promissory estoppel. Either way, they can't sue.

But enjoy the hat, I'm sure it's very shiny and stylish.

Re:Oh good!

[bmo]

In one case, only ecma mono is covered.
In the other case, all of mono is covered.

The former is the Microsoft promise.
The latter is the Novell contract.

http://www.osnews.com/story/21784/C_CLI_Under_Community_Promise_Mono_Split_in_Half

There you go.

--
BMO

Re:Oh good!

[Abcd1234]

In one case, only ecma mono is covered.

So only use the ECMA bits (which will soon be factored out), and use the free software stack bindings for everything else, just like, well, basically all existing Gnome Mono applications (last I checked, F-Spot wasn't using winforms).

There, problem solved. Happy now?

Re:Oh good!

[Anonymous Coward]

http://live.gnome.org/GNOME3Myths#GNOME_3.0_depends_on_Mono.21

Re:Oh good!

[Anonymous Coward]

Cute alias Linus. Shouldn't you be working on ext4 or sound or fuse or something? Take your pick, it all needs improving.

Re:Oh good!

[Grishnakh]

Exactly right. The biggest key here is that Mono is Microsoft technology reimplemented on Linux. Qt isn't a reimplementation of any MS technology, so if there are any patents covering it, they're just incidental. With Mono, however, the likelihood is high, simply because it is a direct clone of MS technology.

Re:Oh good!

[moosesocks]

Actually, Microsoft does benefit indirectly from Mono (even if nobody gets sued), because it's bringing more C# developers into the world at a time when Java is arguably the most common language.

The "standards and interoperability" thing is bullshit too. Any claim you could make against Microsoft with regard to C# could just as easily be made against Sun and Java a few years ago. They've come a long way too, in terms of supporting open standards in good faith -- targeting IE is essentially no longer an issue for web developers; Interix (Windows' little-known Unix subsystem) is surprisingly good; Microsoft are contributing code to JQuery; and PHP on top of IIS is officially supported (while Microsoft also contributed an open-source SQL Server driver for PHP)

All of these are technologies that originated elsewhere, and compliance with the standards is generally pretty good. OOXML was a debacle to be certain, and Microsoft faced a pretty hard backlash from their customers for it and Vista. A major course-correction seems to have taken place from within the company at some point during 2008 or 2009. They know they're losing the desktop market to Apple, and the server market to Linux, and are doing an admirable job of "catching up."

Re:Oh good!

[fredrik70]

Well, wine is a far worse patent trap than mono nad companies still use it as basis for some linux ports

Re:Oh good!

[Anonymous Coward]

Surely, you mean "less freedom"? The users lose freedom because they get more proprietary programs in the market. If anyone gains freedom, it's developers, and among them proprietary and selfish developers.

Re:Oh good!

[ThatsNotPudding]

So say we all.

Re:Oh good!

[Grishnakh]

First, WINE reimplements the Win32 API, which has been around for ages, although it's evolved some. Maybe not quite 20 years, but pretty close, so if there is any patented stuff there, it's about to expire. Mono reimplements .NET, which is much newer ( 10 years).

Second, WINE isn't needed for any Linux distro, and is only an optional install. I'd guess that most Linux users either never or rarely use it. No distro uses it for any standard included software. Contrast this to Mono, which many distros have included as standard, along with various apps that require it. The people behind Mono are pushing hard for Mono to be included as standard in every distro, and for as many apps as possible to use it, so that it becomes as integral to a standard Linux desktop as glibc, and thus much harder to remove or work around.

Re:Oh good!

[niittyniemi]

I don't give a rat's ass because Microsoft will never, ever go after an insignificant, individual end user like me for patient infringement.

Frankly, what you do to your patent's disgusts me. I hope Microsoft go after you with the full force of the law.

Re:Oh good!

[aztracker1]

Well, I can run GTK# stuff without dealing with the non-ecma MS portions of the tech, so "yeah"... not to mention that ASP.Net MVC is under MS-PL license, which is pretty libre.

Article is bogus, needs proofreading

[Peter Steil]

I know it's sin to actually RTFA, but I've been a GNOME user since the 1.x days and figured I'd take a read. The author seems to use the number "3.30" to refer to the current release..

GUADEC

[ReinoutS]

The GNOME Conference (GUADEC) will be in The Hague (NL) this year from July 26-30. You can bet there'll be a lot of GNOME 3.0 hacking going on there. More information: see the GUADEC website.

Please upvote.. The panel can be everywhere

[Anonymous Coward]

See subject: The panel can be rearranged to to the side, bottom, whatever. On my netbook I tend to delete the top panel and add the things I need (task switcher, main menu, clock) to the bottom panel.
Right or left should be no problem either, though.

Re:Please upvote.. The panel can be everywhere

[Nerdfest]

In 2.3 yes, in 3.0, no, or not last time I checked.

Re:Please upvote.. The panel can be everywhere

[Anonymous Coward]

Ok, this may seem as a bit of pedantry but, it's actually 2.30. It's been almost a decade since Gnome had it's 2.3 development versions.

Not the same stuff - much worse!

[_greg]

Yes, I DO remember the early days of Gnome and how much better it was than now:
    - automatic save and restore of multi-workspace sessions
    - handy window operations like maximize-vertically and maximize-horizontally
    - easy to change settings like which app to handle movies, etc.
I remember when clicking on a menu button gave an instant response,
not a several second delay for the first time in a session.

Gnome has become bloated and slower while becoming less stable and less powerful.
It is neither easier nor harder for beginners. It has more eye candy.

Gnome clients have also gone downhill: Evolution used to support my mh mail folders.
Now it uses a database that crashes when I try to load my old mail and fails to work
with my rules. It still doesn't integrate the contact manager with the mail rules.

I'd switch to KDE but they've been destroying themselves even faster!

Re:Not the same stuff - much worse!

[thoughtsatthemoment]

I have some similar impression even though I only use it occasionally. It seems the desktop itself is trying to match features with other desktops, but not much improvements on application development on the desktop.

Re:Not the same stuff - much worse!

[Arker]

This is my impression also.

All the stuff you mention plus keybindings and WM choice. Once upon a time I was totally in love with Gnome. Unix keybindings and WindowMaker integration made it very useable and useful for me. Gnome2 took away all that - took away everything I liked about it - and to ad insult to injury the developers made it a practice to abuse anyone that didnt like the change. It's true I havent tried it recently, and it's also true I probably never will. I am not a masochist.

