Google To Replace GTK+ With Its Own Aura In Chrome

sfcrazy writes "Google's Chromium team is working on an alternative of Gtk+ for the browser, called Aura. Elliot Glaysher, a Google developer explains, 'We aim to launch the Aura graphics stack on Linux in M35. Aura is a cross-platform graphics system, and the Aura frontend will replace the current GTK+ frontend.' The Free Software community is debating: is Google trying to do Canonical? Couldn't Google just switch to Qt, which is becoming an industry standard?"

Comments

Google's Aura

[rossdee]

Is looking darker and darker every year

Re:Google's Aura

[Dcnjoe60]

Is looking darker and darker every year

Actually, Aura has been part of the Chromium project for quite some time, so it isn't any darker today, than it was yesterday, or even last year or two. Most likely, this has more to do with ChromiumOS than Chromium/Chrome.

Here's the link: http://www.chromium.org/develo...

Re:Google's Aura

[higuita]

QT with LGPL could be used freely by google... maybe the problem is control... they could not control GTK and may have fear that QT could neither be controlled by then... Or is just another NIH attack!

Re:Google's Aura

[Anonymous Coward]

QT with LGPL could be used freely by google...

People on Slashdot keep saying this, but corporate lawyers see "GPL" in "LGPL" and flat out say "no way."

Re:Google's Aura

[RaceProUK]

QT with LGPL could be used freely by google...

People on Slashdot keep saying this, but corporate lawyers see "GPL" in "LGPL" and flat out say "no way."

Maybe those corporations should hire actual lawyers instead of corporate lawyers.

Re:Google's Aura

[MrNaz]

God forbid someone make an on-topic reply to a joke. Noooo we can't have that, anything on topic after a joke *must* be whooshed.

Re:Google's Aura

[BravoZuluM]

The incompetent lawyers hire the additional incompetent lawyers. It was hell just getting our lawyers to understand that ftping source over the network through open source routers and servers did not make the transmitted code open source. The name LGPL needs to be changed so that the letters "GPL" does not appear in the acronym. Our lawyers screech, "GPL! GPL!" "Uh no, it's LGPL" "But it's GPL!" Our lawyers are fine with the BSD license so maybe we can call it the LBSD license. :-)

Our lawyers, quashing innovation one bad decision at a time.

Re:Google's Aura

[Anonymous Coward]

What was that Shakespeare quote... the one about the first thing we should do...

Re:Google's Aura

[shutdown -p now]

I'm pretty sure that Google's corporate lawyers are quite familiar with GPL and LGPL. They do, after all, ship Linux in commercial products.

As a Qt fan

[Anonymous Coward]

Atleast Gtk+ isn't gaining.

Re:As a Qt fan

[Anonymous Coward]

I agree. GTK is disgusting crap that tries to cram every bad design decision of every platform it works on into one horrible framework to make the absolute worst GUI toolkit ever conceived. I strongly suspect that it was a product of the line of thought that lead the GCC internals to be so convoluted: some bullshit paranoia that proprietary vendors might make something nice out of it. The GCC are feeling the effects of that bad decision as LLVM\Clang eats their lunch and are just now beginning to improve their design. I sincerely hope that the GTK devs will react the same way and try to salvage the heap of crap they manage.

Re:As a Qt fan

[kthreadd]

Lots of adjectives. Explain what you yhink is wrong with it instead of just saying that you dont like it.

Re:As a Qt fan

[Anonymous Coward]

You're obviously trolling, but I will bite. What makes you think Aura is better than QT? Why throw in the yardstick of "serious" work, when the difference doesn't affect workflow?

Re:As a Qt fan

[Artifakt]

I'll let the AC explain what he thinks is wrong, if he will actually step up to the plate.
But, you do realize that this story starts with Timothy mentioning what a small percentage of the OS community thinks, and doesn't mention a somewhat more likely possibility - that Google is dissatisfied with the GTK, finds it very difficult to work within its limits, and doesn't feel it can get any cooperation from the GTK designers. If that is how Google feels, then the AC would probably say Google's position is reasonable. I tend to agree with that, myself. But, what's the point of asking the AC to defend his position, when that same position was totally left out of Timothy's original summary, and the position of those who don't see any problems with the GTK is presented as the default of the whole open source community?

