The Last GUADEC?

An anonymous reader writes "How can we ensure, together, that this will not be the last GUADEC? Last year, during GUADEC, there was that running joke amongst some participants that this was the last GUADEC. It was, of course, a joke. Everybody was expecting to see each other in Brno, in 2013. One year later, most of those who were joking are not coming to GUADEC. For them, the joke became a reality. People are increasingly leaving the desktop computer to use phones, tablets and services in the cloud. The switch is deeper and quicker than anything we imagined. Projects are also leaving GTK+ for QT. Unity abandoned GTK+, Linus Torvald's Subsurface is switching from GTK+ to Qt. If you spot a GNOME desktop in a conference, chances are that you are dealing with a Red Hat employee. That's it. According to Google Trends, interest in GNOME and GTK+ is soon to be extinct."

Comments

Its NOT smartphones.

[Anonymous Coward]

Gnome sucks. Its a UI made not for normal users but for the designers imaginary friends.

Re:Its NOT smartphones.

[Anonymous Coward]

Not necessarily. I know a lot of "normal users" who are just fine with Gnome 3. What I've found is that the problem is not Gnome 3, the problem is that Gnome 2 went away. I think there would have been a much better situation if Gnome 2 had been maintained in parallel with Gnome 3, so that users had been able to have both installed and move freely between them until Gnome 3 was able to fully replace Gnome 2. It was just so unnecessary that we had to rename everything to Mate.

Re: Its NOT smartphones.

[Anonymous Coward]

What? Let Gnome 2 and 3 coexist in the same world? That's just crazy talk. What would people say if they tried to do that with something like Python? Or even the Linux kernel itself?

Re:Its NOT smartphones.

[serviscope_minor]

Not sure why this was modded funny. There is more than an element of truth in it. I can't speak from the perspective of a "normal user" since I'm not one. But I can speak from my own.

Firstly there's GTK. For those not aware, GTK is based on GObject. I've tried using GObject. It may be a very fine object system, and since it's based in C, I imagine that bringing over the runtime is relatively easy, but really it is no fun to use. And by no fun, I mean awful.

The primary reason is that it's all in C. Actually doing heavyweight dynamic style OO in C (basically like Python or Ruby or Javascript) is possible, but it is very, very verbose. Essentially you have to do vast amounts of stuff by-hand and it means that the program logic ends up being very sparsely interleaved with the heaps of required boilerplate. Not only that but the learning curve is very steep. IIRC Rust is a bit like C with native GObject support, so perhaps that will help things. But at the moment programming in GObject is unpleasant.

Secondly, frankly the UI is bad. They seem to be determined to abandon the long held principles of heirachal filesystems and the current working directory. Why oh why oh why when I start up a program in $HOME/projects/foo does the file dialog now default to "favourites" which is something not in the heirachy at all, or the last place I was working. This sort of change is completely unnecessary. If normal users don't start from anywhere else but $HOME then ignoring the current directory won't affect them at all. It only hurts power users.

The thing is, that's just one example. For another example, how many steps does it take to print a document of any sort at 6-up in a GTK program versus the appauling old style dialog like "xpdf"? The answer: lots. Adobe (of all people) proved it was unnecessary by making a really nice system that was simple for all normal stuff, but in the "advanced" box, the GUI options simply added things to an LPR line.

Examples abound of where GNOME is essentially "simplifying" to the point of making things less simple (how is ignoring $PWD unlike every other system simple?) especially for advanced users. The thing is advanecd users are the ones that hack on it and the ones that go to conferences.

If they systematically put off all advanced users, then basically it will be nothing but a commercial project. That's fine if they want it but it will kill off any ecosystem.

Oh and about GLib. A good fraction of the stuff is about providing things like linked lists, resizable arrays, hash tables, essentially all the sort of stuff that's in the STL, except much more verbose, vastly less efficient and not even remotely type safe. And two of those are unnecessary in C if you're prepared to (a) put up with complex macros and (b) have C++ like compile errors.

And other stuff just seems to be there for the sake of it, like the lexical scanner compared to (e.g. FLEX). They could simply have integrated the FLEX runtime into glib and used that as a scanner. It would have been better documented and have much nicer syntax that way.

I know people complain about C++ being complex, but once you add on a library the size of GLib, you've made almost a new language and the complexity arguments pretty much vanish at that point.

Don't get me wrong. For a C library, GLib is very impressive. It allows you to do all sorts of things with C that would otherwise require vast amounts of work and huge amounts of pain. But the problem is it's built on a very low level language and that makes life difficult.

Oh yeah, where was I. Gnome. One other example I remember from a while back.

My mum had an old computer. After Windows 98 became untenable, I put ubuntu on it, since she was feircely opposed to spending money on a new computer if the old one worked. Being young and naive I still did family tech support.

Anyway, it worked pretty well. I hd taught her what a heirachal filesystem was and she was happily able to arrange files in a nice organised manner. Actually

Re:Its NOT smartphones.

[serviscope_minor]

Gah.

Said Rust, meant Vala.

Re:Its NOT smartphones.

[Anonymous Coward]

I know a lot of "normal users" who are just fine with Gnome 3.

Lies.

Re:Its NOT smartphones.

[cheesybagel]

GObject is not the problem. If you are a C++ programmer which dislikes having to type macros all the time you can just use gtkmm.

The problem is they broke compatibility with GTK+ 2.x with the release of GTK+ 3 and the platform portability remains low. I know several developers who are revolting and porting their code to Qt instead of making the changes to migrate to GTK+ 3.

You can port Qt applications to mobile devices, Windows, and Mac OS X for example while GTK+ applications run well on Linux and poorly everywhere else. The Windows port has always been a bit of a mess even if it works. Slow and full of graphic glitches. The MacOS X port has been in limbo since like forever. The mobile port is non existent.

Re:Its NOT smartphones.

[serviscope_minor]

GObject is not the problem. If you are a C++ programmer which dislikes having to type macros all the time you can just use gtkmm.

You can, yes. Trouble is, not much of the infrastructure is written in C++ which makes hacking on the system some wort of hell.

Re:Its NOT smartphones.

[Anonymous Coward]

I know several developers who are revolting

Me too.

Re:Its NOT smartphones.

[KugelKurt]

Gnome sucks. Its a UI made not for normal users but for the designers imaginary friends.

Funny, I never liked the Gnome 2.x UI (which Xfce also mimics) but I like Gnome Shell a lot. I like it second best after Plasma Desktop.

Re:Its NOT smartphones.

[DrXym]

I think a better strategy would have been to offer GNOME 2 like functionality in GNOME 3. They sort of had this from the beginning in the fallback mode, and now have a proper "classic" desktop. Personally I think the default experience is extremely usable, certainly by people who see a desktop as a thing on which they do stuff rather than something to endlessly tinker with.

Re:Its NOT smartphones.

[DrXym]

C is the lowest common denominator and bindings including C++ can be and have been written that sit on top of it. So if you want C++ with GTK, use gtkmm.

QT bindings also exist though I suspect it would be far more difficult to write and maintain them and present them in a language-sane way due to the class hierarchy the bindings must sit over.

Re:Its NOT smartphones.

[HiThere]

The classic "fallback" mode was so unusable ugly that I quickly switched to something else after Gnome2 disappeared from the repository. I'm still not as happy with the replacement (KDE4) as I was with Gnome2, but then I was never as happy with Gnome2 as I was with KDE3.

I'm not really after a lightweight desktop, I'm after a usable desktop that doesn't interfere with my workflow. This means that I can't be needing to re-fix it every month or so.

Additionally the original Gnome3 wouldn't work with my hardware. I believe that this has been fixed now, but I find it's actually worse than Unity.

Re:Its NOT smartphones.

[Anonymous Coward]

Stopped reading at C++.

Re:Its NOT smartphones.

[heteromonomer]

That may be. The funniest thing is that Unity sucks even more. But still Unity is becoming popular (at least according to TFA).

The best thing I learned from TFA is that someone is developing an incognito version of CyanogenMod. When and if it gets rolled out, the guy deserves to be given money by every privacy-conscious Android user who decides to use it.

Re:Its NOT smartphones.

[dragon-file]

As a "normal user" leaning toward linux noob, I'd have to agree. Gnome 3 brought with it no inherent problems for me. In fact, at the time I encountered it, it was shortly after upgrading to mint 12 and, frankly, I didn't even know they had upgraded to Gnome 3. It was only after stumbling across an article online, more of a rant really, that I discovered that some people were having serious issues with 3.

Re:Its NOT smartphones.

[lytles]

i was an active user and *very* minor contributor in 1999-2005. gtk, glib and gobject are definitely awkward to use. some of that is due to being written in c, but much of the awkwardness is due to being written in a way that allows binding from other languages (bindings exist for javascript, python, java, c++, ...). eg, the callback "marshaling" made stepping thru code in gdb painful

i always hoped that someone would strip out the language-agnostic support and make a pure c library. but nobody seemed excited about the prospect

another problem was that gtk et al attempted to be incredibly generic. the language-agnostic support is one example. another is that i proposed a key binding that would break focus and guarantee that alt-f would activate the file menu and was told that there was "nothing special about the menubar" in the context of gtk. and then gnome was at the other extreme - attempting to be very minimal and rigid. in between was a vast no man's land

Re:Its NOT smartphones.

[Synerg1y]

I think a better strategy would have been to offer GNOME 2 like functionality in GNOME 3

I think common sense would have been to offer GNOME 2 like functionality in GNOME 3.

What's a little bit of functionality in place of slapping your own code in place of someone else's, surely the code will be at least be 100% better right?

Re:Its NOT smartphones.

[AliasMarlowe]

I know a lot of "normal users" who are just fine with Gnome 3.

Lies.

GP's statement is manifestly untrue, but not necessarily a lie. GP may fervently believe in some "imaginary users", and be utterly sincere. They go along with the "imaginary friends" that Gnome developers have. Faith overrides the fact that users are fleeing, as usual.

Re:Its NOT smartphones.

[hairyfeet]

Exactly, Gnome simply went the Windows 8 "Hey lets flip off our users and tell them to 'deal with it' because we don't give a fuck what they want" route and are now paying the price for that....and?

Its something any little shop guy could have told you, you give the people what they want or they go somewhere else. The gnome devs thought they had a captive audience, thought they could say "deal with it" and nobody would do anything about it, and instead the userbase evaporated...uh huh? I mean who couldn't have predicted flipping off your users would get them to abandon you? That the application devs would likewise go where the users are, so that you end up with the chart on TFA where the whole thing just nosedives.

I hope this serves as a lesson down the line, you can't force the customer to take shit if they have choices. Doesn't matter who you are, Gnome, Ubuntu, MSFT, you can't just throw a shitburger down in front of the customer if they can just go across the street and get a real burger, it just doesn't work. I hear that the gnome guys are trying to backpedal but its probably too late as once folks move away its DAMN hard to get them to move back so probably the best they can hope for now is to get that line to level off and stop the bleeding, but its gonna take years to bring that line back up if ever. For an example just look at how many abandoned Ubuntu for Mint, never to look back.

Re:Its NOT smartphones.

[hairyfeet]

Not really as i have found in the shop there is ALWAYS a minority that will like ANYTHING, hell I've met users that hung on to the last possible second to Windows ME because they really liked it and I have a handful that will be on Vista right up to support ending in 2017, no matter how funky or off or just plain terrible something is you'll find somebody that loves that shit, see Troll 2 being a cult hit as an example.

That said i have to throw a flag, bullshit on the field, when it comes to Chrome OS becoming the "dominant" OS, total crock of shit. Maybe if you live in Asia where ultra high speed and free WiFi is as common as dirt but it will NEVER become anything more than a niche everywhere else because like all thin clients once you lose Internet it becomes a giant worthless brick. Hell even Torvalds who bought the crazy expensive ChromeBook (Pixel I think its called?) is talking about putting Linux on it as the limitations of a thin client is starting to become grating, after using a rich OS most folks aren't gonna be willing to put up with the limitations of a thin client and all the spying and sharing data with the feds certainly isn't gonna give folks a warm fuzzy when it comes to Google products.

Sadly you wanna see what the future is go to your average phone store, its walled gardens controlled by the carriers that never get updates and which you are forced to toss yearly as the carriers refuse to support anything but the latest and greatest and even those get piss poor support at best. Frankly it bums me out as we were getting all this incredible hardware for cheap...then here comes the smartphones and its back to the 90s, horrible data caps, proprietary everything, locked down hardware, most of these phones have more in common with game consoles than they do PCs.

Re:Its NOT smartphones.

[Gerald]

FWIW we're in the process of porting Wireshark to Qt.

Re:Its NOT smartphones.

[SeanBlader]

+1 for Kurt, I like Gnome 3. It's easy, gets the hell out of my way, and lets me use the programs I want to use, rather than forcing me to use the operating system. It has the least amount of desktop clutter next to Windows 8 which takes it 2 steps too far, while still maintaining contextual awareness of where you were and where you are going. Really, it's the brilliant happy medium after operating system interfaces of the 20th century. I would say the Gnome team got it right. More of you should really give it a solid look, and take the few minutes to figure out it's use case that would really fit what you want to do.

They shot themselves in the foot

[msobkow]

Most people assume GTK+ is a dead end seeing as it's tied so closely to that abortion of a desktop known as Gnome 3.

Re:They shot themselves in the foot

[Anonymous Coward]

LGPL 2.1 vs LGPL 2.1?

Re:They shot themselves in the foot

[segedunum]

GTK was dead fifteen years ago, just no one realised it. When you have a toolkit that programmers are supposed to be using pulled out of another project (GIMP) as an emotional response to the license of another toolkit (Qt) you're already on to a loser. You have to win on the software, and producing a core development toolkit for GUI and desktop developers is spectacularly difficult. New features look cool and you have to keep moving things on but there are a spectacular amount of bugs to fix and that takes full-time manpower. Red Hat weren't going to plough lots of developer money into GTK because it made no money for them for the manpower they would have to put in to get it on a par with Qt and other GUI development software. It's not as if they were selling licenses or anything. You can't be emotional. Either free desktop software is good enough to compete or it isn't.

Qt was miles better fifteen years ago, miles better ten years ago, miles better five years ago and the gap has only widened. You throw in the LGPL license now and you have a situation where there is no reason whatsoever you wouldn't use Qt in a Linux based environment.

It's just a pity it has taken us so long and there has been so much blood letting for us to get to a situation where we probably have a de fact standard GUI development path decided upon by natural selection.

Re:They shot themselves in the foot

[gl4ss]

LGPL 2.1 vs LGPL 2.1?

well qt has a commercial option? I dunno how that's being less nice though.

Re:They shot themselves in the foot

[aliquis]

I'm all ok with people switching to Qt.

I'm ok with people writing applications for KDE to.

I personally don't agree with whatever guidelines KDE people have for how to design user interfaces though. Toolbars, ugly drag holders, vertical tabs with rotated text! WTF?! Tabs too wide for the window where you have to scroll between them and so on..