Re:Not the same stuff - much worse!

[shutdown -p now]

> I'd switch to KDE but they've been destroying themselves even faster!

This. First the disaster that was (still is, from my experience) KDE4, and now those brain-dead Gnome 3 UI mockups. What are they all smoking? Can we please have one sane full-featured DE left?

Re:Not the same stuff - much worse!

[Anonymous Coward]

2010 is the year of GNUstep on the desktop!

Re:Not the same stuff - much worse!

[Grishnakh]

- automatic save and restore of multi-workspace sessions
        - handy window operations like maximize-vertically and maximize-horizontally

Modern Gnome doesn't have those? That's pathetic. KDE4 has them.

I'd suggest giving KDE4 a second try. Make sure you're using a very new version, though, preferably 4.4. The early versions were very bad, and should never have been released for general consumption.

Re:Not the same stuff - much worse!

[Anonymous Coward]

GNUstep could've been a winner if they'd only put forth the small degree of effort to at least look and feel like other desktops.

Instead, they chose to stick with those fucking vertical menus that basically nobody else uses. They made it fucking impossible to have GNUstep apps look like GTK+ or Qt apps. And the end result is that nobody uses it.

It's clearly not a language or API problem, as there are may people who like Objective-C and Cocoa, and develop fantastic apps using both. The problem is clearly with GNUstep, due to all of the effort they put in towards not working will with other apps and desktop environments.

Re:Not the same stuff - much worse!

[jcupitt65]

Gnome has these features, or mostly anyway.

Maximise vertically is in (on Ubuntu anyway) System / Preferences / Keyboard Shortcuts. Scroll down to Window Management and you'll find lots of useful shortcuts you can use and configure.

What's difficult about setting an app to handle movies? You can either click System / Preferences / Preferred Application and pick a generic "multimedia" application, or you can right-click on a particular file and set the program you want to use for that exact file type (it's in properties, under Open With).

I'm not sure what the situation with gnome-session is.

In what way is Gnome slower? The file manager is much, much faster, Startup is much faster. The filechooser dialog (which everyone still hates, hehe) is much faster. Gnome-terminal is finally faster than xterm, phew.

Re:Not the same stuff - much worse!

[pedestrian crossing]

KDE is still (very gradually) recovering from the move to 4.x.

They made Amarok suck, and while it is slowly getting better, it's not what it used to be.

And don't get me started on Digikam! It's still unusable.

Plasma sucks resources like a 5 dollar whore.

Still, I'm one who abandoned Gnome after 1.x and I'm not going back. I just feel fucked from all sides...

Re:Not the same stuff - much worse!

[amorsen]

Maximize-vertically and horizontally are available from the Compiz "maximumize" plugin. It should be installed already if you're running Compiz, but it isn't enabled by default. It's easy to enable with CompizConfig.

Re:Not the same stuff - much worse!

[Anonymous Coward]

- handy window operations like maximize-vertically and maximize-horizontally

Try clicking with the middle or right mouse button on the maximize icon and you'll notice they're still here :)

- easy to change settings like which app to handle movies, etc.

Ever tried right clicking, properties and change the file association?

Gnome is not that bad.. the combination of gnome, gnome-do and compiz make it a fast, pleasant desktop experience. But it would be good is the gvfs would be automaticly part of any gnome/gtk+ application. I'm looking forward of what they planned for gnome 3.0. And with companies like Canonical and Novell behind it which tend to focus more and more on usability this could work out well. Or in the worst case we all switch to KDE and forget all about Gnome ^_^

Re:Not the same stuff - much worse!

[moonbender]

or you can right-click on a particular file and set the program you want to use for that exact file type (it's in properties, under Open With)

I always thought that was an odd place. File association is a system property and not a property of an individual file (unless you want only that one file to be opened with a particular app). At least there should be an "Always open with this" checkbox in the Open with... dialog. Preferred Application doesn't really do it either because it's very broad; I'd either replace it altogether or at least include an "Advanced" dialog.

Re:Not the same stuff - much worse!

[jcupitt65]

At least there should be an "Always open with this" checkbox in the Open with... dialog.

That's how it works, doesn't it? Open With adds an application as a handler for all files of that type, not just this file.

If you right-click and select "Open With Other Application" you can quickly add a new type-handler association, if you right-click and select properties you can edit all type-handler associations.

I'm not sure how you add a new file type through the GUI though. I think .desktop files have to do it for you (erm, I think).

Re:Not the same stuff - much worse!

[moonbender]

Hm. When I select Open With -- Other Application... from the context menu, the interface does say "Open X and other files of type Y with". But the actual file association (as configurable in the file properties) isn't changed, and double clicking still opens the file in the old application. So that would seem to be a bug. (Ubuntu 9.10)

Re:Not the same stuff - much worse!

[jcupitt65]

It doesn't change the association, it adds a new one.

For example, .xls files can be opened by Gnumeric, OpenOffice and no doubt some others too. If you right-click on a .xls file, select "open with other application" and pick gedit, you'll find all .xls files now have an "open with gedit" item in their right-click menu but the default handler will remain OpenOffice (on Ubuntu anyway). I guess this is a safety thing: you don't want to be able to break associations too easily.

If you want to change the default association, right click on an .xls file, select Properties, pick the Open With panel, and pick the radiobutton next to gedit. Now doubleclicking on .xls files will open them (uselessly, heh) with that program.

Re:Not the same stuff - much worse!

[jenningsthecat]

Hear hear! I've been so annoyed with Gnome that I tried switching to KDE, but K-bloat was even worse than G-bloat, so now I'm back on Gnome, but using Dolphin whenever I can because Nautilus is so brain-dead.

Before Gnome programmers add more whiz-bang crap that needlessly chews up processor cycles, I hope they bring their file manager into the 21st century. Click-and-drag to select multiple files/folders, along with a full-featured INTEGRATED search capability, would be a good place to start.

Re:Not the same stuff - much worse!

[moonbender]

Oh ok. That's what I was referring to in the first place, I think the file properties is an odd place for changing the default associations (since it is not a property of one specific file). And for the sake of convenience there ought to be a checkbox in the Open with dialog that makes the selection into the default association. I agree that changing the default association should be an explicit choice.