Summary: Ooooohhhh! Anybody who doesn't like FOO is a rapist of dead baby seals and unmutual to boot! We're gonna just assume that absolutely everybody reasonable likes FOO, and raise only the questions those reasonable people would ask mean old unreasonable Google.
AC: Well I don't like FOO because it's smelly and might let girls into the Sekret club...
You: AC, you need to explain mo' betterer

Yes, AC probably should present some specific facts, if this was a debate over GTK's quality. But even if you turn this whole thread into a debate with the AC and others like him, win every point, and leave the rest of us impressed with your clarity and logical superiority, do you really think that will prove Google's reasons are as invalid as your debate opponent's?. The facts are, there is an ongoing debate o in the OS community over the conduct of the GTK developers. The summary needs to be written like the community is still seriously divided, not like the only questions being asked are from people who don't see a problem with the GTK and assume that Googgle can't really have a good reason.

Re:As a Qt fan

[Anonymous Coward]

I am not sure I agree with you. I'm a Qt fan because it allows you to have native looking applications with little effort.
I don't use Chrome, but I used to use Chromium and usually a very recent development release. The removal of native-looking widgets in web pages pretty much coincided with the introduction of Aura in the development version. I wouldn't be surprised if Aura isn't capable of drawing native widgets and Chrome never gets them back.
I've since switched to Firefox for obvious reasons, and I my only regret is not switching sooner.

Re:As a Qt fan

[Anonymous Coward]

GTK is the lousy toolkit here. Don't know about Aura, but given the fact Chromium used GTK in the first place I say their judgement is impaired and that's the reason they are going Aura and not QT.

Re:As a Qt fan

[Anonymous Coward]

sorry , but who/what is AC ????

Re:As a Qt fan

[benjfowler]

You haven't tried coding against Motif, have you? That manages to be worse.

Re: As a Qt fan

[Anonymous Coward]

Anonymous Coward

Re:As a Qt fan

[Dcnjoe60]

But, you do realize that this story starts with Timothy mentioning what a small percentage of the OS community thinks, and doesn't mention a somewhat more likely possibility - that Google is dissatisfied with the GTK, finds it very difficult to work within its limits, and doesn't feel it can get any cooperation from the GTK designers. If that is how Google feels, then the AC would probably say Google's position is reasonable. I tend to agree with that, myself. But, what's the point of asking the AC to defend his position, when that same position was totally left out of Timothy's original summary, and the position of those who don't see any problems with the GTK is presented as the default of the whole open source community?

That is pure speculation. Google hasn't stated why they are switching, just that they are switching. A just as valid reason could be that since Aura's purpose is to produce a new desktop window manager and shell environment with modern capabilities. The UI must offer rich visuals, large-scale animated transitions and effects that can be produced only with the assistance of hardware acceleration., that it simply fits in better with their vision for Chromebooks or something they want to do with android. If your vision is that your browser is the desktop, then it makes sense to use Aura which is its own windowmanager that includes acceleration and compositing.

Aura has been part of the Chromium project for quite some time. Initially, it was Windows specific. It looks like it is now working on Linux (and I assume OS X), too.

Re:As a Qt fan

[Anonymous Coward]

I agree. GTK is disgusting crap that tries to cram every bad design decision of every platform it works on into one horrible framework to make the absolute worst GUI toolkit ever conceived

Qt is a p.o.s. too, with numerous lousy design decisions. Some people in the community still can't forgive the developers the idiocy surrounding the original licensing either. Hence the decision to choose something different from either.

Re: As a Qt fan

[Anonymous Coward]

Anal cunt (it's a band, for real)

Re:As a Qt fan

[Anonymous Coward]

That is pure speculation. Google hasn't stated why they are switching, just that they are switching.