So much crap in the user interfaces. Also on this low-end machine which should had been killed a long time ago and be replaced by something much more powerful and will be so installing Qt and KDE programs in general pull in a lot of other stuff, like install digikam and get close to 500 MB of stuff. What's much better over in GTK camp is that you get many more programs to pick from, in general they are more independent (good and bad but then again some "KDE" programs (not really I guess) like Clementine doesn't botter with the common stuff anyway) and much smaller in size and you can pick a more freely of what you want without getting everything else at the same time.

The KDE philosophy may be fine IF you can throw more or less unlimited resources on it and IF their solutions was the best ones available and IF everything used it. But that's not the case. Now it may still be quite ok if you at least have the first one solved..

I guess OS X give less to pick from in general but often better / more clevery designed software and user interfaces and with Windows you of course got a shitload to pick from and something likely works good for you.

Re:They shot themselves in the foot

[JanneM]

I don't have any religious attachement to either toolkit, and I don't care one bit which one I use as a user. But there's GTK bindings to all kinds of languages out there since it's relatively easy to do, whereas there's far fewer for Qt. I can't seem to find any decent bindings for Scheme for instance. I can only assume it's because Qt is quite tightly tied to C++, and languages that don't mesh well with it will have trouble interfacing with it.

Re:They shot themselves in the foot

[serviscope_minor]

well qt has a commercial option?

Well, at this point LGPL 2.1 is a perfectly fine commercial option. Honestly if your lawyers say it isn't then you should probably fire your lawyers and get some new ones. After all, the famously well lawyered Apple seem entirely happy to integrate the LGPL Webkit into their very proprietary operating systems.

Commercial means commercial, not "palatable to ingnorant lawyers" :)

Re:They shot themselves in the foot

[cheesybagel]

Great. Then tell me of all the wonderful Qt applications people commonly use. I use GIMP, Inkscape, Gnumeric GTK+ applications quite frequently and have used zero Qt applications so far.

Re:They shot themselves in the foot

[JanneM]

"Wow, you really had to think about that one, didn't you."

Nope. I'm interested in Ubuntu Mobile, and I recently wanted to see if there were bindings to my current favourite language to play around in. Since I looked for it only a few days ago it's what came to mind.

Again, I have nothing much against Qt. With Unity I can stay away from both Gnome and KDE and their multi-year infight. And with so much memory and disk in even very small machines today I simply don't have to care what toolkit other people use to develop their software. Having two (or three) sets of libraries installed and running at the same time doesn't really affect performance in any meaningful way any longer.

Re:They shot themselves in the foot

[uncle slacky]

The recent announcement of a merger between LXDE and Razor-Qt is another nail in the coffin of GTK+3, and should produce something rather lighter than KDE.

Re:They shot themselves in the foot

[Anonymous Coward]

For starters

Re:They shot themselves in the foot

[Desler]

In what way? Qt is multi-licensed as LPGLv2.1, GPLv3 and commercial. How is GTK+ "better"?

Re:They shot themselves in the foot

[cheesybagel]

Ok so VirtualBox and VLC it is. VLC's interface is horrid but that is probably the developers fault. VirtualBox has a pretty good interface.

The rest seems to be mostly closed source applications. Some of which are in common use like Skype or the Google applications.

Re:They shot themselves in the foot

[cheesybagel]

I assume the Google application they are talking about is something like Google Maps since Chrome seems to be using GTK+.

Re:They shot themselves in the foot

[Desler]

Well, at this point LGPL 2.1 is a perfectly fine commercial option.

Except if you develop for any of the platforms that only get support under the commercial license. LGPLv2.1 is fine for linking code with proprietary, but the commercial license from Digia encompasses far more than just a proprietary-friendly license.

Re:They shot themselves in the foot

[cheesybagel]

s/Maps/Earth/

Re:They shot themselves in the foot

[tibit]

What about it? All you need to provide is a static library (.a/.lib) with the objects for the rest of your project. As long as anyone can statically relink the thing with their copy of Qt, you're good.

Re:They shot themselves in the foot

[tibit]

Writing a gtk-to-Qt translator say re-using the LLVM frontend would be a very nice Ph.D. project :) Whether the developers realize it or not, GIMP, Inkscape and Gnumeric will end up as Qt applications or they'll become irrelevant. I mean fucking come on, only a year ago did Gimp manage to wrestle GTK to work on OS X. That's stuff that was mostly said and done in - wait for it - 2001. Yes, Qt 3.0 had OS X support, albeit only for commercial users.

I've worked on manually porting Inkscape to Qt and it was pretty much a menial job once you figured out common patterns of code transformations that were to be applied. I've lost interest in it after about a month's worth of work, but there's nothing fundamentally hard about it.

Re:They shot themselves in the foot

[chris.alex.thomas]

and that favourite language is??

Re:They shot themselves in the foot

[Anonymous Coward]

While I agree with most of your points, there's the burning question of would we have had Qt move to LGPL if GTK had never existed.

Re:They shot themselves in the foot

[Anonymous Coward]

GTK wasn't an "emotional" response to QT, but a practical response. If a license is preventing you (by law) from doing what you need to do, or abiding by the terms of that license would be impractical, then the only remaining option is to do it yourself.

With that understood, there is no reason to bring up "emotions" at all -- unless of course your own perspective is emotional.

Re:They shot themselves in the foot

[JanneM]

Ah - Scheme, as I mentioned in the original comment.

But once again, I have little opinion either way between Qt and GTK. I've worked with both at times, and with well-supported languages (such as Python) there's no major difference in usability. Most difference people really care about is really about KDE and Gnome, and I can frankly do without either.

Re:They shot themselves in the foot

[KugelKurt]

Great. Then tell me of all the wonderful Qt applications people commonly use. I use GIMP, Inkscape, Gnumeric GTK+ applications quite frequently and have used zero Qt applications so far.

Common people use Skype, VLC, etc. and not Gnumeric.

Re:They shot themselves in the foot

[KugelKurt]

Whether the developers realize it or not, GIMP, Inkscape and Gnumeric will end up as Qt applications or they'll become irrelevant.

Doubt it. Contributors to OpenOffice/LibreOffice cope with its homegrown VCL toolkit since many years because it's easier than porting everything to Qt.

Re:They shot themselves in the foot

[segedunum]

Trust me I've heard all the arguments, and the 'bindings' argument is usually somewhere near the top of the list. Most of the GTK bindings are pretty poor and really, what's the point of having all the bindings in the world when your underlying API quality is so poor? If it's poor in one language it's poor in another.

Re:They shot themselves in the foot

[aliquis]

Didn't knew about it but I like Razor-Qt.

I also like the desktop in Parted magic which I think is among other things running LXDE? Parted Magic for whatever reason seem very fast to (I don't get why Ockular feel quicker in some distributions / setups than other.)

Re:They shot themselves in the foot

[segedunum]

Scribus, Krita, Digikam, Kontact...... The quality shows through in their development. Evolution is all but abandoned these days despite the years of funded full-time effort into it and the GIMP's interface is a poster for how poor GTK GUIs are. Gnumeric is a bizarre thing to be using when Libreoffice is around.

GTK apps are crushed under their own weight of bugs and required manpower.

Re:They shot themselves in the foot

[segedunum]

We most probably would have eventually anyway, but it's a moot point. Qt was only ever good enough because a company could put dedicated time and resources into it where GTK has never had that. Regardless of license no one in the Mac or Windows development world would ever have took GTK seriously.

Re:They shot themselves in the foot

[segedunum]

It was anything but a practical response. If you're pushing inferior software on people (in almost twenty years GTK has never been on a par with anything else in terms of GUI development) and selling it on the basis of a license then that is simply an emotional response. The software is either good enough or it isn't - that's practical. We've had to wait a long, long time with the Linux desktop being all decimated to find that out.

Re:They shot themselves in the foot

[Carewolf]

Well Apple did put a BSD licensed wrapper called WebKit around the LGPL WebCore, and they also have a slightly more "liberal" interpretation of the LGPL than most, since they ship all the iOS specific changes to the LGPL WebCore library as precompiled binaries only.

Re:They shot themselves in the foot

[jedidiah]

It was an imminently practical response.

You can't have a gratis product dependent on commercial components. It's a competitive death sentence in a market where none of the commercial alternatives have the same limitations.

This is the same reason that proprietary Motif, CDE, X servers, and OpenGL were all discarded.

The GPL is often portrayed as anti-business but that's nothing compared to expensive per-seat developer licenses.

Re:They shot themselves in the foot

[darthscsi]

Writing a gtk-to-Qt translator say re-using the LLVM frontend would be a very nice Ph.D. project :)

No it wouldn't. I doubt that would even be a publishable paper in a decent conference. (I appear on both http://llvm.org/developers.cgi and http://llvm.org/pubs/ )

Re:They shot themselves in the foot

[MrEricSir]

Well, at this point LGPL 2.1 is a perfectly fine commercial option.

Not if your commercial plans involve getting in to certain app stores.

Re:They shot themselves in the foot

[kthreadd]

And now both both Motif and CDE is LGPL.

Re:They shot themselves in the foot

[KugelKurt]

It was an imminently practical response.

No. The GNU project had proper UI frameworks for years before KDE existed: Gnustep. However it had practically no support from any GNU member. It took KDE to enter the scene and the GNUs suddenly ran wildly around like headless chicken. So instead of deciding to use a proper frameworks set (Gnustep) they chose to chop an image editor apart and rip out its toolkit that was never designed to be a general purpose one.

Re:They shot themselves in the foot

[JanneM]

"But you can't expect the Qt developers to create and maintain bindings for everyone's pet language out there..."

Those bindings are very incomplete and for a dialect I don't use. And of course I don't "demand my bindings" or anything like it; that would be silly, even presumptuous. All I did was point out one small area where in my (no doubt rare and unusual) case GTK ends up being the better choice.

Re:They shot themselves in the foot

[KugelKurt]

The KDE philosophy may be fine IF you can throw more or less unlimited resources on it

"Will Stephenson from openSUSE giving a talk on how to make a more lightweight #KDE at #akademy2013 . KLyDE (the optimized version) has a memory footprint of only 119 MB!"
https://plus.google.com/106620870500514995860/posts/GuXFD4wqj3V

Btw, funny how you attempt to prove "KDE philosophy" the Extragear application Digikam that is not even part of mainline KDE.

Re:They shot themselves in the foot

[chris.alex.thomas]

ok, so basically a language nobody gives a flying fuck about......ok, just wanted to clear that up....

Re:They shot themselves in the foot

[JanneM]

"ok, so basically a language nobody gives a flying fuck about......ok, just wanted to clear that up...."

You know, there's good ways to make people care about and like a project. Constructive responses, attempts to help the curious potential new convert - even a simple straightforward admission that some particular thing is not (yet) possible can go a long way towards creating a good feeling about it.

That's not what I got from posting about my small, idiosyncratic corner case. I've gotten a mix of verbal abuse (including all-new to me words for feces), diatribes about why the alternatives all suck, and people basically telling me I'm an idiot for wanting to do something odd or off-beat.

This may surprise you, I know, but it does not make me more likely to jump on the bandwagon, try to find a solution with the software or want to communicate with the user community in any way. It makes me a lot less likely to want to do so.

Avoiding Qt and its user community, and just continue with GTK2 seems like a good short-term solution to me. Longer term I'll just focus more on Android I guess. People in that development community are really friendly and approachable.

Re: They shot themselves in the foot

[MatheoDJ]

A default desktop environment will probably do more to further Linux adoption than fancy widgets ever would... Simplicity has been and will be the number one draw. It was simple to use the same OS that we used at work 20 years ago, when computers were thousands of dollars and floppy disks were incompatible across OS's. It was simple to just use the default OS that came with the Dell my mother bought 10 years ago, which is why I was never able to get her using Linux for more than a week data time. It was even simpler for her to buy and use a Mac 5 years ago, and she never looked back - it was simple, intuitive, and reduced the number of family tech support calls to virtually zero. Simple and elegant (and functional) doesn't have to be expensive. I hope we are finally headed in that direction.

Re:They shot themselves in the foot

[Rakarra]

I have rarely heard of people using Skype on GNU/Linux and I do nothing but support GNU/Linux.

It's difficult to chat with friends on IM these days without using Skype, since that's what everyone's moving to. I'm not too fond of it, but my skype plugin for pidgin works ok. At least it's not Steam.

Re:They shot themselves in the foot

[tibit]

Well, said translator would need to include a knowledge base with translation rules. Those rules would need to be expressed in some high level language, with formal semantics and all that. Presumably, to save on the drudgery, many rules would be generated from some descriptions of semantics of "similar" concepts in Gtk and Qt. It would be rather far from simple stuff like pattern matching, and even that was subject to plenty of papers on code generators I'm sure.

Done properly it'd not only be a dissertation-sized piece of work, it would require some original results to implement it in a way that's fun to deal with. I mean, come on, people do master theses with nothing but a description of a C-derived programming language and formal semantics of the same with some proofs. We're talking about something perhaps an order of magnitude harder in scope.

Sure it could be hacked together from a bunch of M4 macros, it's Turing complete after all, right? But that's not what I'm talking about.

Re:They shot themselves in the foot

[tibit]

I've looked at some of the papers, and I don't think that such translation would be anywhere in the same order-of-magnitude ballpark of complexity. It's a much bigger job than, say, proving some properties of SSA optimization. Of course I'm only an amateur and whatever I know about programming language semantics is self-learned, but I just don't see how it's anywhere near "easy" if you want to do it right. My gut feel is that, scope-wise, it'd be more along the lines of formal semantics of C and compcert work of Xavier Leroy et al. After all, you need some description of semantics of both the source and target frameworks, and then a way of describing or even just aiding/guiding the transformation between the two. In order to be fit for human consumption, it'd all need to be done in some sort of a declarative framework so that you don't deal with the drudgery. Well, maybe these days it's all a walk in the park for the theorem-prover-wielding yung'uns :)

Re:They shot themselves in the foot

[tibit]

And no, I'm not proposing to do it myself. I just think you haven't thought it all through - it only appears simple; that's why nobody has done it yet. Leveraging the LLVM front-end to extract interfaces from GTK and Qt code would be a tiny aspect of the job, done out of convenience.

Re:They shot themselves in the foot

[chris.alex.thomas]

I'm not interested in constructive responses, since it was those very constructive responses that led us down this path in the first place, to a world where we accomplish a tiny fraction of our potential and if thats the case, fuck constructive responses.

if my responses make you think about jumping off the bandwagon, then well, since you contributed nothing in the first place, you surely won't be missed when you're gone, right?

Re:They shot themselves in the foot

[segedunum]

It was an imminently practical response.

You can't have a gratis product dependent on commercial components. It's a competitive death sentence in a market where none of the commercial alternatives have the same limitations.

You can't have any market based on software that isn't good enough. The end. That's why GTK was complete total and utter waste of time and effort.