Re:Not the same stuff - much worse!

[Grishnakh]

It's also been a standard part of KDE (kwin specifically) for pretty much as long as KDE has been in existence. What kind of moron would take it out of any window manager? Are they trying to make Gnome more like Windows or something, by removing all the handy features that make Linux a joy to use? What are they going to do next, push to remove the bash shell altogether from Gnome-based distros and replace it with a stripped-down shell that's a clone of DOS?

If you have to install a different window manager (one which uses 3D, even worse; not exactly a viable option if, say, you're accessing the machine remotely with VNC or NX) to get basic window management features, then that just tells me that Gnome isn't usable at all.

Re:Not the same stuff - much worse!

[DrXym]

Of course its become easier for new users. GNOME 1.0 was a passable first effort but it suffered from lack of functionality and too many rough edges. A new user couldn't even set up a network connection, or change the screen resolution because there were no tools to do those things. Many of the apps were unforgiving, crashy and lacking features. Ximian smoothed out a lot of the rough edges and standardized installation and look & feel. The biggest influence of course was the HIG which whipped the whole experience into shape making it easy for novices and power users to use the thing.

I think its very obvious why GNOME has gained mindshare while KDE has floundered. GNOME has made the occasional wrong turn but by and large it's a very usable desktop. By which I mean it's attractive, intuitive, forgiving, simple and task centric. By comparison KDE 4 may have more features but the experience is a disaster.

Re:Not the same stuff - much worse!

[Anonymous Coward]

If you long for the old Gnome try XFCE, I did about 3-4 years ago and have found it excellent. Thunar is super fast and simple but does all the file management you expect, and although it looks like a Filer clone it's far better.
Although it's handling of the desktop icons is a little annoying, it would be nice to have a "auto arrange icons" option and an option to make sure device/home dir/trash links come before the other icons. But it's steadily getting better and even the beta releases have been rock solid.

Re:Not the same stuff - much worse!

[amorsen]

You don't have to install a different window manager, Compiz is what you'd typically use with Gnome.

And your rant about 3D is very 90's. If Linux doesn't give you remote 3D 20 years after Silicon Graphics got it working, then complain about that. Don't complain that the Gnome desktop has finally entered the 21st century.

Re:Not the same stuff - much worse!

[Knuckles]

- automatic save and restore of multi-workspace sessions

Works for me in Lucid, and I think it's for the first time that it also works for Firefox.

- handy window operations like maximize-vertically and maximize-horizontally

Middleclick or rightclick the maximize button, and IIRC it's been like this for a long time (I use Compiz though, so YMMV)

- easy to change settings like which app to handle movies, etc.

I don't know what's your problem there, rightclick file -> Properties -> Open With works fine for me, and for disk media it's even simple because it asks on insertion, else go to Edit -> Preferences -> Media in a file manager window in Ubuntu; IIRC stock Gnome has a settings entry in the desktop-wide menu System -> Preferences, too.

I remember when clicking on a menu button gave an instant response,
not a several second delay for the first time in a session.

I don't try to be an ass, but works for me, I think. At least I never noticed anything annoying. And I think that your nostalgia clouds your judgement, because on the topic of delays, I remember that opening the application menu in earlier Gnome 1.x took several seconds for the first time in a session due to scanning for menu icons. And I'm sure about this because I was the one who filed the bug. It's almost instantaneous now.

I don't know about your mh trouble, but there are at least two workarounds that took 1 sec to google: http://www.mail-archive.com/evolution@lists.ximian.com/msg13503.html
Anyway, I think it's a bit unreasonable that all apps will forever support data formats for edge cases just because you are too lazy to convert them. And if you find that evo is missing features that you need, well, what about using a mail client that fits your requirements?

Gnome-screensaver still broken...

[Anonymous Coward]

Call me when the $%%&$%#^ that maintains that part of it allows people to actually tune the Gnome-screensaver modules without ripping it all out and replacing it with xscreensaver.

Re:Gnome-screensaver still broken...

[Knuckles]

Call me when the $%%&$%#^ that maintains that part of it allows people to actually tune the Gnome-screensaver modules without ripping it all out and replacing it with xscreensaver.

If you had cared to, you'd know that the old way of doing this was wrong and sucked, and the gnome-screensaver maintainer is happy to accept patches to restore the functionality in a sane way.

Mobius Story!

[Anonymous Coward]

If you do a search for images of early GNOME releases and compare the results with 2.30

Actually, now when I do a google search for "images of early gnome releases", every result on the first page is just a link back to this slashdot story.

Can anyone tell me what type of images I might have seen before this story was posted?

Why the hang-up with version numbers?

[rudy_wayne]

"2.30 will probably be the final version of the 2.0 series"

I've noticed that open source software generally seems to be more hung-up and obsessed with version numbers than proprietary software. For example Linus Torvalds has said that there will never be a version 3.0 of the Linux kernel. So I guess 2.9.99.99.999 will be the end of the line.

I don't get the big hang-up with version numbers. Who cares if it is 2.30 or 3.0? My current nVidia video driver for Windows is 196.21 -- as long as it works, who cares?

Re:Why the hang-up with version numbers?

[Sponge Bath]

So I guess 2.9.99.99.999 will be the end of the line.

With the current prohibition on a stable kernel/driver interface in the linux kernel, 2.9.99.99.998 drivers will be incompatible with 2.9.99.99.999

Re:Why the hang-up with version numbers?

[Sponge Bath]

Before I get a bunch of "we don't care about binary only" drivers type responses: I'm talking about open source drivers, including those in the main kernel. The constant code churn means constant and frequent changes to driver internals are required just to keep up. New bugs are a frequent result. You are always chasing the latest change with bug fixes to compensate.

Honestly, a well defined and stable interface is sometimes useful.

Re:Why the hang-up with version numbers?

[hduff]

For example Linus Torvalds has said that there will never be a version 3.0 of the Linux kernel.

Because there will be no more major revisions to the kernel? That would make sense.

Re:Why the hang-up with version numbers?

[Anonymous Coward]

In many open source projects the version numbers have a technical basis, e.g. ABI or API compatibility. In proprietary software it's usually marketing fluff. Unfortunately much open source is heading in the same direction.

Re:Why the hang-up with version numbers?