Speculation that would be well enough founded, considering that I have never heard anything good about GTK. Hell, even Linus the most important figure in Linux development could not get useful support from the community, as a result one of the projects he started had to migrate to Qt. Having such high profile cases go wrong and even developers throwing around excuses "It can be done, you should look how yourself instead of wasting time", "you do not understand our design paradigm" and such gems as "windows support was never intended to be more than an experiment" makes any migration away from it look like the sane thing to do.

Fact is unless you are a GNOME developer there is zero reason to use GTK and every reason not to.

Re:As a Qt fan

[Anonymous Coward]

To be honest, after taking a look at the Subsurface source code before they migrated to Qt I can understand why they did it. It usually helps if you understand GObject before you design your application. A typical C application designed from the point of view that the GUI is just another library that we can call here and there. Compare it with the typical GNOME application and you see what I mean.

Re:As a Qt fan

[kthreadd]

I wouldn't say native-looking, but rather native-inspired. I'm using OS X occationally and it's very clear when an application is built with Qt. Things usually look almost native, but there's always some things here and there that's not right.

If you want native, build a native UI per platform and share common code between them.

Re:As a Qt fan

[Anonymous Coward]

Look man, GTK (That's GIMP Tool Kit) was written in the 90s as a replacement for Motif. (If Motif doesn't mean anything to you, that should be your first clue to fuck off. If it does mean something to you, I'll have to assume the brain damage was permanent and severe.).

Re:As a Qt fan

[MrNaz]

If you want native, wxWidgets.
Cross platform. Open source. Sane licensing. True native drawn widgets.

Re:As a Qt fan

[Aighearach]

Gtk is actually the best example of OOP in pure C I've ever seen. It still works great, lots of apps use it. Mostly, apps that don't bother advertising their toolkit because it is besides the point.

These fanboys should lay off the drugs. Toolkits are not the current fad, you get no style points for loving Qt in 2014. People even hate on the code, guaranteed if they had to find an actual example of bad code it would be their first time opening it up, and they'd eventually just point to some app that uses it.

I've read the Linus flames, I'm not convinced at all. Clearly, somebody pissed him off when he was drunk and he Made a Decision. If you don't believe me... google it.

Re:As a Qt fan

[Aighearach]

wxWindows was a nice promise. I remember trying it out over a decade ago. And then again a few years later. The problem is, the code isn't written in a portable style; instead they try to bugfix their way to portability. So it often has trouble compiling. You can't just distribute a wxWindows app and have it work everywhere, for real, the way Gtk and Qt apps do. It will crash and burn for some percent of users; it will crash and burn if the compiler settings need to be adjusted; it will crash and burn just because a dev wants to use a new version that hasn't matured enough to get enough bug reports.

Re:As a Qt fan

[Anonymous Coward]

OOP in C, how is that supposed to be? I thought you didn't even have classes in C.

Re:As a Qt fan

[paraax]

struct MyObject { ... };

struct MyObject* MyObject_create() { /* constructor code */ }
void MyObject_method_1( struct MyObject* obj, args... ) { }
void MyObject_method_2( struct MyObject* obj, args... ) { }

True you don't get encapsulation, but its still object oriented.

Re:As a Qt fan

[Anonymous Coward]

To be truly OO, it has to have polymorphism.

Re:As a Qt fan

[shutdown -p now]

Which it does.

Re:As a Qt fan

[Anonymous Coward]

ANSI C, or are we talking GCC extension?

Re:As a Qt fan

[shutdown -p now]

It's not language-level, just as there are no language-level functions methods. But you can always make an explicit vtable in C, and write some macros to hide the virtual dispatch.

Just for a browser?

[Anonymous Coward]

From the story:

I don’t think it makes much sense to write your own GUI toolkit just for a browser.

It occurs to me there are very, very few applications that could possibly warrant the development of a new widget set, but that a web browser is certainly among them.

Gtk emerged from Gimp; the "GIMP Tookit." If a raster graphics editor used by a vanishingly small number of people provides sufficient rationale to make a toolkit, certainly a web browser used by tens of millions is worthy.

Re:Just for a browser?

[Anonymous Coward]

What exactly is the need for a new toolkit? more widgets? you can take an existing toolkit and make something look horrible like unity . Or a simple toolkit can make something look cool like blackbox .