KDE and Gnome are still comparable

[darkHanzz]

Although Qt is going strong, KDE and gnome seem both to be in a downwards trend..
http://www.google.com/trends/explore?q=gtk%2Cqt%2Cgnome#q=gtk%2C%20qt%2C%20gnome%2C%20enlightenment%2C%20kde&cmpt=q
Ah well, the higher abstraction level that C++ offers, does make sense for a UI framework.

Re:KDE and Gnome are still comparable

[segedunum]

Why should anyone be modded up who is basing the general usage of desktop environments on Google trends?

Re:KDE and Gnome are still comparable

[91degrees]

Gnome does have a slight bias in that there are more hits for other meanings of gnome. Look at the highlight markers. One of them refers to a 3D animation about gnomes

Re:KDE and Gnome are still comparable

[RabidReindeer]

And of course Google Trends is a totally unbiased source.

Do no evil and all that shit, right?

I was hoping for confirmation from Netcraft, myself.

Re:KDE and Gnome are still comparable

[mangobrain]

It's no less valid than the original Google-trends-based premise that interest in GNOME is declining. If anything, the fact that we have a GNOME-focussed article about this, but no KDE-focussed one, shows how bogus it is. TFA actually has a lot more to say, so perhaps the summary would have been better not to call out this one small point.

Re:KDE and Gnome are still comparable

[Desler]

Qt is much more than UI framework.

Re:KDE and Gnome are still comparable

[msclrhd]

These Google trends are only part of the picture.

Once you have found what you are looking for (e.g. Qt or Gtk+ documentation), you are likely to bookmark it and go directly to it, so your search result will only apply once even if you visit the Gtk+ docs a lot. Also, if you have downloaded and installed the Gtk+/Qt/KDE documentation, you don't need to search Google to find it!

Q: What are the Debian, et. al. popularity contest stats for the Gtk+, Qt and KDE libraries? What about the dev packages? Documentation packages? Applications making use of these libraries?

A user is not likely to search for Gtk+ or Qt (and maybe not even KDE or GNOME). They are more likely to search for things like "Amarok", "Natilus" or even more likely things like "PDF viewer for Linux". Even then, if they are looking for a specific application they are likely to go through something like the Ubuntu Software Center, or follow the instructions to install a cool new app from a blog/review site like http://www.webupd8.org./

A developer is also not likely to perform a generic search for the UI framework or desktop environment either. They will either search for a specific query mentioning APIs or packages such as "How do I set the value of a GtkProgressBar?", "gtk_label_set_text HTML styling" or "libqt5-gui". Even then, they are likely to try sites like stackoverflow first. They are also likely to ask questions on the relevant forums, IRC channels, etc.

Q: What are the search/question trends for gtk+/gnome and qt/kde tagged questions on stackoverflow?

Q: What are the visitor stats like and comment counts like on the various Qt/KDE/Gtk+/GNOME pages, blogs and wikis (e.g. Allen Day's GNOME blog).

Q: What are the trends for the number of commits and developers to the Gtk+/GNOME/KDE/Qt source code repositories?

I just started working on gnome

[maweki]

I just became the maintainer of a small games project in gnome and I have to say, the lack of (wo)manpower really shows. There are other projects that have many hundreds of untriaged bugs (most of them unconfirmed. We're not talking about unfixed here). There are only a handful of people doing really cool stuff and about nobody doing the menial labour of just making builds stable or working with one-off-contributors who sent in patches on their own.
But all in all I don't believe gnome's development cycle is unsustainable in the foreseeable future, even with shrinking interest in the desktop as a whole.

Re:I just started working on gnome

[segedunum]

Gnome's development cycle has been unsustainable for some time. Relying on a toolkit that continually struggles for manpower to fix bugs is no something you want to be doing as an application developer.

Re:I just started working on gnome

[KugelKurt]

I just became the maintainer of a small games project in gnome and I have to say, the lack of (wo)manpower really shows.

Even tough I share the criticism here about GTK, your example does not really mean a lot. In many big FOSS projects small subprojects are driven by maybe just a single developer. It's exactly the same in striving projects like KDE which also has fringe subprojects like some games or screensavers. Heck, recently even two Calligra apps lost their maintainer and nobody could step in.

Sad, if true

[jandersen]

It is sad, in a way, although not surprising to me.

Sad, because it was once so promising; GNOME was once my absolute favourite desktop, but when they started becoming more and more a Windows clone, I lost my faith in them. And then they started removing useful features, upsetting their core community - those who were on Linux because it is OPEN, extremely configurable, very inclusive etc - and the GNOME developers became more and more unapproachable and sectarian. I suppose, in a way they chose to follow their own closed set of ideals and lost their way.

Now I use KDE - it is not perfect, but I don't need perfect, I only need good enough, and KDE is good enough for my purposes.

Re:Sad, if true

[jez9999]

The Windows desktop is extremely popular, and you think becoming more of a Windows clone is the problem? Windows GUI is pretty well-designed. The problem is they totally fucked up GNOME 3 and programming GUIs in C without any decent dev tools is not exactly fun.

Re:Sad, if true

[Tough Love]

Doing that OOP stuff in plain old C is just a travesty. Should have ended many years ago, but for all the self-interested spin and PHBs involved. Anybody remember that company that lost $40 million writing a file browser for Gnome (Nautilus) that ultimately was completely discarded except for some of the expensive artwork?

Re:Sad, if true

[kthreadd]

Well you can just use any of the many language bindings. Gtkmm for C++ for example, or Vala which a lot of new Gnome development uses. Vala translates to C, so there's nothing extra needed at runtime.

Re:Sad, if true

[91degrees]

I agree with all your criticisms. But what ui doesn't have the destructive window closing button issue? As for #6, has anyone implemented a start menu alternative that is actually any good?

Re:Sad, if true

[RabidReindeer]

It is sad, in a way, although not surprising to me.

Sad, because it was once so promising; GNOME was once my absolute favourite desktop, but when they started becoming more and more a Windows clone, I lost my faith in them. And then they started removing useful features, upsetting their core community - those who were on Linux because it is OPEN, extremely configurable, very inclusive etc - and the GNOME developers became more and more unapproachable and sectarian. I suppose, in a way they chose to follow their own closed set of ideals and lost their way.

Now I use KDE - it is not perfect, but I don't need perfect, I only need good enough, and KDE is good enough for my purposes.

The problem is that Gnome's creator, Miquel de Icaza is a bitch for Windows. Among his sins you can include something like, but even more horrible than, the Windows Registry. Add to that the hubris of the Gnome 3 team and it's a recipe for something that looks good but doesn't do what you want it do do.

KDE's bouncing icons were its biggest turn off for me. I want GUI that "just works" and doesn't attempt to be an art form or a distraction and doesn't require switching off "vanity features" before I can use it.

As a developer, it's also rather important that the UI toolkit is easy to inject into projects, doesn't have bizarre interactions with other subsystems (or itself), supports multiple languages in a developer-friendly way, and last, but not least, has usable documentation. Including sample code and instruction manual. A set of "javadocs" isn't enough.

Re:Sad, if true

[Tom]

but when they started becoming more and more a Windows clone, I lost my faith in them.

ditto.

Gnome was very promising once, I even worked on it for a while. But this exactly, there was a point where it turned into a me-too project, where ideas for making things better were shunned in favour of making things familiar, which at that time meant copying windows.

Gnome is a major example of Free Software fucking it up because of bikeshedding and copying instead of innovating.

Re:Sad, if true

[mangobrain]

Have you seen the hoops that older versions* of Qt jumped through to maintain ABI compatibility? It wasn't really C++, it was C-with-classes-plus-a-bunch-of-horrible-macros. Based on when the project started, I think C was the better choice on the grounds of stability, compatibility and portability. If you don't want to do it in C, use the C++ bindings, which avoid (most of) the ABI compatibility issues by virtue of being an OO wrapper around something with a stable ABI.

* I say "older versions" because I don't work actively with Qt, but have read things here and there suggesting the situation is improved in Qt 5.

Re:Sad, if true

[Anonymous Coward]

I was driven away from contributing when I tried to submit some bug fixes for a GNOME app I used pretty frequently a number of years ago.

It was a real pain in the ass to submit the patches. The mailing list got no traffic, I couldn't find a bug tracker, the sole maintainer didn't respond to direct emails, and IRC was the only place he hung out.

Whenever anyone tried to submit a patch over IRC, he'd respond with something like, "Your patch doesn't tickle my balls right. I can't accept it."

He'd literally say " doesn't tickle my balls right " time and time and time again when presented with patches. I got this treatment, and I know several other patch submitters got the same treatment.

It was really absurd, you know? I've dealt with some unusual folk throughout my professional programming career, but nobody was as weird and obsessed with scrotal tickling as this GNOME developer was. His attitude drove me right away from all GNOME software, and I've been using KDE ever since.

Re:Sad, if true

[loufoque]

Doing that OOP stuff in plain old C is just a travesty.

It's fine, really.
The use of C++ that Qt makes is much more of a travesty.

Re:Sad, if true

[kthreadd]

Gtkmm is horrid. It makes me want to go back to using MFC instead, it's truly that bad. Once you've started using Qt, you can't seriously go back to a toolkit like Gtkmm.

What's so bad about it? You give very few arguments for your case.

Re:Sad, if true

[The Cat]

How the fuck does anyone spend $40 million writing a fucking file manager?

Re:Sad, if true

[tibit]

Demonstrate using some code examples, please, and elaborate what is the travesty you speak of.

Re: Sad, if true

[AvitarX]

How did gnome become a windows clone over time? The most windows like version was 1.x IMO, and that was like windows up through vista. As time went on its config got less control panel like, the layout changed to so etching fairly unique to gnome, but closer to osx than windows (2.x). Then they tried something completely new and unprecedented (it didn't work, but apple appears to have taken some of the concepts into os x).

Gnome was a windows clone for a very short period, and at its best (2.x interface e just felt uncluttered, and relaxing to use for me) when they invested in ui research and went with it (sun spent the money I think).

KDE has always felt windows clonish, but more customizable, but the interface has always felt stressful to me (just a feel). Windows 7 may be my favorite ui though (it makes flipping through many windows of the same application easier than any other system, snapping windows, and a fast searching launcher, if only it had and always on top button.

Re:Sad, if true

[tibit]

What a load of uninformed bollocks. I've been using Qt since late 1.x versions, and things got much improved with every major release since then (2.x, 3.x, 4.x, 5.x). What have the macros got to do with maintaining ABI? What macros are you exactly taking about? The macros that substitute empty strings and are only used by moc (signals, slots, Q_SIGNAL, Q_SLOT, Q_INVOKABLE)? Qt provided an introspection data extractor (mod) since the first release. The functionality provided by moc was never available in C++, and it's not yet possible to implement this using merely a C++ compiler. Even functionally equivalent stuff is unlikely to make it into C++, given that AFAIK there's no strong, maintained proposal in WG21.

The "hoops" that Qt has jumped through to maintain ABI are the hoops any C++-based library that cares about BC will have to jump through. Qt is not special in this respect. You might be confused since many other Qt-unrelated C++ libraries don't give rat's ass about maintaining BC.

Re:Sad, if true

[Oligonicella]

"Sure it also still runs the usual work stuff, but that has not been the reason for its existence for a long time now."

Nice vacuous affirmation.

Re:Sad, if true

[dbIII]

The problem is that Gnome's creator, Miquel de Icaza is a bitch for Windows. Among his sins you can include something like, but even more horrible than, the Windows Registry.

That wasn't him. I also accused him of gconf, on this very website, and he replied that it wasn't him but another of the gnome group that came, set up that travesty, and left it half finished. The facts checked out. I suspect that gconf is completely dead now with gnome3 and good riddance to it, whether the idea was sound or not is immaterial since the execution was not sound.

Re:Sad, if true

[loufoque]

MOC, duplicates everything that is in the standard library, braindead usage of COW everywhere, avoids usage of good language features, type unsafe when it does not need to be, etc.

Re:Sad, if true

[Anonymous Coward]

Eh?

I have to use a virtualized Win7 for work. here are some observations:

1. Destructive Logoff command is the top-level default when I just want to lock or disconnect the session
2. Destructive window closing command button is co-located with minimize and restore buttons
3. Applications steal focus when I am typing
4. Some applications don't close when I DO intend to click the closing button, they just hide themselves
5. Can't hide the window decorations for full-screen apps ( unless the app has this function built-in )
6. The Start Menu is a mess

1: pretty sure it is a simple config change to change this.
2: watch what the fuck you are doing and it's no problem, it's been that way for the last 20 years.
3: with win7 I have had the opposite problem, popups don't come forward most of the time ( especially OS message popups). Maybe it is just the specific applications you use?
4: That is an application choice, turn it off in the applications preferences.
5: I don't use a potato to view my PC output, even in Linux I don't care about a few pixels on the top of my screen. That said explorer and browsers just need a simple F11 keystroke to hide the window decorations. If explorer can do it I would assume any other native applications can as well, but can't be assed to confirm since I never use that "functionality."
6: I agree, 9x / XP start menu was organized into sub directories much better than 7 or *shudder* the mess that is win8.

Re:Sad, if true

[Rich0]

The Windows desktop is extremely popular, and you think becoming more of a Windows clone is the problem?

Keep in mind that 99.999% of those running Gnome are running it on a PC that could be running Windows, and the reason they're not using windows isn't because they don't want to fork out the $0 to buy a license for their PC and the 0min of time it takes to get it installed.

Gnome users already jumped through a lot of hoops because they think it offers something that Windows doesn't offer, so moving back towards Windows is not a given.

Hey, if there is a good idea in Windows go ahead and borrow it, but you have to keep in mind that your user base is not aunt tilly, but a bunch of geeks who don't mind installing Linux and going through all kinds of pain to configure modelines and printers. (Yes, I realize that both of those are much easier today, but much of the Linux userbase are people who didn't mind doing that work in the past.)

Re:Sad, if true

[kafka.fr]

"moc" provides introspection you would hardly get another way, at the cost of a single macro (Q_OBJECT) and automatically generated code. That's still OOP and C++.

Qt does not duplicate the STL, it provides similar functionalities, but arguably simpler to use. Indeed it uses COW almost everywhere it might make sense. Sometimes it's a real nice thing to be able to return a QString by value for (almost) free - don't forget all of this was created long before we had rvalues and move semantic in C++.

"Good" language features are usually "modern" language features (sure, not always, I know). Because Qt tries to be available for older compilers, even awful proprietary ones (Visual C++ anyone?), it has to do tradeoffs here and there.

About typesafety, well... I won't really disagree on this one ;-) but it's a real minor concern really, given the enormous benefits the whole framework provides.

By the way, I've been using Qt for... oh my, 16 years already??

Re:Sad, if true

[tibit]

MOC provides introspection. Until introspection gets wrapped into C++, there's no way around it. No, template metaprogramming doesn't provide anything to bridge that gap. Seriously. When you use GObject, you're writing all the introspection data yourself, or vala does it for you just like moc does it. How is writing introspection data by hand going to make you more productive I just don't know. Aren't we supposed, you know, to use tools to get rid of menial tasks? Why do C/C++ users generally scoff at code generation if it's a part of a framework? That's just silly IMHO. MOC is a good thing. It saves you from doing stupid shit that the compilers stubbornly refuse to do. Why is that so hard to understand?