[Anonymous Coward]

no. after some version 2.9 will come 2.10 (provided it is a non-major upgrade)

OSS Version rundown

Major.Minor.BugFix-Build

In general Major version numbers are only incremented for what is essentially a new project. (Kernel 1.X to 2.x, Gnome 2.x and 3.x)
Minor is incremented when new features are added, or the API is changed in some way (Gnome 2.26 and 2.3)
Bugfix is incremented when code is changed, but the API remains identical (Gnome 2.26.0 2.26.1). When a shared object (.so) gets a bugfix increment, it should be a drop-in replacement for the older version.
Build is hardly used in most projects. If you make an error building you bin, you can re-release based on the same exact c-code, a new bin.

Firefox used to play with these numbers. Then came the jump to 3.5 (much like netscape's jump from 4 to 6). Even so, a bug-fix update should still work with you extensions. A minor update may require a little recoding, and a major update requires rewriting and may require re-design.

Re:Why the hang-up with version numbers?

[ookaze]

"2.30 will probably be the final version of the 2.0 series"

I've noticed that open source software generally seems to be more hung-up and obsessed with version numbers than proprietary software. [...] Who cares if it is 2.30 or 3.0? My current nVidia video driver for Windows is 196.21 -- as long as it works, who cares?

Seems like to me you are the only one hung up about these version numbers here, and you seem to care a lot about these version numbers indeed, contradicting your question of "who cares?".
The OP was only stating a possible fact, which seems to irk you to no end.

Re:Why the hang-up with version numbers?

[Just Some Guy]

I don't get the big hang-up with version numbers. Who cares if it is 2.30 or 3.0?

3.0 is in an entirely different series from 2.30. For comparison, KDE 4 isn't just KDE 3.6 with an updated major version number.

key-bindings

[smoothnorman]

Does this version of GNOME allow for easy global key rebinding? There was a version not long ago that sent me off to KDE that appeared to impose some rather autocratically determined key-bindings.

Re:key-bindings

[Albatrosses]

I ran into that too. I've had Compiz configured with Meta+Button1 = move a window, Meta+Button3 = resize a window for as long as I can remember. It's quite handy - no more groping for that little resize handle, just click-drag click-drag done.

Then, I tried out whatever build of GNOME that was, and suddenly Meta+Button3 brings up the window context menu (with "close" helpfully placed at the top so I hit it every time I try to do a quick resize).

Try to set the binding in CCSM? It claims to have changed, but Meta+Button3 still brings up the menu, and if I close and reopen CCSM it's back to not being set. I checked keyboard preferences, keyboard shortcuts, everything. Couldn't find the source of it. It just overwrote whatever I told CCSM to do, with no indication of why. I tried hacking it in gconf, I tried the "Simple CCSM", no avail. And here I thought Linux was about user choice... maybe I'll buy a Mac instead :)

Re:key-bindings

[drinkypoo]

You might try using ccsm, actually, because it has keybindings too, and they seem to win over everyone else's. They're in general settings. Meanwhile, file a bug report.

Helsinki Syndrome

[Anonymous Coward]

HelsinkiSyndrome(TM)

The mono trap and GNOME

[jmorris42]

> And Gnome has been adopting mono like it doesn't matter.

You are out of date. Have Fedora 13 Alpha + all updates in a VM right now and behold:

[root@Fedora13 ~]# rpm -qa | grep mono
dejavu-sans-mono-fonts-2.30-2.fc12.noarch
liberation-mono-fonts-1.05.2.20091019-5.fc13.noarch

Everything works just fine. They ditched F-Spot for Shotwell and replaced Tomboy with the C++ port GNote. With those gone mono doesn't need to be installed. Somebody caught the cluetrain and stopped Novell from infecting GNOME with their patent poison.

Re:The mono trap and GNOME

[bmo]

That's encouraging.

Let's hope it stays that way.

But is that the "installed" or did they remove Tomboy and the rest in the repositories too?

--
BMO

Re:The mono trap and GNOME

[jmorris42]

> Let's hope it stays that way.

I'd think the die is now cast. Enough folks yelled "STOP!" loud enough they backed off from making mono a core dependency and replaced a pair of otherwise keeper apps out of the default install. It would be pretty hard to introduce a must have dep now, what would they say to the authors of F-Spot and Tomboy?

> But is that the "installed" or did they remove Tomboy and the rest in the repositories too?

Mono, F-Spot and Tomboy are all still in the repo. No problem with that, use it or don't. Heck, outside the US where software patents don't exist there isn't any reason not to use em. The problem was if we allowed those camels to put their nose under the tent there wouldn't be any way to stop the conversion of GNOME into a .NET project. Having a few .NET apps or using mono to run foreign code isn't going to be a problem because Microsoft won't drop the patent bomb on those uses. They would lose more than we would.

Much the same as Wine is probably an even greater patent minefield but nobody objects to it being in every distro's repo. But we would be daft to let some Microsoft sockpuppet con us into using winelib as a reason to adopt Win32 as the basis for the Free Software desktop stack. Think of the arguments in favor that could be made. Apps run as fast, or faster, under Wine as Windows so performance isn't an issue. With only a little care binaries would run unmodified on x86 Linux and x86 Windows; just like mono/.net! Lots more potential developers know Win32. Develop Free Software on Windows with their 'superior' tools. And so on. Then once a critical mass of can't live without apps were Win32 only the patent hammer comes down and we all run Windows.

Re:The mono trap and GNOME

[Anonymous Coward]

I don't have a problem with mono. The C# language is java fixed up, the apps are nice. It's unlikely that MS will do this, if they do then so be it.

In the meantime, whats the point in all that ram remaining empty.

I think a lot of KDE users disappeared with KDE4.

[aussersterne]

I used KDE from KDE 1.0, when I switched away from TWM. I was fully integrated into the KDE "way of life," and reliant on lots of KDE apps.

I tried to use KDE4.0 but after about two weeks it got the boot. Though it has theoretically improved and I keep a KDE 4 installation on my Fedora 12 personal machine, logging into KDE thus far provides no incentive to switch back, despite updates.