Re:Just for a browser?

[IWannaBeAnAC]

The reason why Gtk was written is because at the time that GIMP was starting out, there wasn't a good alternative. Nowdays there are.

Re:Just for a browser?

[peppepz]

Because GTK2 is bit-rotting and GTK3 is bound to GNOME.

Re:Just for a browser?

[kthreadd]

If I remember correct the Gimp guys had started with Motif. Early versions of Gtk+ was more or less a non-strict reimplementation of Motif, but that changed quite rapidly after a while.

Re: Just for a browser?

[Anonymous Coward]

Right, and Motif wasn't free.

Re:Just for a browser?

[ChunderDownunder]

there are very, very few applications that could possibly warrant the development of a new widget set, but that a web browser is certainly among them.

There is only XUL.

Re:Just for a browser?

[gbjbaanb]

IIRC Gtk was written because there was a high-quality alternative but that wasn't free and open.

Years later Trolltech released Qt under the LGPL and suddenly there is no longer any need for Gtk (or other toolkits written it seems just to be a bit "special")

Re:Just for a browser?

[kthreadd]

You don't need to run Gnome to run GTK 3. I'm using it right now on just fvwm.

Re: Just for a browser?

[Anonymous Coward]

It is now (LGPL). Time to move back?

Re:Just for a browser?

[Zukix]

From previous releases its clearly the Chrome team is being mismanaged and has lost its way.

They really cannot get the basics right. A web browser is basically text in windows that can be styled by the page author. Lets see you they are doing:

i) They don't fix the appalling font rendering issues on Windows promptly and as a priority. Most of Google's own web fonts are unusable in production because of this.

ii) They don't follow standard most-recently-used order when ctrl-tab between tabs and they don't see the problem and close any bug report as won't fix. How can Chrome be the platform for office tools and applications when you can't flick between applications?

iii) They start adding animations to html elements you can't restyle with CSS e.g. the zoom ease-in they added to select elements in a recent Chrome. What possible justification was there for that? If you need to use more than a couple select elements on a form the animation effect of using each one is terrible.

iv) They add forced behind the scenes updates (ok) but they then push poorly tested unstable releases. There were wide-spread issues on their recent releases. You can only auto-update if you are rock-solid.

v) They fork from the web-kit project, a once high-point in cross company collaboration for the betterment of the web. Now... beginning of the end.

vi) And now they are going to spend their time re-implementing a cross-platform widget toolkit. How about fixing the fucking font rendering first?

I don't know how the team is being led but it can't be right. Google, time to take an axe to your chrome team...

Re:Just for a browser?

[Anonymous Coward]

Hopefully the new toolkit uses standard C++ instead some weird kitchen sink object system and their own data types scattered everywhere. Perhaps the STL is not perfect, but it is quite likely already used all over the application itself, and not too many want to write their otherwise perfectly portable using some Q* classes.

Re:Just for a browser?

[Anonymous Coward]

GTK3 is bound to GNOME.

Errr... it supports Windows, for a start...

Re:Just for a browser?

[Anonymous Coward]

GTK3 is bound to GNOME.

Errr... it supports Windows, for a start...

your post is so funny I almost fall from the chair laughing.

Re:Just for a browser?

[hobarrera]

I agree on this point, but, what about Qt? Qt is superior to gtk in almost every way, cross platform, and not "bound" to anything. Plus, the license is pretty liberal.

Re:Just for a browser?

[K. S. Kyosuke]

Except that this "standard C++" is total crap that doesn' even offer proper reflection, which is the whole reason why Gtk+ went on to create this "weird kitchen sink object system" (so that creating the bindings to all the dozens of different languages used in a typical Un*x system were as simple as possible). And why - wonder of wonders - Qt did the same, to improve this "object shitstem" of C++. I think a discerning observer might start seeing a pattern at this point...

Re:Just for a browser?

[Anonymous Coward]

You need lots from GNOME project to get GTK3

Qt would be much better choice but Google engineers just dislikes Qt but they dislike even more GNOME, so they go to third option.