Duplication of the standard library was necessary since Qt started when the "standard" libraries weren't so standard at all and had bugs and inconsistencies that were impossible to work around. IIRC Qt 4.x still had to support MSVC 6. There's nothing particularly egregious about having a yet another implementation of a container library that the toolkit can depend on instead of working around the shortcomings and bugs in standard C++ library implementations. I mean, give me a fucking break, you seriously think std::string is anything more than mental masturbation? It's useless at anything that's not an academic exercise.

To see what a large-scale project that uses C++ standard library had to go through over the years, have a look in lyx-devel. LyX is nowhere near as portable as Qt is. It simply won't compile on a lot of platforms that Qt itself compiles on, even though LyX uses Qt, through a platform abstraction layer going back to the times when they used an X-only toolkit and were unsure about whether Qt would "catch on".

COW is not braindead. It simply defers copying until it absolutely has to be done. That's what you had to do to work around lack of rvalue references in pre-C++11. C++ standard library implementations do the very same thing, yet you seem not to bitch about that...

Where is Qt type unsafe when it does not need to be? Do realize that runtime-checked pre-Qt5 signals and slots are typesafe . A type-mismatched connection won't ever crash your program. It may fail and you get a runtime warning on the debug output, but that's it. You can have type safety without full compile-time checking; you seem to confuse the two.

Exactly what good C++ language features does Qt avoid? All that time I was thinking that C++ was a reasonably good indicator at what C++ features do actually work across multiple compilers that people use - as opposed to what the fanboys advocate people must be using or else!, hurr durr.

Re:Sad, if true

[jedidiah]

Windows is a legacy support OS and always has been. It thrives purely on the fact that everyone bought into the idea that they needed it to run MS-DOS programs that are long forgotten.

It's a vicious cycle that causes everything else to be cast aside and this effect predates Linux entirely.

Re:Sad, if true

[RabidReindeer]

The problem is that Gnome's creator, Miquel de Icaza is a bitch for Windows. Among his sins you can include something like, but even more horrible than, the Windows Registry.

That wasn't him. I also accused him of gconf, on this very website, and he replied that it wasn't him but another of the gnome group that came, set up that travesty, and left it half finished. The facts checked out. I suspect that gconf is completely dead now with gnome3 and good riddance to it, whether the idea was sound or not is immaterial since the execution was not sound.

It's possible that Icaza's name has been used in vain these many years, but I've read enough news reports about his baby where he seemed to be taking credit for Windows-like features.

Truthfully, I'm not sure if there's a distinction between "gconf" and the underlying configuration data. To me, "gconf" is the app that was supposed to maintain them, and I'm not sure if it existed way back when I discovered that my home account was unusable thanks to some obscure uncorrectable corruption in the gnome configuration. I'm not an expert on the internals of gnome. Partly because there were never decent design docs on it.

Re:Sad, if true

[loufoque]

"moc" provides introspection you would hardly get another way

Not true. You can get it with plenty of other ways, including normal usage of the embedded preprocessor.

That's still OOP and C++.

Associating C++ with OOP is a problem. It's a multi-paradigm language. OOP has very few valid use cases. If you want OOP abused everywhere, use Java.

Qt does not duplicate the STL, it provides similar functionalities

It has its own set of containers, strings, iterators, all of which use are not conforming to the standard concepts. They might have better implementations in some cases (for example a custom hash map implementation optimized for a given usage pattern might be faster than the general-purpose one of your standard library implementation), but the design is not as well-crafted and is significantly less flexible.

Indeed it uses COW almost everywhere it might make sense.

COW never makes sense unless it's used to implement partial data sharing within a larger data collection. That's not the case in Qt. In Qt it's just used because the bad design forces the framework to copy many objects even though it's not needed.

Sometimes it's a real nice thing to be able to return a QString by value for (almost) free

I don't understand how that's related. Returning a local variable or temporary variable by value is always "free" (doesn't call copy constructors) regardless of COW. It's called NRVO, though that it is a silly name. It's not really an optimization. It's more a matter of ABI.

don't forget all of this was created long before we had rvalues and move semantic in C++.

rvalue references allow to distinguish lvalues and rvalues at the language level. There was no problem implementing move semantics by handling rvalues explicitly. Before C++11, people used swap when they needed to move. Qt containers didn't even have swap before 4.7 (2010).

"Good" language features are usually "modern" language features (sure, not always, I know). Because Qt tries to be available for older compilers, even awful proprietary ones (Visual C++ anyone?), it has to do tradeoffs here and there.

Current Qt only works with Visual C++ 9 (2008) and higher, Visual C++ 10 (2010) or higher is even highly recommended. Decent C++ support has been available since Visual C++ 7.1 (2003).

About typesafety, well... I won't really disagree on this one ;-) but it's a real minor concern really

Type safety, a minor concern!?
The point of C++ is that it has a powerful type system that can catch errors. Type errors are ones of the worst type of errors you can have in a program.

Re:Sad, if true

[Tom]

and I've been using KDE ever since.

I knew something was missing from my original post:

KDE sucks just as badly as Gnome does. In fact, the total failure of the desktop environments on Linux is part of what drove me to OS X. I can't speak about today, since it's been years since I've last used a Linux desktop (I still run plenty of servers on Linux, but that's commandline only), but back when I switched, it was an awakening. It was like having driven a 1950s car all your life and then you buy a modern car and you realize that there are all these amazing tiny things that make everything just so much better, smoother and less painful to use.

Plus you don't have to take it apart and fix something yourself every thousand miles or so.

Re:Sad, if true

[Eravnrekaree]

Windows 95-Windows 7 was a popular, well designed UI, that was indeed popular and actually was well designed. Emulating the taskbar start menu model made since becasue they are actually practical, well thought out models that made sense. Windows 8 is a total disaster, for pretty much everyone from a programmer to granny. So trying to emulate that has simply extended the disaster to Gnome.

Re:Sad, if true

[msclrhd]

Qt3 to Qt4 required you to change your code to support Qt4.

There will also be porting needed for Qt4 to Qt5 (e.g. QtWidgets being a separate module, changing QtGui include to QtWidgets). See http://qt-project.org/wiki/Transition_from_Qt_4.x_to_Qt5 for the full list of changes.

I used QAudioFormat from QtMultimedia in a project I was developing, This got moved from the QtMultimedia package to the qtmobility-dev package and from the QtMultimedia folder to the QtMultimediaKit folder.

Are you going to use the Qt Q... variants or the KDE K... variants of the different classes? Oh, and you cannot use them interchangeably (e.g. the file open/save dialogs).

Also, there is more emphasis in Qt on the QtQuick platform which is another moving target. Even then, are you going to use pure QtQuick, or use the new Ubuntu Phone/Tablet APIs they are creating?

Re:Sad, if true

[HiThere]

If Vala would ever get ready to use. The documentation seems to require that you already know the language, or possibly just GObject, before you start using it.

Re:Sad, if true

[kafka.fr]

"moc" provides introspection you would hardly get another way

Not true. You can get it with plenty of other ways, including normal usage of the embedded preprocessor.

I wrote "hardly", not "impossible". Now you may be of the elitist kind of people which assume everyone should be able to write and use something like Boost::MPL for breakfast, but in the real world easy tools are sometimes welcome.

That's still OOP and C++.

Associating C++ with OOP is a problem. It's a multi-paradigm language. OOP has very few valid use cases. If you want OOP abused everywhere, use Java.

Notice I said "OOP and C++": I did not say "OOP is C++" nor the other way around.

Qt does not duplicate the STL, it provides similar functionalities

It has its own set of containers, strings, iterators, all of which use are not conforming to the standard concepts. They might have better implementations in some cases (for example a custom hash map implementation optimized for a given usage pattern might be faster than the general-purpose one of your standard library implementation), but the design is not as well-crafted and is significantly less flexible.

Less flexible, quite true. But again, overall easier to use. Now their containers provide the usual "::iterator" and "::const_iterator", making them looking quite similar to STL containers. Even QString provides begin() and end().

Indeed it uses COW almost everywhere it might make sense.

COW never makes sense unless it's used to implement partial data sharing within a larger data collection. That's not the case in Qt. In Qt it's just used because the bad design forces the framework to copy many objects even though it's not needed.

The thing is, quite often data is shared in a heavyweight GUI app. It may be mostly useless in Qt itself, but it's quite usefull when you *use* Qt. Now we can talk about "good" or "bad" design: I tend to prefer a design which may not be academically perfect, but clean and easy to read and use. When you have to maintain several millions LOC, perfection may have different meanings.

Sometimes it's a real nice thing to be able to return a QString by value for (almost) free

I don't understand how that's related. Returning a local variable or temporary variable by value is always "free" (doesn't call copy constructors) regardless of COW. It's called NRVO, though that it is a silly name. It's not really an optimization. It's more a matter of ABI.

ABI which is not standardized, last time I checked. Here the optimization is provided by the framework, not relaying on the compiler. That said, if you don't like it, then don't use it... as simple as that.

don't forget all of this was created long before we had rvalues and move semantic in C++.

rvalue references allow to distinguish lvalues and rvalues at the language level. There was no problem implementing move semantics by handling rvalues explicitly. Before C++11, people used swap when they needed to move. Qt containers didn't even have swap before 4.7 (2010).

Don't take my poor example for more than it is... There's more in sharing data than just moving. Granted, it's pretty useless for a "Hello world" label.

"Good" language features are usually "modern" language features (sure, not always, I know). Because Qt tries to be available for older compilers, even awful proprietary ones (Visual C++ anyone?), it has to do tradeoffs here and there.

Current Qt only works with Visual C++ 9 (2008) and higher, Visual C++ 10 (2010) or higher is even highly recommended. Decent C++ support has

Re:Sad, if true

[loufoque]

MOC provides introspection. Until introspection gets wrapped into C++, there's no way around it. No, template metaprogramming doesn't provide anything to bridge that gap. Seriously. When you use GObject, you're writing all the introspection data yourself, or vala does it for you just like moc does it. How is writing introspection data by hand going to make you more productive I just don't know. Aren't we supposed, you know, to use tools to get rid of menial tasks? Why do C/C++ users generally scoff at code generation if it's a part of a framework? That's just silly IMHO. MOC is a good thing. It saves you from doing stupid shit that the compilers stubbornly refuse to do. Why is that so hard to understand?

I'm all for automatically generated code. I use various kinds of automatic code generation everyday, and it's even related to my research topics.
MOC just isn't a good way to do it. There are plenty of ways to have better introspection.
Yes, you can have introspection with template meta-programming, and that's probably the most useful kind of introspection ever, since introspection is mostly useful for generative programming, and generative programming is better done at the meta level. But I digress.

There's nothing particularly egregious about having a yet another implementation of a container library that the toolkit can depend on

Except it's not an implementation of the standard containers. It's not conforming at all and largely inferior in design.

To see what a large-scale project that uses C++ standard library had to go through over the years, have a look in lyx-devel. LyX is nowhere near as portable as Qt is. It simply won't compile on a lot of platforms that Qt itself compiles on, even though LyX uses Qt, through a platform abstraction layer going back to the times when they used an X-only toolkit and were unsure about whether Qt would "catch on".

I have worked on many projects with millions of lines of code of C++, working on many compiler versions, operating systems and hardware architectures. Targeting standards with multiple implementations is indeed much more work than targeting a single piece of software, simply because the code might rely on non-standard behaviour or the implementation might have some bugs.
But that's pretty much obvious. The same problems exist everywhere, be it interoperability between video-conference systems built by different vendors to different browsers rendering websites in standard HTML/CSS differently. You can also choose to stick to a single standard implementation for all your targets (some C++ compiler/standard library combinations are available on all operating systems of note).

COW is not braindead. It simply defers copying until it absolutely has to be done.

Here is a better idea: don't copy until you absolutely have to do it.
See other parts of the thread for my replies on the subject.

C++ standard library implementations do the very same thing, yet you seem not to bitch about that...

The C++ standard have never allowed COW for containers, and has outlawed it for string since C++11.
This is for obvious iterator validity reasons.

Where is Qt type unsafe when it does not need to be? Do realize that runtime-checked pre-Qt5 signals and slots are typesafe . A type-mismatched connection won't ever crash your program. It may fail and you get a runtime warning on the debug output, but that's it. You can have type safety without full compile-time checking; you seem to confuse the two.

A type safe program doesn't cause type errors at runtime. Qt slots do not help in writing type safe programs, though they do point out when type errors occur.

Exactly what good C++ language features does Qt avoid? All that time I was thinking that C++ was a reasonably good indicator at

Re:Sad, if true

[loufoque]

I wouldn't call "decent" the C++ support in Visual 7.1. Not even in Visual 2010. Visual 2012 is getting close to be "decent", although "decent" is pretty far from "good".

The kind of code I write is often more advanced than the average Boost code (I am also a Boost developer).
Yet I can tell that MSVC10 is quite decent. It's still swarmed with fundamental bugs everywhere, but MSVC11 and even MSVC12 are not really significantly better in that regard, despite sporting more C++11 features. Nothing that cannot be worked around.

For C++ that doesn't require too much advanced features (no advanced usage of ADL, SFINAE, overloading and partial specializations), MSVC7.1 is ok enough that it's possible to target it.

It's a minor concern in the context of using Qt. Feel free to write typesafe code around it.

So I should wrap all the code around signals registering and invoking?

Re:Sad, if true

[AvitarX]

1) agree, but rare, and I use command-L anyway
2) annoying, though it's generally not destructive (to changes) without confirmation. This is a major pet peeve of mine, and always but space in there with a custom KDE button layout
3) annoying, a lot, and the opposite happening at times too (things staying behind and hanging other operations because I can't click them)
4) this is a pretty universal thing
5) Almost every app supports this, and most all use F11 now
6) I like it, type three letter and click. Shouldn't need to browse it at all.

Also, the side and top snap are great features. The taskbar is the best solutions I've found for managing lots of windows (blanking out all other windows when hovering on a thumbnail makes it super fast to flip through open windows, and find them when selected.

It may not be the best over-all. I also really like a customized KDE, or gnome2 (always on top being a big win), but Win7 is a contender. the snapping windows (which is in KDE too) and new taskbar being the new things that moved it beyond Win95, and on par with OS X, KDE and Gnome.

Re:Sad, if true

[Grishnakh]

KDE is the answer to both of your questions.

2) While in stock form on most/all distros, KDE is set up with the same destructive window closing button issue as Windows and Mac, it's easy to reconfigure it to change the locations of all the window buttons. You can add some extra space between the close button and the minimize/restore buttons, or you can even move the close button to the other side of the window if you want. It's completely configurable. They probably keep it just like Windows/Mac to avoid confusing people.