Dolphin is still intolerably slow. Important apps still don't share a consistent appearance; Firefox, Chrome, and OpenOffice in particular look good in GNOME but are full of distracting artifacting and other appearance problems in KDE. GNOME apps in general don't mix well with KDE themes right now. The graphics still don't work right. A notification balloon is likely to take out half the taskbar, etc. They blame this on the radeon driver and I believe them, but that's the hardware I have, and GNOME shows none of the same problems. Desktop management for multiple monitors doesn't behave as I expect it to, and it's difficult to create a configuration that jostles well amongst varying configurations of external, internal, or both, monitors without taskbars disappearing or desktops shifting from display to display unexpectedly. The default icon theme is far too colorful and luminous for focused desktop work of the kind that I do (lots of writing, editing, and calculating) but there are few replacement icon sets to be found. The wireless connectivity manager seems incapable of working with my simple home WiFi installation without needing constant reconfiguration and tinkering, while in GNOME it "just works."

Yes, some of these things could be fixed, but to trudge through each one of them would require rather a lot of time and effort that I just don't have to spare. So despite the fact that I'm still not wild about GNOME either, KDE4 is simply not on the cards in the near future for me. What's missing everywhere is polish. Not the kind that makes widget corners have a "glass" appearance, but the kind that keeps widgets from disappearing or artifacting unexpectedly, or the kind that doesn't leave you wondering why the hell the widget doesn't work, or there isn't a widget for that at all, in the first place. Details work. Not big thoughts. KDE needs to cut out the innovation for a while and patch roof leaks.

I wouldn't be surprised to hear that many other KDE users right up through KDE 3.x switched to GNOME with the KDE4 release.

Re:I think a lot of KDE users disappeared with KDE

[anshulajain]

I quit from being a vocal KDE4 supporter and moved on to GNOME with the KDE 4.4.x release. The forced Akonadi/Nepomuk/MySQL messup coupled with the perennially unstable Kontact made me throw up my hands in disgust. Why the fsck do I need 100MB+ of Akonadi/MySQL stuff (Nepomuk can be disabled, mercifully) just to run Kontact? Besides, Kontact cannot seem to work properly with anything else other than itself. It cannot sync to Google Calendar, cannot sync to Exchange. Using KDE4 in a corporate environment as a desktop became a chore for me to do. I simply gave up and moved on to GNOME, where I am more than happy with its offerings.

summary?

[Anonymous Coward]

Shouldn't that be "the free desktop in general, and GNOME in particular"?

Mono considered harmless

[steveha]

If there is a patent trap in Mono (which exists regardless of its license), then that means that MS can then sue all those businesses and governments for patent infringement.

I'm getting sick of this meme.

Tell me, are you a patent attorney? What is your expertise for making claims like this one?

Now I'm not a patent attorney either, but here is my understanding: If Microsoft does assert some kind of submarine patent, the main effect will be to cause GNOME and everybody else to yank out Mono. At that point, we will just have to port the Mono apps to Java or something. That is the absolute worst case. Can you give me an example of any time where some company had a submarine patent, then suddenly asserted it, and successfully extracted a bunch of penalties from businesses and governments?

Furthermore, while I'm still not a patent attorney, I have read Groklaw for a while, and I read some essays there about the "unclean hands" doctrine. If a company has patent rights, and discovers that someone is infringing, that company has a duty to inform the infringers as soon as possible; it is not allowed to just let the patent sit there ticking like a bomb, and then demand extra damages because the infringer was infringing for so long.

So, let's review: Mono is a technology that is very similar to the JVM, which in turn is similar to other virtual systems, going all the way back to the UCSD P-system. The amount of prior art is staggering. Besides that, the only danger is a submarine patent, not a new patent: the .NET stuff has been around for years and years, and you have to file for a patent before you publicly disclose a technology, or you lose your chance.

So, the alleged threat is that there is a patent already granted, that nobody has noticed, on technology that has a ton of prior art; and Microsoft is deviously not asserting the patent, but is going to later. Microsoft won't care about the negative publicity for itself and for .NET, because it stands to gain so much and is certain its patent will survive all challenges. And anyone infringing will somehow be on the hook for penalties.

I for one don't believe any of it. C# is as safe as Java and Mono is as safe as the JVM.

steveha

Re:Mono considered harmless

[fredrik70]

Also, MS extended their community promise (apparently legally binding) to not go after mono with patents:
http://arstechnica.com/microsoft/news/2009/07/microsoft-issues-patent-promise-dispels-mono-concerns.ars

and directly from MS:
http://www.microsoft.com/interop/cp/default.mspx

Re:Mono considered harmless

[Just Some Guy]

Also, MS extended their community promise (apparently legally binding) to not go after mono with patents:

No, they didn't. From your second link:

Microsoft irrevocably promises not to assert any Microsoft Necessary Claims against you for making, using, selling, offering for sale, importing or distributing any implementation, to the extent it conforms to one of the Covered Specifications, and is compliant with all of the required parts of the mandatory provisions of that specification ("Covered Implementation")

If Mono (or any other .NET implementation) fails a test suite, all bets are off.

Re:Mono considered harmless

[supersloshy]

If there is a patent trap in Mono (which exists regardless of its license), then that means that MS can then sue all those businesses and governments for patent infringement.

I'm getting sick of this meme.

Tell me, are you a patent attorney? What is your expertise for making claims like this one?

Now I'm not a patent attorney either,

I stopped reading right at that point.

Re:Mono considered harmless

[Anonymous Coward]

I stopped reading right at that point.

Well, good for you! We all really wanted to find out whether you read the whole thing, or not, and now we know. What a relief!

Just one thing: I have to know... did you read the GP posting, or did you stop reading it too? GP isn't a lawyer either, but perhaps you read it anyway. I just can't wait to find out!

Re:Mono considered harmless

[aztracker1]

*Sigh* The community promise is relevant to more than just .Net, if you look below, you'll see there's a number of specifications offered, and covered under the promise. Try finishing what you read, and comprehending what is there. (Including ISCSI, which is on a lot of *nix based NAS hardware).

I mean, lets be honest, if they haven't come after SAMBA, WINE and a number of other technologies that more directly affect their bottom line, why would they go after an application toolchain that does nothing but promote their tech and wares?

Re:Mono considered harmless

[aztracker1]

You are misinformed, ignorant, or a troll. The community promise refers to more than .Net/CLI/Mono, the "Covered Specifications" means an item in the list of specs covered by the promise.

Re:Mono considered harmless

[Just Some Guy]

How many times are you going to reply to me?