Re:Just for a browser?

[fuzzyfuzzyfungus]

I'm not sure why GIMP originally wanted a Toolkit; but Gtk mostly emerged because Qt was proprietary at the time. Given that, for all its failings as a toolkit, Gtk, possibly along with other factors that coincided with it(I'm perfectly willing to listen; but don't know of any offhand) succeeded in getting GPLed Qt. I think Nokia even LGPLed Qt part of some aspect of their flailing-death-spiral strategy.

That, to the best of my understanding, is what confuses people: Qt is generally considered superior to Gtk, is now LGPL, and is generally well liked; so why would somebody say 'Well, Gtk has issues, so I'm going to make my own Gtk; but better!' when they could just use Qt?

Re:Just for a browser?

[kthreadd]

You need lots from GNOME project to get GTK3

Can you be more specific? GTK+ does not have a lot of dependencies to begin with, and most of them is not directly related to the Gnome project; GLib, GdkPixbuf, Pango, ATK and GObject according to the documentation [0]. The rest of them are external.

[0] https://developer.gnome.org/gt...

Re:Just for a browser?

[serviscope_minor]

I agree with most of your points, and I don't use Chrome, or chromium, but I disagree here:

v) They fork from the web-kit project, a once high-point in cross company collaboration for the betterment of the web. Now... beginning of the end.

I think this is actually a good thing sort of. The cross compatibility should come from good, implementable open standards, not a single open source renderer. It was beginning to get to the stage where Webkit was getting so dominant that it was looking to a return to the bad old days of "Best viewed with Nutscrape at 640x480, 8 bit colour". It's now back up to 4 mayor engines: Webkit, Blink, Geko and Trident with a reasonable split between them. For now, the first two are nigh on identical, but after a while they should diverge a good bit.

Re: Just for a browser?

[Anonymous Coward]

Except that this "standard C++" is total crap that doesn' even offer proper reflection, which is the whole reason why Gtk+ went on to create this "weird kitchen sink object system"

Wrong. GTK+ did their weird object system because it's in C, not C++. Reflection isn't a factor, and as a C++ programmer I don't feel like I'm missing anything without it.

And why - wonder of wonders - Qt did the same, to improve this "object shitstem" of C++. I think a discerning observer might start seeing a pattern at this point...

Qt started down that path because the C++ compilers of that era had poor or non existent compatibility with C++ template specs. They decided that it was just easier to write a precompiler stage than deal with templates.

Re: Just for a browser?

[Entrope]

Qt has a lot of overhead that can be useful for writing desktop apps but requires extra work for a web browser. Qt wants all apps to be web apps, except you get your "choice" whether to write layout and logic in Qt Quick, C++ or overhead-added HTML; this gives you some degree of interop with the other two, but web browsers don't need that or the overhead it brings. Qt also pointlessly reinvented lots of the C++ standard -- witness QString and all their container classes -- making it hard to integrate with libraries written in non-Qt C++. People who use Qt are mostly allowing themselves to be locked in to a dead vendor's proprietary library.

Re:Just for a browser?

[martin-boundary]

This is exactly how Microsoft fought the browser wars in the 1990's. By introducing deliberately incompatible, piecemeal and nonstandard ways of doing things, and making them defaults on Windows. Google does the same on Android and ChromeOS. Microsoft had Windows on every PC, except Apples. Google has Android on every smartphone, except Apples. You might ask, why should the Free Software Community care? Well, it used to be that Android was basically a linux distro. Linux distros are nice, because the user can replace anyting he likes and make everything just so. That works because the free software isn't just gratis, but has the kind of license that promotes collaboration, and a community of devs and users. Google have horrible licences, and want to replace existing software that runs unchanged on lots of OSes and architectures with their own, then bundle it with Android. In this way, you end up with an OS that runs proprietary software nobody except Google can change (realistically), and runs on every piece of hardware you can buy in stores (except Apples). It spies on you, and rams ads down your throat. And the ubiquitous Free Software we became used to in the internet boom years will become less available, since it won't run on the hardware people will want to buy.

Re:Just for a browser?