6) The start menu (actually "K menu") in KDE is organized very well. When it opens, the first tab is "Favorites", and shows your most-used applications, plus a box where you can type in the name of an application to search for. Or you can switch to the "Applications" tab to get to a menu of all your installed applications. In there, all the applications are grouped according to their function: Development, Games, Graphics, Internet, Multimedia, Office, Settings, etc. This is completely unlike Windows where all the applications are a big mess, only grouped by the application's maker with no indication as to what it actually is or does. Then, inside one of these groups, each application has its own icon, plus is usually listed by its description, along with its name, such as "Image Editor", and then in smaller gray text, "GIMP Image Editor"; or "Media Player" and "Kaffeine" in smaller gray text; or "Spreadsheet" with "LibreOffice Calc" in smaller gray text. (note: some of this is probably distro-specific. I'm using Mint KDE for reference.) A new user doesn't have to know the name(s) of the spreadsheet programs on a Linux/KDE computer like this, he or she just needs to open up the K-menu, go to Applications, click on "Office", and then look for the programs that say "Spreadsheet", and LibreOffice Calc will be there (along with any others; that's all I have on mine at the moment). OR, they can click on the K-menu, and type "spreadsheet" into the search box, and all the spreadsheet programs will be listed there for selection. OR, for a shortcut, they can type Alt-F2 which brings up a litle search box at the top of the screen, type "spreadsheet" into that, and all the spreadsheet programs will be listed there. In my case, since I have only one (Calc), when I type Enter it automatically launches Calc. Windows doesn't do any of this stuff.

Re:Sad, if true

[Grishnakh]

Top-of-the-line hardware, Aeron chairs, a big fancy office with insanely high rent in some prime location, ridiculously-high salaries (esp. for the executives), all-expenses-paid trips to "important" conferences, I could go on and on.

When you're playing with someone else's money instead of your own, there's no reason to be frugal. Might as well have a field day with it.

Re: Sad, if true

[mangobrain]

Wow, what an angry response. MOC itself is one huge hoop, the necessity of which proves my point that Qt hasn't ever really been C++. The slots & signals stuff *can* be done in pure C++, see libsigc++ (as used by the glib and gtk C++ bindings), similar functionality in boost, or the contents of standard . Introspection is overrated IMHO, but it does now exist in the gobject world as well.
I know exactly what sort of hoops have to be jumped through for ABI compatibility, and I am aware of how many libraries piss all over them, which is why I stand by my statement that writing gtk in C made sense.

Re:Sad, if true

[Tough Love]

Agreed, the MOC was and is idiotic and unnecessary. But lies firmly in the "wart" category compared to the inoperable liver disease of GTK(1/2/3/)(+)

Re:Sad, if true

[Darinbob]

But most of the early UIs were built with C. Yes it was clumsy at times, but C is still the universal language, even today.

With C++ you bring in some baggage immediately; do you use STL (still not 100% uncontroversial) or rely on Boost (lots more naysayers)? Or are you object oriented or instead have heavy use of templates? Do you have a small embedded variety of C++ (no use of exceptions) or require full C++ standard?

A better thing perhaps is to have a basic C-compatible library (ie, easily usable without special bindings from C, Fortran, Pascal, assembler, etc). Then around that you have a C++ wrapper or utility kit. If someone wants Ruby or JavaScript or Lua bindings then separate wrapper libraries can be created without touching the base library. Yes, it's true that you could create the base library in such a way that other binding libraries become nasty messes, but libraries can be done better than that.

Re:Sad, if true

[Tough Love]

Qt3 to Qt4 required you to change your code to support Qt4.

Annoying, but pretty easy from first hand experience. Compared to nigh on impossible according to reports I have seen from devs attempting similar "updades" across GTK versions that share little other than name and warped attempt to make C sit up and do OOP tricks it just isn't suited to.

Re:Sad, if true

[Darinbob]

C is still a great choice though. Embedded systems are a very large market and Qt sells into that market. Simpler is better when you need to cram the code into something small, not everything is as large and luxurious as a smartphone or PC.

Re: Sad, if true

[Tough Love]

MoC is just one, survivable irritation. I programmed it out of my sight without too much effort. Compared to never, ever being able to fix C + GTK to be anything remotely resembling pleasant to work with.

Re:Sad, if true

[Tough Love]

Doing that OOP stuff in plain old C is just a travesty.

It's fine, really...

Black knight: "'tis but a scratch."

Re: Sad, if true

[tibit]

Signals and slots can be done in C++. Introspection - can't. Moc is not about signals and slots, it's about introspection. Introspection data is then used for signals and slots, but that's just a happy coincidence, because introspection is used for other things too. I'd have hoped people would actually look at the code without repeating the tired old non-arguments. You know, it's pretty damn easy nowadays, no need to download/unarchive the tar ball.

Introspection is not overrated. It's what you need to do the things that are done in Qt, day in, day out. Qt nowadays supports both compile-time-checked and runtime-checked bindings, something that you don't get without introspection. Libsigc++ only solves the compile-time end of the deal.

Re: Sad, if true

[tibit]

Huge hoop, my ass.

Wait a minute, does that mean that you program without using code generation from things higher level than C/C++? I almost can't believe that you can do any big project without speeding up development and avoiding errors and the mundane by generating code in a way that's not really possible merely utilizing the template metaprogramming. Almost anything you do eventually needs a lexer or parser (data formats, comm protocols), state machines, etc. Writing those even in C++ is a big hassle, since the compiler isn't really designed to do arbitrary computation at compile time, and that's ultimately what you need in order to be productive. Yeah, you can do it all by processing bytecodes at runtime, but that bloats your product with fragments of tools that are not needed past build stage. Bloat is bad.

It's really silly IMHO to insist that the make utility, compiler, assembler and linker are on a pedestal, and you're OK using those, but no other tool could ever come into the picture because it's somehow unkosher. Get over it - moc is a code generator, like many others that you're supposed to be using to stay productive. LLVM has got tablegen for a reason - good luck implementing it using template metaprograms to run during C++ compilation, ha ha. Are you going to bitch that LLVM hasn't ever really been C++, too?

Template metaprogramming quickly degenerates into mental masturbation where you express things that take a few lines of easily comprehensible imperative script in a convoluted mess of declarative functional code that makes people who do real functional programming in Haskell or OCaml feel quite happy at not having to deal with C++.

Do you really believe that there is any benefit in re-expressing simple code generators into dozens of template metaprogramming headers that not only slow down the compilation, but are a royal pain to debug should things go wrong? I consider eigen, for example, to be a fairly top-shelf example of what it takes to get C++ to produce well performing numerical code. It's a huge freakin' clusterfuck, and some things still can't be done due to limitations of either C++ spec or of implementations.

I am, in a way, happy about people "disliking" moc. I expect they have a similar religious aversion to other code generating tools. That only makes me more productive and more competitive. So, yeah, thank you.

Foreshadowing

[Zouden]

Is this a preview of what might happen to Linux distros at some point in the future? Android has had a bigger impact than anyone expected. I wouldn't be surprised if it leads to Linux becoming more marginalised (servers only) and fewer people adopting it on the desktop.

Re:Foreshadowing

[MichaelSmith]

If anything the growth of Android makes its shell the predominant UI for Linux.

Re:Foreshadowing

[Zouden]

The Google Trends comparison is interesting.

Re:Foreshadowing

[king neckbeard]

I'm not sure about that. GNU/Linux distros seem to be the ones still leaving open options for people who want to get shit done. Not that there aren't a lot of distros that are jumping on the toy bandwagon, but there are still options.

Re:Foreshadowing

[Kjella]

Is this a preview of what might happen to Linux distros at some point in the future? Android has had a bigger impact than anyone expected. I wouldn't be surprised if it leads to Linux becoming more marginalised (servers only) and fewer people adopting it on the desktop.

Quite possibly, with AIOs like this coming out now in September, 21" IPS touchscreen, keyboard and mouse using Android. It'll take a while before the OS and apps catch up, but fundamentally this is what most people are looking for in a "real" computer, it's not the CPU, not the GPU, not the RAM it's a big screen, keyboard and mouse. It's the kind of PC you can crank out a novel at or work on a big spreadsheet (not in Excel, but there are alternatives), anything where it's not about x86 compatibility or computer horsepower. That said, the big migration potential is certainly Windows users but if there's a landslide like in the mobile market Linux is sure to notice it too.

Re: Foreshadowing

[tepples]

21 inches for one application? Google has strongly resisted any attempt to add tiled or overlapping window management policy to stock Android. This means applications have to be coded to each hardware manufacturer's proprietary window management API. I can dig up Google employees' explanation of this "all maximized all the time" policy and why Cornerstone isn't in CyanogenMod if you want.

Re: Foreshadowing

[Kjella]

21 inches for one application? Google has strongly resisted any attempt to add tiled or overlapping window management policy to stock Android. This means applications have to be coded to each hardware manufacturer's proprietary window management API. I can dig up Google employees' explanation of this "all maximized all the time" policy and why Cornerstone isn't in CyanogenMod if you want.

Well I might be the odd man out among tech-savvy users, but I run most applications maximized on my 24" screen, I've tried doing it differently, I've been on Linux for some years and tried multiple desktops, I've tried working with multiple monitors as well but I come back to the KISS model - maximized apps and rapid switching. I don't think I've ever used such a thing as drag and drop between apps, it's select in one app, Ctrl-X/C, switch, Ctrl-V. I very often flip-flop between two apps, like say code and specs by clicking the same app icon to show/hide instead of side-by-side. It might not be for everyone but I think many users will get by just fine without tiles or overlapping windows.

Re: Foreshadowing

[tepples]

Would you use a desk big enough to hold only one sheet of paper? Alt-tabbing between your textbook and where you are writing your solutions to the exercises can get tiring.

Re: Foreshadowing

[Mad Merlin]

Well I might be the odd man out among tech-savvy users, but I run most applications maximized on my 24" screen...

This is frustratingly common, it is incredibly painful when I see other people do this. Currently I have 19 windows visible on the current virtual desktop and a total of 76 windows open across 5 virtual desktops. Needless to say, alt tabbing through 76 windows doesn't cut it.

Re:Foreshadowing

[msclrhd]

How about pizza vs chips [http://www.google.com/trends/explore?q=android%2C+linux#q=pizza%2C%20%20chips&cmpt=q]. The search for "chips" is flat and a lot lower than the growing trend for pizza.

Conclusion: no one likes chips anymore, so all chip shops, McDonalds, etc. are going to die out!

Re: Foreshadowing

[CronoCloud]

I've been on Linux for some years and tried multiple desktops, I've tried working with multiple monitors as well but I come back to the KISS model - maximized apps and rapid switching.

Glad I'm not the only one, though I've only been using Linux since 2002. (With usage of KDE1x, FVWM, fluxbox, E16, Gnome2, and XFCE over the years)

I've also used windowshading in the following manner:

1. have one non-maximised window (something like X-chat or IDJC) I want to keep right on top, for easy access but out of the way. Right click > Always on Top.

2. Then use mousewheel to windowshade it as needed.

Re:Foreshadowing

[Darinbob]

I have an android phone, and I can not possibly imagine in any universe using that on a desktop computer. But then, a year ago I probably would have said the same thing about the Windows Phone UI...

Re:Foreshadowing

[MichaelSmith]

It just needs to support non-maximised windows, and that is on the way for tablets anyway. I can imagine Android being used now for a server appliance. One which needs a simple control panel.

Re:Foreshadowing

[Darinbob]

Well, they have widgets which are sort of non-maximized windows.

vala

[samjam]

Shame - vala is a really cool c(+) style language that hides a lot of the glib rot that was too hard to use.

The same trend for the whole PC infrastructure

[maweki]

Looking at Gnome and GTK as an Example for them going extinct specifically is pretty stupid. You see declining trends for microsoft, dell and KDE as well while playing with Google Trends.

Re:The same trend for the whole PC infrastructure

[Ploum]

Looking at Gnome and GTK as an Example for them going extinct specifically is pretty stupid. You see declining trends for microsoft, dell and KDE as well while playing with Google Trends.

Indeed, it's not like Microsoft, KDE or Dell are declining, it is imposs

Oh, wait!

Re:The same trend for the whole PC infrastructure

[Clsid]

Correct me if I'm wrong, but all that measures relate to are Google searches. Would be better to actually have download numbers or even sample data of desktop usage.

ChromeOS...

[Anonymous Coward]

... is becoming the dominant OS? Now in which universe is that?

Re:ChromeOS...

[KugelKurt]

How funny then that Chrome browser chose GTK....

Chrome only uses GTK for minor stuff like the Open/Save window. The actual Chrome UI is written in Google's own Skia library, not GTK.

Yup Gnome 3 sucks

[DMJC]

It's sad but true, Gnome 3 with it's stupid tablet interface completely sucks. Gnome is trying to double down on fail and it'll lead to complete extinction within a few months. They need to massively reverse course but too much ego will probably prevent it from happening. The Linux desktop has basically shot itself in the foot right when it's finally achieving mainstream gaming success. Personally I'm banking on GNUstep actually getting finished and offering an osx-alike experience on Linux. Gnome was a really nice desktop but by choosing to rush into tablets they've pulled a Microsoft and shot their desktop users in the face. The desktop isn't going anywhere, it's Gnome that's gone away.

Re:Yup Gnome 3 sucks

[swalve]

Gnome had the right idea (merging tablet and desktop interfaces), they just did it wrong. Microsoft, with a couple of exceptions, did it mostly right. Maybe I'm the only one who thinks that, but after 5 minutes of playing with a Windows 8 tablet, I was hooked. I can do tablet stuff AND desktop stuff? Sold.

Re:Yup Gnome 3 sucks

[DrXym]

You whine about GNOME 3 being like a tablet (though it is clearly nothing of the sort) and then hold out for GNUStep offering an OS X like experience. When GNOME 3 and OS X are very clearly aiming for the exact same space in terms of user experience, usability etc.

Re:Yup Gnome 3 sucks

[Clsid]

I thought the same as you regarding GNUStep, but it is too messy and the mindset towards creating an attractive GUI is simply not there. They want to achieve perfection by emulating NeXT, when they should just be trying to copy something like OSX which looks nice.

Linux is either a Fluxbox (or similar) affair, or you are better off with Windoze or better yet, with OSX. It's the fonts, the hardware, the games and the it just works attitude. Sad to see projects going these way but it wasn't hard to predict. I kind of feel the same way as Miguel De Icaza regarding OSX right now.

Re:Yup Gnome 3 sucks

[BlackPignouf]

Linux Mint + MATE has replaced my Ubuntu + Gnome/KDE, and probably will be replaced some day by Mint + Cinnamon.
I don't care which directions Gnome 3 and KDE 4 are going.

Re:Yup Gnome 3 sucks

[Clsid]

Windows 8 is aiming for the tablet experience. OSX might be merging some stuff from the iPad, but to me, Apple has been doing great at holding out the unusable parts of the tablets in the desktop OS. The addition of the launchpad is good and it will lead the way to deprecate the odd style of having an Applications folder in the Finder. Gnome in the other hand, tried to create something artsy but failed to achieve the whole look & feel. I guess it could have worked if they had better default fonts and not being so extreme as removing the maximize and minimize button.