I'm excited

[tthomas48]

The stuff that's happening with gnome desktop is fantastic. It's especially nice on small laptop/netbook/tablet machines. The latest Ubuntu (Lucid Lynx in beta) has built in social networking that actually jumps ahead of OSX or Windows. The fact that I have something like TweetDeck built into my OS is pretty cool. Sure there are some rough edges. OSX has rough edges too. But I rarely find myself explaining away huge deficiencies. It's just a different bug from your OSX or Windows bug.

I'm excited. But then again I'm trying to be accepting of change as I get older, rather than screaming at the kids to get off my lawn.

Re:I think a lot of KDE users disappeared with KDE

[Tehrasha]

I have always thought that KDE's early popularity was a combination of its similarity to Windows (look and feel), and the popularity of Knoppix as the first wide-spread live-CD.

How about fixing these before 3.0:

[spitzak]

1. Make it possible to hide the taskbar/panel/whatever. In every version of Gnome for years now the best you could do is make it hide to 1 pixel tall, and this requires gconf. The default "hide" is an incredibly stupid looking thick 6 pixels. PS: KDE has had this working for about 10 years now, so you can't claim it is some technical defect in X or some other excuse.

2. Fix the setting so that clicking in windows does not raise them. Apparently this configuration option causes it to ignore ALL attempts to raise the window, including the program itself calling XRaiseWindow()! This is completely broken. What we want the configuration option to do is stop clicks inside the window from raising it, NOTHING ELSE!!!

Of course I am wondering if they are purposely leaving these things broken so that they can later claim "nobody uses those options so we will remove them".

How far have we come? About a quarter-inch.

[jensend]

For those who were around for GNOME 2.0 back in 2000, the 2.30 release stands as evidence of how far GNOME in general and the free desktop in particular have come in the last decade in usability and design.

For those who were around for GNOME 1.2 back in 2000, the 2.30 release stands as evidence that Linux on the desktop and GNOME in particular have made awfully little progress in the last decade. GNOME 2.0 was released in 2002, not 2000, and it was horrid; maybe if your first experience with GNOME was 2.0 then you might think 2.30 was a vast improvement- heck, TWM is a vast improvement on GNOME 2.0. 2.0 was extremely bug-ridden, and if you wanted to change anything from its mind-numbingly bad defaults you had to putz around with finding where in gconf's xml you could go to change things.

If you were around for 1.0, the RH 6.1 "October GNOME" release, or 1.2, you know that GNOME made a lot of progress, was centered on the needs of those most likely to use Linux rather than on unsubstantiated usability claims, and was becoming quick, convenient, and powerful. The progress GNOME made between 1998 and 2000, the big improvements in the 2.2 kernel series, and a host of other developments made it seem like Linux really would overtake Windows for desktop use soon. But I really don't find much about modern versions of GNOME that really improves on 1.2 or maybe 1.4; the last 9 years have seen little improvement in the Linux desktop IMO.

Re:How far have we come? About a quarter-inch.

[Jollyeugene]

And the small, lean desktop apps of those days have been replaced with bloatware as well. Abiword and GNUCalc were fast and useful back in the GNOME days, Now they are a joke-- they crash as soon as you do anything with them.

Numerous little GNOME 1.4 applications existed (time trackers, media players and cd rippers) that were fast and light and that worked well. Every GNOME release one of these apps would get booted out in favor of some bloated project that was "new" and "great". The "new" and "great" always meant twice the memory size, unstable habits while running, and much less functionality-- provided in drop down windows and wizards.

GNOME ended up with HAL, BONOBO, and mono. And guess what? All that crap in 2.0 that they added over 10 years-- that is now going away again because it was all so piss poor to begin with. LOL.

Hopefully GNOME 3.x is more like GNOME 1.5x.

Hopefully GNOME stops trying to include default applications and goes back to providing the API's and the HID best user practices and requirements.

I don't want to hear that web browser X is out of GNOME, and IM client Y is now "blessed" to be in GNOME. That crap is retarded. When your GNOME package starts depending on this silly list of dorky applications it gets really annoying. Either the app meets the GNOME usability and code and API usage guidelines or it does not. Give it a gold star, or a silver star, or a black dot, or whatever. But making productivity apps a core part of the desktop, and then changing the apps in every release to reward whomever has kissed the most ass that release has pissed me off to no end these past 10 years.

Re:How far have we come? About a quarter-inch.

[gravis777]

I agree. I hope I don't get marked troll, but I just did a Google image search for 2.0, 2.1, 2.2 and 2.3. Images posted a few days ago versus stuff posted in 2001, and guess what, I could not tell a difference. Except that you can have strange fonts in the app window name. No, it looks EXACTLY THE SAME! No, even back when I was a Linux user, I was unimpressed with Gnome. Back in the early days (1.x release), it was beautiful, but ran SO SLOW compared to KDE. However, it seems to me that very little progress has been made in the UI of Gnome. In the past ten years, on the Mac side, we have seen the introduction of OSX, and the evolution of it, which is now functional, fast, and beautiful. On the Windows side, we went from Windows 2000 to XP to Vista to Windows 7. Gnome 2.3 has a look and feel that reminds me of Windows 2000. Welcome to the year 2000 everyone!

Even having left Linux totally about three years ago, I feel that I have used it enough to recommend ANY X manager over Gnome. You ever wonder why most Linux distros bundle KDE with it? I mean, look at the history of KDE: http://images.google.com/imgres?imgurl=http://itpencil.files.wordpress.com/2008/02/kde3.png&imgrefurl=http://zebardast.wordpress.com/2008/02/02/kde-history-in-screenshots/&usg=__DhWX0VcaoSaOkdSFmFvFf8umTlo=&h=864&w=1152&sz=347&hl=en&start=2&itbs=1&tbnid=m6-4wpQ_HueotM:&tbnh=112&tbnw=150&prev=/images%3Fq%3Dkde%2B3%26hl%3Den%26gbv%3D2%26tbs%3Disch:1

I would also happy recommend BlackBox over Gnome. I am honestly shocked to hear that Gnome is still around.

Okay, you can mark me troll now.

Re:How far have we come? About a quarter-inch.

[gravis777]

Sorry, should have posted the link to the website rather than the Google cache: http://zebardast.wordpress.com/2008/02/02/kde-history-in-screenshots/

Re:How far have we come? About a quarter-inch.