[renoX]

I agree with you but not for the same reason: my reason is a variant on 'cannot get the basics right': Chrome is not pleasant to use at all on a 4GB RAM PC, it use too much memory which make the PC swap..

I used to prefer Chrome to Firefox, due to its snapiness, good separation between tab, but when it used frequently so much memory as to make the PC swap, I switched back to Firefox.

Re: Just for a browser?

[Anonymous Coward]

Can we please stop spreading this myth? Even with C++11 and upcoming C++14 you can't just replace the Qt object metatype system. See e.g. http://woboq.com/blog/reflection-in-cpp-and-qt-moc.html for a detailed analysis ...

Re:Just for a browser?

[BitZtream]

The could have just used SWIG like the rest of the planet.

Re: Just for a browser?

[K. S. Kyosuke]

Wrong. GTK+ did their weird object system because it's in C, not C++.

Yes, they did. And they also had a number of other reasons. Most of which perhaps disappeared in time, but not all of them.

Reflection isn't a factor

Sure, if you're happy to be forced to recompile all binding libraries, generate new packages and force hundreds of thousands of people to upgrade them whenever the interfaces change instead of just upgrading Gtk+.

and as a C++ programmer I don't feel like I'm missing anything without it.

Oh wait, you're C++ programmer. Of course, in that case your brain is already so horribly damaged that you can't perceive the comparatively mild benefits this would bring to a C++ programmer (removing only one cut from the thousand ones already killing you).

Re:Just for a browser?

[Greyfox]

The C++ mentality is that you should catch as many errors as you can at compile time. When you're launching a satellite, you want to be sure your code's as bulletproof as possible before it leaves the ground. Every time I've seen reflection used, it was by some terrible programmer who'd just learned about reflection and was looking for an excuse to use it. And every single time, they were using it as a crutch so they wouldn't have to think about the system they were building. You can almost hear them thinking "Cool! I don't really know what this bit here is supposed to do, so I'll just use reflection and once I've collected my paycheck and left, some other programmer can drop random objects in there later!" I've never seen its use actually improve a design. I've never even seen it make more loosely-coupled or reusable code, now that I think about it. Somehow they ended up with introspective systems that were so tightly coupled that you couldn't just break one object out to run a unit test on it. If they'd written unit tests. Which they didn't, because they were bad programmers.

Re:Just for a browser?

[Anonymous Coward]

Hopefully the new toolkit uses standard C++

Read the Google Style for C++ and weep. The best you can hope for is a Frankenstein mix of C++98 and Java 1.5.

Perhaps the STL is not perfect, but it is quite likely already used all over the application itself

Again the Google Style at fault, it does not like the STL: value semantics, operator overloading, assignment operators (combination of the other two), templates, add in some of the more functional concepts and the STL is one big style violation.

Re: Just for a browser?

[netsurfer912]

It's now back up to 2 mayor engines: Webkit and Gecko. Webkit and Blink are the same (as of now) and Trident is not an engine.

Re: Just for a browser?

[Anonymous Coward]

Wrong. GTK+ did their weird object system because it's in C, not C++. Reflection isn't a factor, and as a C++ programmer I don't feel like I'm missing anything without it.

That's kind of like saying that as a loyal Soviet citizen, you don't feel like you're missing anything without democracy or free markets. You may not understand the poverty and limitations you live with, but that doesn't make them or their consequences go away. You live in poverty and under tyranny, and because you have never seriously experienced anything else, you don't even realize it.

Without reflection and dynamic typing, and relying on templates instead, you get bloated executables, difficult extensibility, and complex APIs; in different words, your typical C++ library. In many cases, to add insult to injury, you get poor performance too.

Re:Just for a browser?

[swillden]

i) They don't fix the appalling font rendering issues on Windows promptly and as a priority. Most of Google's own web fonts are unusable in production because of this.

I haven't used Windows since about 2000, so I have no comment on this. I will point out that it appears work is in progress: https://code.google.com/p/chro...

ii) They don't follow standard most-recently-used order when ctrl-tab between tabs and they don't see the problem and close any bug report as won't fix.