Re:Yup Gnome 3 sucks

[DrXym]

Actually I think it's the opposite for once.

GNOME 3 got right things that Microsoft got wrong. For example Microsoft are bring back the start button (not menu) because jamming the mouse into the corner is non-obvious. GNOME 3 always allowed you to click Activities and discover the corner behaviour later. Metro doesn't bother to show you any visual context when you're in metro of what is happening on the desktop whereas GNOME puts the windows into its launcher so you can rapidly see them. Microsoft heavily relies on swipes and gestures to bring up overlays which don't work on the desktop whereas GNOME relies on conventional menus and a control panel.

In fact I don't see why people say GNOME 3 is a tablet interface at all. It's certainly a better place to produce a tablet interface from (larger buttons, scope for gestures etc.) but it's still a desktop window manager. Conversely Microsoft very clearly redesigned Windows so that it tablet friendly.

Re:Yup Gnome 3 sucks

[couchslug]

"Personally I'm banking on GNUstep actually getting finished and offering an osx-alike experience on Linux."

To succeed, copy success.

Re:Yup Gnome 3 sucks

[DrXym]

I think GNOME went too far by cutting configurability. The inability to choose fonts and some other basic settings is greatly annoying. It doesn't mean it has to offer a kitchen sink of options (which IMO is even worse), but as a rule of thumb if OS X offers an option then GNOME 3 probably could too without compromising usability. But fundamentally I don't believe the design is broken or unsound. It's a very slick, attractive and usable user interface and for once Linux has a desktop which actually holds its own vs Apple and Microsoft.

It's also entirely customisable so there is no reason that people who don't like some of the things it does have to put up with it, e.g. use the "classic" mode or MINT extensions and it resembles GNOME 2 while still benefiting from a fast compositing desktop.

Re:Yup Gnome 3 sucks

[dkleinsc]

Gnome had the right idea (merging tablet and desktop interfaces)

That's not the right idea at all.

Really easy to do on a tablet:
- Poke a particular point on the screen
- Swipe and other gestures
- add your signature or otherwise scribble in cursive
- turn the screen

Really easy to do on a desktop:
- mouse click
- move the mouse to a corner of the screen
- mouse hover
- press a key or a simple combination of keys

It's possible to do desktop-like things on a tablet, and some tablet-like moves on a desktop, but trying to use one kind of input on the other device is like trying to hammer a nail with a screwdriver - you can make it work, but it's a bit awkward and difficult. The only point of overlap is mouse clicks versus finger-pokes of large areas of the screen.

Different input methods implies different user actions implies different interfaces.

Re:Yup Gnome 3 sucks

[dbIII]

It shouldn't take five minutes to find the things that should be on the screen and in your face.

Re:Yup Gnome 3 sucks

[Luthair]

Windows 8 is an abomination as a desktop operating system. As of Windows 7 I would have said it was unlikely that I'd use anything else as my primary desktop, but after putting Win8 on my laptop at release (and leaving it there until now) there is no way I'll be switching to it on the rest of my PCs. (8.1 solves none of the real complaints)

Re:Yup Gnome 3 sucks

[Rich0]

Maybe I'm the only one who thinks that, but after 5 minutes of playing with a Windows 8 tablet, I was hooked. I can do tablet stuff AND desktop stuff? Sold.

Sure, if you're running on a tablet then a tablet UI makes tons of sense. However, the problem is that it is clumsy when you're running on a desktop.

I don't get the case for merging the UIs. I can see where you might try to move in a convergent direction where possible for consistency, but the desktop UI is HIGHLY optimized already after decades of evolution. To think that a revolutionary change is going to be an improvement is pretty audacious.

Re:Yup Gnome 3 sucks

[Eravnrekaree]

Thats a joke right? Windows 8 is a total useability disaster for most desktop users and I know many who throw up their arms in despair. Its a train wreck. The desktop and mobile form factors require completely different user interface models. The taskbar and start menu works best and is simply the most practical for a desktop. On a mobile device you are dealing with a small screen, no (real) keyboard, no mouse, an imprecise touch pointing interface that has less precise pointing accuracy that a mouse, none of these are limitations found on a desktop system, so hence a different UI might be needed for a mobile because of the severe limitations, but make no sense on a desktop where limitations do not exist. I cannot see mobile devices being anything more than a niche device for use when on the go because of these very reasons, the devices are extremely inefficient for actual content creation, anything more than a few one liner messages. These are acceptable limitations when you are walking down a sidewalk or on a subway, but they are absurd when you are at a desk at work where a desktop machine would be far, far better and in that case the confined nature of the mobile device is pointless, you have plenty of space on your desk for full desktop input output devices for a much better computing experience. If mobile devices came first,and the desktop came later, we would be talking about how the mobile device was going extinct because the desktop can offer much more computing power, a 100 times larger screen, a full size keyboard a real mouse, much better than trying to work on a 3" chiclet keyboard. Are people really going to walk into a store and say "I want to trade in my 20" monitor, my full size keyboard and my mouse for a 3" screen, with a chiclet keyboard and 1/30th the power". This i going backwards. Mobile devices are niche products and people will suffer from greatly reduced productivity and user experience if they try to use them in lieu of a real full featured desktop.

Re:Yup Gnome 3 sucks

[HiThere]

Unfortunately, MATE depends on a lot of Gnome2 applications which are no longer being maintained. Cinnamon is probably a better long-term bet...but not if Gnome3 really goes away.

Re:Yup Gnome 3 sucks

[KugelKurt]

Gnome had the right idea (merging tablet and desktop interfaces)

That's not the right idea. The right idea is to create a set of building blocks that can easily be assembled into very different UI paradigms without compromising the others. KDE's Plasma framework has already been used to create desktop, netbook, tablet, and mediacenter shells.

Re:Given the # of comments here it's already dead!

[flyingfsck]

GTK is the Gimp Tool Kit. I don't see Gimp going away any time soon.

GUADEC?

[bmomjian]

I am sure I am not the only one who doesn't know what GUADEC is, and in fact even the event homepage (https://www.guadec.org/) doesn't spell out what it is. It is the GNOME Users And Developers European Conference.

Re:GUADEC?

[MortenMW]

I was wondering about the same thing, I've never heard about GUADEC before. Thanks for clearing it up.

Re:GUADEC?

[citizenr]

Its ok, I didnt even know Gnome had any users left.

Re:GUADEC?

[Nimey]

If Slashdot only had editors...

Re:GUADEC?

[Just Some Guy]

I am sure I am not the only one who doesn't know what GUADEC is

I'm guessing "Gnome User Aggrevation/Deprecation Enhancement Committee", based on their past output.

Re: GUADEC?

[CronoCloud]

Thank you Anonymous Coward for giving me a good laugh today, the moment I saw that I heard Christopher Judge in my head saying it.

Yes, people it is Indeed funny if you've ever watched SG-1 even a little bit:

http://www.youtube.com/watch?v=xYnzZ2rczSo

Re:GUADEC?

[afranke]

You must be kidding. The first line on the event homepage says "The GNOME Conference" and the first paragraph after the menu, on the picture of gathered contributors taken at least year's GUADEC, reads: GUADEC is the annual conference of the GNOME community, held in Europe since 2000. GUADEC 2013 will be held in Brno, Czech Republic, a city which has played host to several other successful GNOME-related hackfests in the past. People modded you Informative for that comment. I guess I shouldn't be surprised that the slashdot crowd doesn't bothered checking facts.

Re:GUADEC?

[bmomjian]

No where does it say what GUADEC means, only that it is a GNOME conference. The larger point is that it wasn't even clear the news item was related to GNOME until almost the end of the news summary.

gnome three

[Anonymous Coward]

Qt is just nicer to use and these days - it even works fast enough.
I liked the look and feel gnome2, but in its early days its components were glitchy and not very usable. However, when it matured -- it got replaced with gnome3, which is (or at least was) a nightmare.
I used xfce4 instead (yes, it wasn't exactly the most mature ones either, but worked better).
Now, about 5 after the release of kde4, i tried a kde4 based live system and was very surprised -- it was fast enough!
4.0-4.8 were all sluggish, but all sudden, the latest version seems to be just fine. Now, i'm afraid of the project getting replaced with kde5 or something.

I blame removing configurability

[zakkie]

Gnome 3 was a fuckup, but it started way back, when Havoc Pennington declared that too many options confused users. That was the start of the slippery slope that led us to this scenario. Taking away options completely instead of just offering basic & advanced configuration options was a fucking stupid idea. A desktop or any interface needs to get out of the way and make your day-to-day experience as painless as possible, but Gnome was hijacked by look-at-me designer types with nothing better to do than find ways of breaking shit that worked pretty fucking well. End result? A clusterfuck that nobody wants to use.

wonderment

[vexon]

What has been happening with GTK and GNOME is demonstrating effects of drugs abuse and hard madness from developers.

Re:wonderment

[syockit]

You mean Guano decline syndrome?

Re:Can't say I'm surprised

[Clsid]

In my case I have abandoned Linux for the desktop some time ago, and just keep Gentoo around for fiddling with it. In my case I went back to Windows for gaming and OSX for the desktop experience, after having tried console gaming and desktop Linux for quite a bit.

No

[The Cat]

People are not abandoning the PC to use phones tablets and "services in the cloud." That is propaganda designed to sell you phones, tablets and services in the cloud.

Phones, tablets and services in the cloud will never replace the PC, because a desktop or laptop computer is the proper control form for the human body.

People want a full keyboard, a full-size monitor and a mouse. They don't want to do real work on a 2" x 3" screen.

This "exodus from the PC" is pure bullshit advanced by mobile device companies to get you back on the hardware upgrade treadmill so they can sell you a new device every two years.

Let me say it again: it's BULL. SHIT.

Re:No

[TheDarkMaster]

Very well said

Re:No

[Anonymous Coward]

It's only true if you have employees beneath you + administrative assisstant and or secretary,,,,at that point,, you dont use your computer so it becomes irrelevant to own a computer since they moslty do e-mails but these same people think that it applies to the rest of their company.

Just check who always wants to move to a tablet,,,,it's usually the person who doesnt use the PC other than e-mails.

Re: No

[AvitarX]

And yet I see lawyers typing/editing briefs in court from an iPad with citrix (and lil keyboard case). And taking notes. True, they may not be typing the bulk of it, but work is getting done.

At home I am seeing people not replacing computers as they die.

Computers aren't going away, but they are becoming office devices, with laptops getting replaced by citrix and an iPad (portability and battery life are winning for the type of person that reads figures and types short emails), home browsing getting replaced with safari and an iPad.

I'd say 1/3 of the work that used to be done on computers surrounding a trial is now being done on tablets, and it's leading to about a 25% reduction in the stock of computers in the war room/court room. Just one isolated example, but it's not total bullshit that things are changing dramatically. I moved recently, and didn't setup my desktop due to not wanting to sacrifice desk space, now, if I have desktopy stuff to do at home, I just stay at work a little longer, it's rare.

Re:No

[JDG1980]

People are not abandoning the PC to use phones tablets and "services in the cloud." That is propaganda designed to sell you phones, tablets and services in the cloud.

But there have been two important changes. First, smartphones and tablets have become good enough that users with limited needs (web surfing, Facebook, simple games) often find them preferable to full-blown PCs, since they are easier to use and don't require as much annoying maintenance. Secondly, regular PCs reached the "good enough for average users" point about 7 years ago. Most people are still just fine with a Core 2 Duo or Athlon 64 X2 system from 2006. Sure, if you do heavy work with photo/video editing, or HPC, or want to play the latest Doom clone, you probably need something better, but only a relatively small percentage of users fall into that category. For everyone else, upgrading to a new PC, especially in these tough economic times, seems like a waste of money.

All this means that the desktop should become more oriented towards power users, not less. The casual users have already bolted in favor of iOS or Android. People who use a desktop PC these days have good reason for doing so, and don't want it to try to imitate the portable ARM devices that they've already rejected as inadequate for their needs.

Re:No

[Eravnrekaree]

Even a laptop is subpar to a desktop , to have the best ergonomics you need to be able to seperately position the monitor and the keyboard, you need a desktop PC for the best ergonomics . Remember back in the 90s the concerns over RSIs and the trends towards ergonomic keyboards for desktop computers? all of that was well advised. Why people now want to actually trade in their comfortable ergonomic desktop system for something that is much more painful and contortionist to use, for a 4 inch screen and try to subject themselves to the pain of using vastly inferior IO devices and a cramped tablet form factor that there is no way that this can come close to the comfort of a desktop setup, vastly inferior to a desktop is beyond me. It seems like they have gone the other direction Away from ergonomics now, to lets make it popular to use torturious form factors with the tablets and the laptops are simply an ergonomic disaster that are hard to use and actually are painful. You also do not want the EMF emissions from the main electronics of the motherboard and the power supply so close to your body so you want all of that positioned a distance away. We are probably going to see an epidemic of cancers caused by handheld tablets and brain cancer from smartphones in a few decades. Brain cancers can be slow moving and may not become noticeable for decades. Then you have the cloud nonsense which really is an attempt to get people to store their lives in the cloud so it can be data mined by the NSA.

Re:No

[Eravnrekaree]

We also dont want to forget that the tablet thing is a fad of people trying to look like they are on star trek. But they are subjecting themselves to painful, difficult to use form factors to look cool because its the hip, in thing. I compare tablets and smartphones to high healed shoes and corsets, and mens leather business shoes, they do terrible things to the body and people subjected themselves to the pain of this to look fashionable, even though there were much more comfortable attire available. In fact they make themselves look like idiots. Il'll take my sneakers and my desktop computer any day over the tablet and business shoes.

Re:No

[The Cat]

and probably will run circles around you on any job.

How about assembly language debugging? Or perhaps animating pixel art? Try that with your voice interface, son.

There are some tools that are the right tools for a job. Learning why things work is just as important as how they work. Someday you'll understand that. Until then, enjoy your toys.

Then game on consoles

[tepples]

There are also fans of console-exclusive franchises (e.g. anything whose characters appear in Smash Bros. series except Sonic), fans of genres that rarely get ported to PC (e.g. fighting games, platformers, and certain JRPGs), and people who prefer one-button convenience in their gaming. These people may do their gaming on a console and everything else on OS X or GNU/Linux.

Gnome Classic works for me

[dltaylor]

I tried XFCE, but it wasn't quite there. Can't really warm to KDE, either. I miss OpenLook and Saw{mill,fish}.

Looks like I'll totally be out of luck when Gnome dies and X is replaced with Wayland. Might as well run Windows at that point.

Re:Gnome Classic works for me

[Eravnrekaree]

X will not be replaced by Wayland. X is going to be moved onto the Wayland architecture and is going to be a user of the Wayland architecture. If you wanted to have a X desktop you will be able to run a fullscreen X server on top of Wayland, and you would never know that Wayland was there.

Re:Gnome Classic works for me

[Eravnrekaree]

PS. you will also be able to run a rootless X server in addition to a rooted or fullscreen X server as well depending on what you need. You can have a full X window desktop with an X window manager, you can have that run within a window on wayland, the rootless X server would run individual X applications within wayland. X isnt going anywhere.