[ookaze]

For those who were around for GNOME 1.2 back in 2000, the 2.30 release stands as evidence that Linux on the desktop and GNOME in particular have made awfully little progress in the last decade. GNOME 2.0 was released in 2002, not 2000, and it was horrid; maybe if your first experience with GNOME was 2.0 then you might think 2.30 was a vast improvement- heck, TWM is a vast improvement on GNOME 2.0.

Well I was around at the time, and I used Gnome 1.4 (actually, I still compile it and still uses a few Gnome 1.4 programs to this day, like gcombust).
I'm not surprised by these reactions, but to me, they come from trolls or from people with a very narrow-minded view of the world that basically revolves around american geeks using Gnome. Because Gnome 2.0 actually brought huge improvements for Gnome, at least in presentation. If I have to select two of the most important ones, it has to be i18n/l10n and fonts handling. But actually there are far more improvements than that.
Comparing the usability worldwide of Gnome 2 and TWM, and saying TWM is better, is just plain stupid hyperbole.
I can't believe people are so dense. Sure Gnome 2 came with its loads of bugs, but to say there was little improvements, wow!

The progress GNOME made between 1998 and 2000, the big improvements in the 2.2 kernel series, and a host of other developments made it seem like Linux really would overtake Windows for desktop use soon. But I really don't find much about modern versions of GNOME that really improves on 1.2 or maybe 1.4; the last 9 years have seen little improvement in the Linux desktop IMO.

So this is due to a very narrow vision of users and the world outside english speaking users then. And blindness too, despite Gnome having improved on usability and disabled people assistance.
The replacing Windows part was always wishful thinking by geeks that don't understand the majority of other people around them, which is perfectly normal, most of us have strong NT personalities, which represent around 10 % of world population.
I argue that big improvements in the 2.4 and 2.6 kernel series (at all levels including audio and video), the Gnome 2 progress, the freedesktop initiative and middleware tools between the kernel and desktop are even bigger improvements than what we saw in those Gnome 1 years.
The experience was smoothed so much that some people don't even see the improvements anymore, despite them being right before their eyes.

Re:How far have we come? About a quarter-inch.

[jensend]

And blindness too, despite Gnome having improved on usability and disabled people assistance.

Here's one thing I think we can agree on: You might think GNOME 2.x was a big improvement over 1.x- if you're BLIND .

The other >99.5% of us who don't need a screen reader haven't seen (sorry about the pun) a lot of improvement. Pango is nice and all if you want your fonts blurred for you or if you only read Sinhalese; I think you're misremembering the situation with respect to i18n/l10n in 1.x- not really that bad.

GNOME in general and the free desktop in particul

[Anonymous Coward]

> GNOME in general and the free desktop in particular
So gnome is a generalisation of free desktop. Cool ^Gool!

Re:I think a lot of KDE users disappeared with KDE

[xtracto]

I quit from being a vocal KDE4 supporter and moved on to GNOME with the KDE 4.4.x release. The forced Akonadi/Nepomuk/MySQL messup coupled with the perennially unstable Kontact made me throw up my hands in disgust.

But those are just apps that you could replace with other apps.

I do not like KDE 4.x, specifically the desktop environment. Because I do not care at all about the "look and feel" of the apps I use, I have no problem using KDE apps in a Gnome desktop. I use RkWard and kMyMoney on a daily basis, in a Gnome desktop.

The problem for me is the KDE desktop which has always felt very "breakable". On every KDE desktop I have used, it is very easy to get the crocodile or now the bomb icon after a "core dump" crash. Meanwhile other desktop environments or operating systems behave normally at the same computers.

Not the end of the line for 2.x

[Anonymous Coward]

"On the other hand, 2.30 will probably be the final version of the 2.0 series."

No it won't. 2.32 will be published in parallell with 3.0. The new gnome-shell (et.al.) that introduces new UI concepts to gnome heavily relies on HW accelerated graphics. The state of many Linux graphics drivers still need to mature to be able to run gnome-shell properly. The 2.x era will live on (and *my* magic crystall ball tells me it will do so for a long time.)

My impression is that Bruce Byfield never seem to get much stuff right. Who is he and why are people listening to him?

Re:Not the end of the line for 2.x

[dejanc]

The 2.x era will live on (and *my* magic crystall ball tells me it will do so for a long time.)

Not necessarily. GNOME being an open source project, it will be hard to motivate people to maintain whatever becomes obsolete in 2.30, and it will be even harder for distribution packagers to maintain a double set of GNOME (which is already notoriously tedious to package).

There is always XFCE for people who don't like 3.0 changes. Even now, differences between GNOME and XFCE are not that big, and since most of Gnome's DE functionality has been switching to GTK anyway (e.g. GVFS), XFCE will have access to it.

Ratpoison

[Jollyeugene]

I am happy with an old version of RatPoison for serious work. Less distractions, more focus on the task at hand, and plenty of scripts and plug ins to get the job done.

When I don't care about productivity, GNOME with bells and whistles keeps me amused. Although, I do miss the 1.4 days.

Re:I think a lot of KDE users disappeared with KDE

[Anonymous Coward]

Yes! I was one of them! I too used KDE since the 1 release and found KDE3 a joy to use. Although I did use KDE4 on the first early releases, it struck me that everything on my desktop was shiny, but I couldn't actually do anything. So I switched back to GNOME, but found that they'd taken away the configuration for *everything* or I had to use GConf. So now I'm trying XFCE or just switching around. *sigh*

Re:I think a lot of KDE users disappeared with KDE

[Bronster]

This. I miss the KDE that was nice and easy to use, but Gnome got good enough right about the time KDE got shit. Maybe if I wasn't on Ubuntu mostly I'd consider switching back now, but Kubuntu is still the poor cousin, and Ubuntu has enough problems with consistency without having to worry about more of them.

Think outside the box

[ThatsNotPudding]

Move the Minimize button to the upper left, the maximize button to the lower right, the close button the lower left and for the upper right: a button that does nothing.

Re:I think a lot of KDE users disappeared with KDE

[optimus2861]

I wouldn't be surprised to hear that many other KDE users right up through KDE 3.x switched to GNOME with the KDE4 release.

/Raises hand!

To this day I don't know what the KDE (and Amarok!) developers were thinking in blowing apart their fantastic software, rebuilding it from the ground up, and choosing to throw out all the great functionality they had and replace it with "Look isn't this shiny!".

I switched to Ubuntu from Mandriva after the first KDE4 Mandriva release came out and haven't looked back. I never particularly liked GNOME but KDE took a flying leap backward and put GNOME in the lead by default.

Light features + heavy footprint = Meh

[AtlantaSteve]

The most interesting thing to me about Gnome these days is that it's memory footprint is still ridiculously fatter than Xfce's, even though Xfce has caught up with Gnome's basic features.

My "family computer" has been running default Ubuntu with Gnome, and my non-technical wife has been happy with it. However, it's starting to show its age, and with each major software update it gets a little slower and slower. So for the hell of it last month I thought I'd experiment with Xfce and see if I could postpone the next computer purchase until the holiday season.

I might postpone a lot further out than that! Thanks to Canonical's packaging of Xfce, it looked pretty much the same as Gnome right out of the box. After 5 minutes of tweaking the panel icons and theme settings, it was almost indistinguishable from my machine's previous setup. My wife didn't notice at all until three weeks later when she went to copy some files from a USB drive, and noticed that the file manager was Thunar rather than Nautilus. She turned out to be happier with Thunar though, because it doesn't randomly freeze up during drag-and-drop operations.

For years now, Gnome's "niche" has been with those who want something more feature-rich than Fluxbox, yet simpler and more lightweight than KDE. However, Gnome's basic functionality has been pretty stagnant for a long time, and lighter-weight desktop environments are catching up with the core expected feature set. Right now, I don't know of any compelling reason to run Gnome other than wanting to use a lot of Compiz visual effects, and Xfce is almost caught up with that too.

Can you over simplify?

[LWATCDR]

Really as long as a Desktop environment has the features you need it can not be over simplified.
I have not used KDE4 yet but I did use KDE3 a good bit. Frankly I liked both KDE3 and Gnome about the same. The thing is I really came to like Ubuntu for it's simplicity and unified look and feel. It is a little plain but very functional.

Re:I think a lot of KDE users disappeared with KDE

[Anonymous Coward]

Not for me. I came from an Amiga background. KDE was cool in the 2.x and 3.x versions because it was fast and complete. Now it's destroyed and has become a developer playground and effects fest.

Re:I think a lot of KDE users disappeared with KDE

[jadrian]

Dolphin is still intolerably slow.

I use dolphin on my 5 year old laptop. Works fine.

Important apps still don't share a consistent appearance; Firefox, Chrome, and OpenOffice in particular look good in GNOME but are full of distracting artifacting and other appearance problems in KDE.

openSUSE does a pretty good job with Firefox and OpenOffice.

The graphics still don't work right. A notification balloon is likely to take out half the taskbar, etc.
They blame this on the radeon driver and I believe them(...)

Works fine more me. Using ATI.

The wireless connectivity manager seems incapable of working with my simple home WiFi installation without needing constant reconfiguration and tinkering, while in GNOME it "just works."

It just works for me in KDE.

Yes, some of these things could be fixed, but to trudge through each one of them would require rather a lot of time and effort that I just don't have to spare.

It just works for me in KDE.

KDE needs to cut out the innovation for a while and patch roof leaks.

They've been polishing KDE4 for some time now.

I wouldn't be surprised to hear that many other KDE users right up through KDE 3.x switched to GNOME with the KDE4 release.

If they preferred KDE 3.5 to GNOME up to there, why would they switch to GNOME just because KDE 4 came out?

What we're still missing

[WheelDweller]

Not to throw dirt in the direction of the GNOME developers- they've done great things. It's the only environment I use. (Other than the classic, "fork-spoon-knife" paradigm of biological input. :)

The Gnome world would grow and expand greatly if ham-handed office workers could shape high-level code into things useful to the office. Think VisualBasic and all that embedding stuff on the other leading brand.

Being able to ask for a (sheet,row,colum) address of a sheet to print a separate document without reading a manual. Or construct a basic database for simple stuff in the office, again without reading a manual and/or immersing oneself in a whole new environment.

Anyone ever use Informix 4-GL's environment, for example?

Allowing the typical office worker to do things that *can't* be easily done on Windows is key to adding desktops. The more ad-hoc stuff an office worker can do without going to a seminar or conference, the better!

I'll bet ALL THESE THINGS CAN ALREADY BE DONE...but it's still not quite 'easy'.

Re:What we're still missing

[Knuckles]

In my job I frequently have to clean up after the monkeys that wrote code in easy-to-learn environments like VBA, and I am 100% convinced that long-term it is better to have a threshold that results in actually having to have a bit of clue before writing code.

with you

[improfane]

I agree that the GUI should be as flexible as possible.

Although you really are dealing with two forces:
GUI code is something that will be executed more than any other application code, it ought to be very fast and efficient. At the same time, humans are rather slow, begging the question if optimizing the GUI code is worth the performance gains.

It's interesting how thousands of objects can potentially slow down GUI controls. Should they?

I presume that GUIs are often object orientated because it makes GUI design easier and more flexible. This makes styles changeable on the fly and hence slower.

If you take a look at Java's GUI related design you will find that a lot can be overriden.

I would love the day when I can select one of a number of default layouts and workflows such as Windows, Mac or a particular Linux distribution and have lots of platform specific details set themselves - while being fully revertable. It would be clever if the policy were completely separate from the mechanism so that even fine grained behaviours like where buttons are placed, dialogue button order. They could even be mixed together if that is desired.

Imagine having a Windows centric 'screen', Ubuntu, Mac and whatever else without leaving the same OS...not that there would be much point.

Yay! Then it sounds like YOU should use KDE

[aussersterne]

and I should not. Since none of those things work for me. I guess they just like you better.

Re:Yay! Then it sounds like YOU should use KDE

[jadrian]

Either that or the problems you're experiencing are not KDE's fault.

Do you think I care whose fault they are?

[aussersterne]

Seriously?

GNOME is working. Therefore I use GNOME. I have better things to do than conduct the Desktop Inquisition to make a determination about whether or not KDE should be exonerated.

Re:Do you think I care whose fault they are?

[jadrian]

Sure, didn't suggest otherwise. That was in reply to "I guess they just like you better". Either that or... see reply.