Re:Gnome Classic works for me

[techno-vampire]

I tried XFCE, but it wasn't quite there.

You might want to try again because there's been a lot of development the last year or two. I migrated there before Gnome 3 came out because the whole "one true way" mindset and the Gnome dev's flat refusal to make any effort not to break third-party extensions repelled me. Right now, I'm using Xfce 4.10 on Fedora and I'm quite happy with it.

Re:Gnome Classic works for me

[noldrin]

Didn't enjoy the later or recent Gnome or KDE. Either XFCE or LXDE are both fine when one decides instead of playing/fighting around on the desktop they just want to get work done. If I wasn't such a sucker for tabs instead of Emacs like key bindings I'd just use Ratpoison/StumpWM

Re:Gnome Classic works for me

[dltaylor]

But I have a functioning X server now, and it does what I want it to do. Why should I have to interpose another engine between it and the hardware?

Re:Good! One less abomination of a GUI Toolkit

[kthreadd]

GTK+ is available for C++, it's called Gtkmm. By using C for the core framework adding additional language bindings is often very easy since almost everything can interface with C.

Re:Grow the hell up, retards.

[91degrees]

What about Nova which means in Spanish (IIRC) "Won't Go". Or in Germany,

This isn't really true. Nova means nova. It's an obvious pun perhaps but it's like calling someone notable because they lack dining furniture.

Pschitt Cola.

Not a problem for something not sold in English speaking countries.

Or Windows. What a fucking stupid name. Office? Ridiculous. Access??? WTF?.

Nothing offensive about these.

Or what about "Squirting"?

Yes. That is a stupid name too.

No, the only problem retards have with GIMP is that it's a damn sight more worthwhile than Photoshop.

In what way?

mate seems to be doing just fine

[emilper]

get your helpful gnome back ;)

  http://mate-desktop.org/

You have got to be joking.

[Anonymous Coward]

Linux users are too smart to buy into that cloud shit. If you use Linux, you have a reason to be on a PC, and not a tablet. The author of this article clearly is either
another stupid futurist who wants things to be like Star Trek at the cost of usability, or a troll bought out by the cloud goons who pay all those tech authors to parrot the cloud while their readers fill their comment sections with hatemail. It's an NSA conspiracy, don't touch it.

Re:You have got to be joking.

[tibit]

Pardon to be so in your face, but don't most tablets in the world run Linux these days?

Re:You have got to be joking.

[Eravnrekaree]

I could not agree more. Someone please mod this up.

Re:You have got to be joking.

[Eravnrekaree]

They do run the Linux kernel, but thats it, its not really a Linux OS, there is no similarity with what you would call an real Linux desktop system. Android has its own userland that usually tend to be locked down and designed to control the user, it provides none of the control to the user that a real Linux OS provides. Just because you have a Linux kernel in there, does not mean that the userland running on top of it is not a locked down, inflexible piece of junk thats designed to take away your freedom and get you hooked into the NSA monitored cloud. I would say a real Linux system has the full Linux/Unix userland, Android has its own userland which is designed to create a black box device that takes away freedom by controlling the user.

Re:You have got to be joking.

[tibit]

And how is this GNU userland helpful on a tablet, pray tell?

This article writer obviously lives in a bubble

[dark_glaive]

"ChromeOS has successfully filled the gap between desktop and mobile devices and is becoming the dominant OS. " Really? Any hard numbers on that? What bubble does this guy live in?

Re:Grow the hell up, retards.

[RabidReindeer]

Please, not that Nova urban legend again. Geeze, when will it die?

In portuguese it means "new", btw.

That's because in Latin it means "new" and Portuguese, like Spanish, is a Romance language, descended from Latin.

However, the word "nova" is only a Spanish word in the sense that it's also an English word - both languages have adopted the astronomical term which was originally "stellum nova", meaning "new star". Which in turn was created to describe stars which suddenly appeared or became brighter.

"No va", however, is a perfectly good cynical term to apply to an automobile and I can think of plenty of English word and acronyms that are equally snarky (Fix Or Repair Daily, for example).

Re:Given the # of comments here it's already dead!

[serviscope_minor]

Given the # of comments

Surely the # of comments really just indicates that it is a comment? Unless it's C, of course.

Nova == explosive

[tepples]

A star that became brighter because it exploded? Might as well call it the Pinto. Oh wait, Ford took that one.

There is no reason to use GTK anymore.

[goruka]

For almost a decade, Qt has been the superior choice for developers.

I used GTK for several years (probably up to version 2.2). The mindset back then was that the minimum functionality should be provided and the developer should build what he or she needed around it. For even a simple item list you had to use the treeview, which in turn was really complex to use. I wonder how much of that remains today.
When I discovered Qt, I ran constantly into the situation of thinking "This behavior I want to do sounds like a common case, i'm sure there is a helper/shortcut to implement it", and 99% of the time there was. Maybe it was more "bloated", but it definitely did reduce development time by a large factor.

Also, if you are doing a desktop app, you are most likely wanting to go cross platform. GTK is terrible at that.
The main disadvantage back then was the license, but that's ancient history. Qt has aged well and moved to mobile without much of an inconvenience. Besides Desktop, It runs on Android and Blackberry 10, and will soon be running on iOS too.

Users didn't leave the desktop

[Trogre]

GNOME left the users.

Also, people are abandoning desktops for "services in the cloud"? WTF?

...that you're already running

[tepples]

Any desktop Chrome browser is running what is essentially the user space of the Chrome OS, just with a different kernel.

The bindings are pure shit.

[Anonymous Coward]

Have you ever tried to seriously use these Gtk+ bindings for non-C languages? Have you? I'm pretty sure that you haven't, because if you had then you'd know first-hand how they're utter shit, and you wouldn't have suggested that they're a benefit of Gtk+.

Gtkmm is kaka, and that's putting it very nicely. It doesn't even compete with the monstrosity that is wxWidgets, never mind a professional C++ GUI toolkit like Qt.

The C#, OCaml, PHP, Ruby, R, Guile, Ada and Fortran bindings are horribly outdated, and they were kaka to begin with, too.

The Python, Java and Vala bindings are the least-kaka of them all, but they're still kaka compared to the other toolkits that are available. If I'm using Python, I'll use PyQt. If I'm using Java, I'll use AWT, Swing or SWT. Nobody actually uses Vala for anything, so it's out of the picture right away.

The rest of the bindings are somewhere in the middle. They're still kaka, and I would not use them.

Then there are the JavaScript bindings. Only somebody with kaka for brains would use JavaScript outside of the browser in the first place. There's no reason for these bindings to even exist.

Yeah, the bindings exist to some degree for several different languages, but that doesn't mean they're any good, and it surely doesn't mean that they're actually usable or useful. Given this, it means that they aren't really a benefit of Gtk+.

Re:The bindings are pure shit.

[Darinbob]

Overall, I think Gtk & Gnome were supported for so long was due to licensing. The chief competitor for a long time did not have a GPL license which made it essentially untouchable by GNU purists, or stuck in "non-free" side channels of linux distributions. It wasn't great but it was kept and supported because it was the only kosher toolkit. But then things changed when Qt got an open license.

GNOME 3/shell is fine

[Kagetsuki]

Personally I like it a lot. Like any big change it takes a few weeks to get used to. The GNOME 2 design is just fine and I really liked it but at this point I would not choose to go back.

I think all the people that hate it really haven't given it a chance and probably only tried an earlier release.

Personally I hope GNOME picks up some more steam and more developers accumulate around it. Personally I'd love it if GTK3 were more widely used and supported (EG with Ruby bindings). GTK2 is actually a really nice cross platform GUI toolkit if you know how to properly stylize it.

Re:GNOME 3/shell is fine

[nine932038]

I like Gnome 3 too, but for a different reason. I just find it simple enough. Thing is, everyone's workflow is different, and mine is extremely simplified - I need a web browser, gterms, and multiple desktops, that's it. So Gnome's ultra clean interface appeals to me.

That said, I can certainly understand not liking Gnome 3, but it really seems out of proportion. After thinking about it a bit, I wonder if the hate is not because of the interface per se, but rather that the Gnome developers made a huge, unsanctioned change which affected a massive number of users. That sort of thing breaks a lot of trust.

Re:GNOME 3/shell is fine

[saxa]

Yes, as anything it needs time to evolve into something way more stable and fixed. First releases were not so well done and i also stayed away from gnome 3 a long time, but after getting used to it, it again feels like at home. Changes are that way.

Re:GNOME 3/shell is fine

[Kagetsuki]

I think you've hit the nail on the head. The change was too big for most people and they stopped accepting that as the "real" GNOME which I very much agree with. GNOME 2 with compiz especially had some incredible functionality and looked fantastic whereas the GNOME Shell at the transition point to GNOME 3 was lackng in functionality, unstable, didn't look that great and above all was missing a decent task bar which if you are used to having a highly functional task bar like in GNOME 2 was a pretty big and damning change.

Personally I'm very much a power user and I have a lot of different things going on on different desktops. It sounds like I am at the opposite end of the spectrum than you, but the simplicity is exactly what makes GNOME Shell more powerful for me. I can hit the super key and accurately change tasks, launch programs and do a variety of other things without moving my hands off the keyboard and as soon as I do whatever I need to do the interface gets out of my way. I still like having a preview of what programs are being run on the current workspace so I am using YAWL: https://extensions.gnome.org/extension/674/yawl-yet-another-window-list/ which is very very nice. That's really about it though - the core functionality just works for me, stays out of my way, and lets me work faster/concentrate on my work.

Misdirected problem

[readingaccount]

People are increasingly leaving the desktop computer

No, they aren't.

They are just not buying new ones because they have reached a level where they are good enough for what they do and have no huge motivation for upgrading. That's why the PC market is crumbling, not because people aren't using PCs anymore, but because they are content with what they have on the desktop.

Now I don't get why GNOME is jumping on the Tablet/Smartphone bandwagon when they don't even have to sell anything.

All in all this has turned out to be a fucking disaster, and it's all the GNOME team's fault for being too arrogant to listen to the screams and cries and often well documented problems people had with this new direction. What a fucking mess.

Re:Misdirected problem

[tibit]

They are just not buying new ones because they have reached a level where they are good enough for what they do and have no huge motivation for upgrading.

I've got some anecdotal evidence to back that up. We've got some i7 desktops at work and I don't see any reason to replace them before 2018 or thereabouts, when Windows 8 mainstream support ends. We will upgrade them from Win 7 to Win 8 in December, 2014.

Re:Misdirected problem

[Robert Frazier]

I suspect that you are correct. That certainly describes my situation and my wife's. I built a desktop with an Intel i3 2100 and SSD storage when Sandybridge mini-itx boards became available. I've no desire or motivation to change it. It does what I want. I run a lightweight "roll your own" desktop, using the awesome window manager as the base.

However, I have bought a number of computers since then: An Intel DM2800MT motherboard to build a router/NAS/owncloud/music streamer/etc., 2 x Raspberry PI's for dedicated NTP servers, 2 x Nokia N800 devices (used) for streaming music, 2 x Allwinner A10 based Android tablets for streaming music, and an Asus Fonepad for phone, tablet, ereader and netbook. (In netbook mode, the Fonepad runs full-fat linux and I use a bluetooth keyboard).

In the near future, I can see myself buying more low power (both in energy use and CPU goodness) computers for specialized tasks.much more easily than seeing myself buying another full-fledged desktop.

Best wishes,
Bob

GUADEC?-GNOME Users & Developer's Euro. Confer

[wonkey_monkey]

Okay, shoot me down in flames - apparently if you don't understand everything in a Slashdot summary some people think you shouldn't be allowed on the internet - but what the hell is a GUADEC*? Wouldn't it be a good idea for a news site that presumably wants to attract and keep as many visitors as possible to at least give a brief definition of the terms used in a headline? You can easily do it subtly enough - you don't even have to spell it out, just give enough context - that those in the know won't notice, and those not in the know will come away better informed, instead of having to open up another tab just to find out whether or not they're interested in the content.

Yes, haha, lgmtfy etc. But you know what, I actually do expect to be spoon-fed my news. That's precisely why I watch TV or read newspapers instead of wandering the streets hoping to catch drama unfolding first-hand.

*of course I've already looked it up.

Everybody was expecting to see each other in Brno

Okay, now I know you're making shit up!

(specifying the country of a not-very-famous city wouldn't hurt, either)

Change your attitude

[pieleric]

As many people have already written, it's not the drive to tablet and phone that is reducing the user community, it's the fact that Gnome has become so bad compared to other DEs that people moved away. The main question is "why has Gnome become so bad?". I'd say it's mostly due to not listening enough to user feedback and lack of good judgment on what is good for the users.

Don't get me wrong. I loved Gnome, used it all the time, even used to send patches for the bugs that were annoying me (actually, I even had SVN commit rights at some points). But I stopped because Gnome 3 was worse than Unity _and_ LXDE, and because developers started to close all my bug reports as WONTFIX or, worse, because the patch would not apply anymore... after 2 years of being ignored.

I'd suggest these changes to all the core Gnome developers:
  * first fix bugs before adding a new feature (or a new app)
  * review and merge as many patches as you get from outside people, as soon as possible (that's how you build a developer community)
  * review the entire interface and especially the fixed/default values so that Gnome is _super_ comfortable to use right out of the box
  * do not ever remove features, and never accept regressions
  * make sure your interface can be used by power users too (yes, that means putting back _some_ configuration options), they are the (future) developers
  * listen a bit to user feedback (that one is difficult because it's typically a very noisy channel, but it's necessary)
  * pick a few known and powerful programming languages, and stick to them for all the core applications. Honestly, just drop Vala: as great as it could be, it's not up to a DE project to develop a new programming language, and almost no one outside of the community knows it. If it was up to me, I'd say, just pick C, C++ and Python.

Keep like this for 3 years, and Gnome will be relevant again.

I'd also suggest to pick 2 or 3 apps and focus on them so much that they are the best for the task among any other competitor. This way, people will have incentive to use Gnome, and all the distributions will make sure these apps and all the dependencies are installed by default and working well. For instance, I'd pick: Evince, Rhythmbox, and Aisleriot.

Re:Change your attitude

[Eravnrekaree]

You're being too kind. Gnome is dead and nothing save throwing out Gnome 3 and going back to a taskbar and start menu model will bring it back. Why not add back all configuration options? The problem is that when you say some that just leads to the problem we have now, someone deciding they dont like an option and forcing their ideology on others. Why not just let users decide? give them them options. They can seperate the options into advanced and regular screens, this allows the advanced options to be provided without any sort of a claim that can be made that they confuse people. The best thing that can be done is to get people like Havoc Pennington away from software projects. Basically the Gnome 3 people are a bunch of jerks. Let the project die and teach them a lesson. There are many much more sane UIs available like MATE.

Re:Change your attitude

[CronoCloud]

And kill Adwaita, which infects some GTK applications under other DE's. Even if you use your own theme under XFCE there are still hints of Ugly Adwaita here and there.

ADWAAAAIIIIIITTTTAAAAAA!

Re:GUADEC?-GNOME Users & Developer's Euro. Con

[wonkey_monkey]

Linking one of the many uses of "GUADEC" to guadec.org wouldn't have been a bad start, instead of to a blog which also doesn't tell you what GUADEC is.

Developers, developers

[Torp]

Been a KDE user for a loong while, but that's not what I'm posting about.
As a programmer, i'd very much rather write an application for Qt than for GTK. Qt (at least until version 4) was much easier to work with, not to mention much more cross platform than the GTK ports.
I'm probably not the only one who feels that way.
But they're probably a bit right. I contract, 80% of my work is Android software, the rest is command line Linux. No one wants me to build a desktop app for them, although they see the desktop programming experience from a few years ago.

Bullshit trends

[vovick]

According to Google Trends, interest in GNOME and GTK+ is soon to be extinct.

For crying out loud, Google Trends compares search terms without any context. All the comparisons I've seen in this discussion make as much sense as mine. Time to write a new story about Gnome losing to goblins and dwarves?

Google Trends vs Debian "popcon"

[Mathieu Lu]

More reliable than Google-Trends: Debian "popcon", a program that Debian users are offered to install and report their program usage.

http://qa.debian.org/popcon.php?package=gnome-shell (40k installed / 20k votes)
http://qa.debian.org/popcon.php?package=xfwm4 (16k installed, 8k votes)
http://qa.debian.org/popcon-graph.php?packages=kde-window-manager (14k installed, 8k votes)
http://qa.debian.org/popcon-graph.php?packages=xserver-xorg (80k installed, 25k votes)

The stats being what they are, you can't really compare the 40k gnome-shell installs with the 16k xfwm4 (gnome-shell is installed by default, which makes the 16k xfwm more impressive, I guess), but you can make some conclusions.

And yeah, I like gnome-shell / Gnome3. Sometimes after a crash (I run debian-experimental packages), I return to fvwm for a few hours, but I always end back onto gnome-shell. "it works", is pleasant to use, and if necessary, there are ways to customize it.

A few months ago, I had forked and published an extension for hiding the top panel. I was surprised of all the feedback and number of users it got. Better yet, someone else stepped up to maintain it and does a great job.

Seriously.. way too many trolls, and most of the rest don't bother to comment.

Re:Google Trends vs Debian "popcon"

[pscottdv]

I really like the gnome 3 shell, but had to abandon it because of stability problems. I've seen the "Oh Noes" screen more times than I can count.

Re:Google Trends vs Debian "popcon"

[iggymanz]

Were Debian a desktop distribution, those stats might have some meaning. It is a wonderful server distro, but only a Poindexter among Nerds would use it as a desktop. Meanwhile, derived distros with much higher membership don't even include the trainwreck in motion that is GNOME3

Re:Can't say I'm surprised

[Anonymous Coward]

Using Windows is like spitting on the graves of all the men and women who died for your constitutional rights

Hyperbole much? Seriously, you need to chill and get some perspective.

I still use GNOME.

[loufoque]

I personally still use GNOME. I've always preferred GTK+ over Qt, not only because I prefer its look, I also prefer the fact that it is focused on being a GUI toolkit, while Qt is a kitchen sink with a horrible C++ design and full of stuff better done elsewhere.
GNOME 3 is shit, but you can still run gnome-panel just fine (and metacity if you want it), even with Ubuntu, and that's what I do.

The more serious problem is that with the obsolescence of GTK+, we may end up not having a good standard GUI toolkit to write applications on Linux anymore.
There is much focus on graphics lately with the alternatives to X.org being developed, but Linux still doesn't have a good solution to make graphical apps with resolution independence, proper text rendering, fluid layout and good accessibility. Not that other operating systems are being that much better at any of this.

Re:I still use GNOME.

[master_p]

I always thought Qt had one of the best C++ designs. Which parts of it do you find horribly designed?

Re:I still use GNOME.

[loufoque]

That's a vast subject.
Mostly it overuses OOP instead of making use of the multi-paradigm nature of C++, stays away from language features that weren't well supported in 1997, and duplicates a lot of things that are done better elsewhere.

Re:I still use GNOME.

[Eravnrekaree]

To actually have cross platform support you must provide cross platform APIs for everything you can possibly need to do is a must, that includes audio, video, networking, video capture, recording, image acquisition devices, you name it. QT has to be much more than a GUI toolkit or you would be very limited with the kind of apps you can write with it in a cross platform way. Thats why the kitchen sink nature of it is a benefit not a drawback, you seem to base your condemnation on really nothing at all. QT need to be is designed that you can write an applications that can do anything and that it can run on anything, it can allow access to any kind of hardware or system interface in a cross platform way without an app having to worry about OS particularities. Instead of having app developers waste time reinventing the wheel creating another cross platform compatability layer for every app they write, it makes sense to create a common set of libraries that provide an OS independant API for accessing every concievable hardware and software feature on all platforms. QT should include the kitchen sink, so that apps can be cross platform. A video conferencing program, a recording studio program, a dekstop publishing program, a video editing program, a 3D opengl game, written for QT shouldbe able to run on Linux or Windows without a single line of code needing to be changed.

Re:I still use GNOME.

[loufoque]

Just use different libraries for audio, video, networking, video capture, recording, image acquisition, you name it.
Do one thing, an do one thing well. Do too many things at once, and none of them will be good.

Re:I still use GNOME.

[Eravnrekaree]

Whether or not the libraries for different capabilities are bundled under the same project heading or not is irrelevant as far as whether or not they are good. It does actually make it a lot easier to have a bundle of libraries that will work on all OSs, so you know that the entire environment will operate on every OS supported by the environment, rather than having to hunt down seperate libraries, some which will work OS X and not OS Y, others on OS Y but not OS X. I cant imagine why having to try to find 10 seperate libraries from different sources each with greatly varying degrees of cross platform support wiould be better than one bundle of libraries that works on the same list of platforms. It makes things much more complex. Why do you want to make writing a cross platform app more difficult if not impossible?

Re:I still use GNOME.

[loufoque]

A library dedicated to a single thing will always do it better. Qt is certainly not the best library at networking or video.
Any serious networking or video application that also uses Qt will *not* use Qt for networking or video.

Re:I still use GNOME.

[ickleberry]

The more serious problem is that with the obsolescence of GTK+, we may end up not having a good standard GUI toolkit to write applications on Linux anymore. There is much focus on graphics lately with the alternatives to X.org being developed, but Linux still doesn't have a good solution to make graphical apps with resolution independence, proper text rendering, fluid layout and good accessibility. Not that other operating systems are being that much better at any of this.

This is already the case in many ways, hence the popularity of web based apps and native applications that run their own webserver to be accessed via localhost. I actually found GTK+ easier to use than Qt, but there is no windows version for GTK3, not even one in sight so it turned out that I was writing a Linux-only program without realising it! GTK3 itself is not too bad to use, it's just a bit limited and needs work.

The Gnome desktop though, is a whole other story, it seems to be controlled by developers who lost the run of themselves and tried the Steve Jobs approach - "We're giving you a UI change and you're going to like it!"

Re:I still use GNOME.

[master_p]

How does it overuse OOP? its design seems very logical to me.

From which features of C++ does it stay away? Qt 5 even supports lambdas as slots.

Which things are done better elsewhere? could you be more specific?

Re:I still use GNOME.

[loufoque]

This is a well-known subject. There is no need for me to detail everything here. Just search the Internet.

Re:I still use GNOME.

[master_p]

I think you are trolling. I searched for Qt's horrible design problems but I found nothing.

Re:Can't say I'm surprised

[cyber-vandal]

And to learn the difference between loose and lose.

Re:Given the # of comments here it's already dead!

[dbIII]

Funny how photoshop now has the multi-window interface that all the photoshop warez fans used to say was the biggest problem with gimp. Also gimp has the single window interface as a choice too.

Comment removed

[account_deleted]

Comment removed based on user account deletion

very misleading

[Nightshade]

Take a closer look at the google trends data. If you click on the "qt" tab you actually see that most of the searches are related to "qt syndrome" or "long qt". these are medical conditions and have nothing to do with UI toolkits. if you click on the "gtk" or "gnome" tab, the search terms are all related to UI toolkits.

Perhaps it's not something specific to gtk/gnome, but maybe all the toolkits including qt are in decline. Either due to smartphones/mobile or ubunut's unity or something else.

Re:very misleading

[93 Escort Wagon]

Yah, this submission smells more like the work of a Qt zealot who was looking for a way to give like-minded individuals an opportunity to trash-talk Gnome. Maybe we'll see a "will the last emacs user please turn out the lights" submission from a vi fan?

I love the level of discussion here too - low even by Slashdot standards...

Post: "I am not crazy about gtk+, but qt has its problems too."
Response: "You are a poopy-head!"

Post: "What about gimp? People use gimp."
Response: "Nobody uses gimp! And if they do, they're poopy-heads!"

Re:the last GUADEC?

[kthreadd]

I would prefer to have one conference for both projects, and hopefully bring them together again.

Re:Can't say I'm surprised

[HiThere]

I have severe problems with the licenses used by both MS and Apple. In particular the provision that has been in several versions of their licenses that allow them to "add, copy, modify, change, or delete any file on your computer". That proviso was originally instituted by MS, but Apple tried to get me to accept it by sneaking it into the license on a security upgrade. So now I have two computers that I can never attach to the internet, but due to data locked in non-portable formats I can't every get rid of. (Well, ever is a long time. I'm probably getting near the end of it. Which is a very good thing since one of them hasn't be upgraded in nearly a decade.)

Fighting stupid limitations

[hendrikboom]

Some time ago I used GTK to do a scrollable tree display, and discovered a stupid limit of about 32K pixels high (I might misremember, it might be something like 64K). If I went past that, further tree expansions would show up at the top of the scroll region overwriting what was already there, instead of at the bottom where they belonged.

Not a limit of 32K pixels isn't too bad for what's actually on a screen, but if it's a limit on what you can scroll through, well, it's ridiculous. Does anyone know whether this kind of limitation is still there?

Or whether Qt does similar?

There's no Win32 port yet either

[Nimey]

It's been... almost 3 years? and there's no stable, supported port of GTK+3 to Win32 yet. It's like they don't care about cross-platform programs written against their toolkit anymore.

Big fish in a small pond.

[Animats]

From the article:

"Only a few years ago, GNOME was at the centre of the creative world. Remember Maemo and the N700?"

No.

What's a GUADEC?

[Jane Q. Public]

GUAno DECoder?

It seems to be some kind of GNOME conference?

But QT is still butt-ugly

[xdancergirlx]

There are problems with Gnome and GTK+, but what surprises me is that KDE and QT have managed to remain consistently, in the words of RMS, butt-ugly throughout the years. It's like a requirement for each new release to be at least as ugly as the last one. Even with plasma and fancy 3D effects, still that same ugly-mugly, kinda-boxy, badly spaced, wierdly rendered fonts, dumb controls, QT look. My love for gnome has dissipated markedly in the last few years with the gnome 3 disaster, but it is hard to imaging replacing it with, ugh, butt-ugly kde.

Re:But QT is still butt-ugly

[MrEricSir]

Qt is perfectly themeable, but if you want a more modern look you should really be using QML at this point.

Re:Grow the hell up, retards.

[Grishnakh]

That's because in Latin it means "new" and Portuguese, like Spanish, is a Romance language, descended from Latin.

I think it's also important to note that Portuguese is much closer to Latin than Spanish is, so it's not surprising that "nova" (new) is common to both Latin and Portuguese and not Spanish. I wouldn't be surprised if "nova" is also used in Galician.

Re-buy

[tepples]

[we] always seem to wind up on the play store wondering why google is selling us games that only run on my damn phone, while we're on the chrome

Probably for the same reason Nintendo is selling Smash Bros. 4 for Wii U and Smash Bros. 4 for Nintendo 3DS as separate SKUs: so that you'll buy both.

And I wish RH would get a clue...

[whitroth]

A year or so ago, I had to install FC17, I think it was, on a user's machine, and he wanted gnome 3.

That was the worst piece of crap I have *ever* seen as a gui.

Huge icons, basically filling the screen, that go transparent/invisible until you roll over them.... Who's this for, some 16 year olds who think it's K3WL!!! My stepson had that, I think, on ubuntu on his laptop, and things ending with a visual explosion? or starting or ending as though you were tearing a sheet of toilet paper?

*rolls eyes*

Yep, as the thing I read last year, I think, said, the typical laptop has more computing power than all the computers in the world in 1970... and 99% of its computing power is used for friggin' eye candy. Make it all run with the *blazing* speed of a '286....

                    mark

RedHat and GNOME

[hduff]

RedHat adopted the GNOME desktop for Fedora which was intended to be the framework for the desktop companion to their RedHat Enterprise server. In this corporate-centric model, there was no need for fancy eye-candy or configurations options because the BOFH determined what you used, how it worked and what it looked like. And it needed to be minimalist, locked down and bereft of choices and meet corporate needs, not user needs.

GTK exists only because people went all derp over QTs inital license terms. Even when the QT license terms were changed to be acceptable to the FOSS community, the Gnomistas continued with their "I hate the evil QT and I'm very cool" attitude. It's difficult to sustain a movement on hate and arrogance and negativity.

GTK and GNOME were, indeed, made for each other.

Is it any wonder that the GTK/GNOME desktop is dying? It does not meet the needs of the majority of the people who want to use it. It is intended only for a very narrow, small audience.

All this yammering of GObject..

[Anonymous Coward]

You can complain on and on about GObject being horrible, but the fact is that GTK+ including GObjects is the best way to write GUI applications in plain C. Sure there would be Athena widgets and Enlightenment.

If Qt is so great, how about C bindings?

Re:All this yammering of GObject..

[kthreadd]

Don't forget Motif, which just became free software under LGPL.

Re:Grow the hell up, retards.

[RabidReindeer]

That's because in Latin it means "new" and Portuguese, like Spanish, is a Romance language, descended from Latin.

I think it's also important to note that Portuguese is much closer to Latin than Spanish is, so it's not surprising that "nova" (new) is common to both Latin and Portuguese and not Spanish. I wouldn't be surprised if "nova" is also used in Galician.

Or something similar. After all, Spanish kept "nova" for new, but mutated it into "nuevo", leaving the original Latin form for astronomers.

Mostly agree,

[aussersterne]

if the KDE4-followed-by-GNOME3 debacle had never happened, I'd still be using Linux. Instead, I went to Mac OS, which is where all of the other Linux users I used to know went as well—a group that had steadily been growing for a decade prior to the last 2-3 years.

Now it's too late to close the stable door; the horse is gone.

o became uo, ue

[tepples]

Italian and especially Sardinian are closer to Latin in some ways than even Portuguese. But in both Italian and Spanish, a Latin 'o' in some contexts became a diphthong: "uo" in Italian or "ue" in Spanish.