I disagree with this one. The Chrome tab ordering is better. most-recently-used sucks when you have 20 tabs and have bounced around between them somewhat randomly (as is normal). It makes ctrl-tab completely unpredictable unless you're just jumping back one or two levels. The Chrome way is better.

iii) They start adding animations to html elements you can't restyle with CSS e.g. the zoom ease-in they added to select elements in a recent Chrome.

Got a link to more information? I'm not sure what you're referring to.

There were wide-spread issues on their recent releases. You can only auto-update if you are rock-solid.

Link? I certainly never noticed any issues, but perhaps that's -- again -- because I don't use Windows.

v) They fork from the web-kit project, a once high-point in cross company collaboration for the betterment of the web. Now... beginning of the end.

Nonsense. There is still cross-collaboration between Blink and Webkit, and Google isn't the only company working on Blink.

I also fundamentally disagree with the common /. meme that forks are bad. There is this mistaken notion that having all of the developers interested in a certain space collaborating on one implementation will improve the pace of development, but that view ignores the fact that software engineering isn't like ditch-digging; with software there are definitely diminishing returns on larger and larger collaborative groups, particularly if the software isn't of a sort that lends itself to crisp, well-separated modules. In practice, we tend to find that with sufficiently-important spaces (e.g. web browsers) the ecosystem is better-served by friendly competition among open source projects. That reduces the amount of inter-group communication needed to reach consensus on approaches and therefore increases the speed of iteration. The fact that all are open source means that when one project implements a great new idea the others can see the details of the implementation and more easily incorporate the idea, even if they can't actually use the implementation.

I think the "we should all work on one implementation" theory has basically the same merits as the old Soviet one-gigantic-factory model for production of goods. On the face of it one would think that producing many different designs for one type of product and then building all of them in separate production facilities, distributed through different distribution networks, etc., is very inefficient. One design, one huge factory to maximize economies of scale should be better, right? But history showed that the opposite is true, that a competitive market produces more goods, better goods and does it at a lower cost. The issues in software are different, but at a high level the emergent properties are similar.

vi) And now they are going to spend their time re-implementing a cross-platform widget toolkit.

They already implemented it. It's been used in ChromeOS for a while. My guess is that they've decided it will take less engineer effort to port and maintain Aura than to keep up with Gtk+. I also wouldn't be surprised if a goal isn't to remove some unused cruft from Chrome on platforms (like Windows) that don't tend to have Gtk+ libs lying around. I doubt Chrome uses more than a tiny fraction of the Gtk+ functionality.

Re:Just for a browser?

[K. S. Kyosuke]

The C++ mentality is that you should catch as many errors as you can at compile time.

No, the Haskell mentality is that you should catch as many errors as you can at compile time (stopping only barely short of executing the whole program at compile time during the type checking :-)). The C++ mentality is that you should improve C in incremental steps without breaking much of compatibility both in terms of source code and ABI (so that you could easily reuse legacy code), until it becomes either bearable, language-feature-wise, or horribly complex, programmer-experience-wise, and ideally both.

Every time I've seen reflection used, it was by some terrible programmer who'd just learned about reflection and was looking for an excuse to use it.

That sounds like a rather lame argument to me. You could say exactly the same thing about C++ itself: the majority of time when C++ gets used (overall population considered), it's some terrible programmer who just discovered that C++ is supposed to be fast, fast, fast (well, it isn't actually all that fast - certainly not on the most recent hardware) and wants to write a game, or perhaps just because the college professor steadfastly requires C++ for his course (and neither of them knows better). Only a small percentage of users ever get to the point that they actually understand the nuances of appropriate language feature usage (that goes for any language). Or perhaps it's like with violins, they don't suck just because most people who try to play them suck. Or do you think that Sturgeon's law somehow doesn't apply to people?

I've never seen its use actually improve a design. I've never even seen it make more loosely-coupled or reusable code, now that I think about it.

Well, perhaps you ought to look at Lisp or Smalltalk one day. The whole Smalltalk IDE effectively runs that way. In fact, any self-sustaining system similar to those essentially *requires* reflection and introspection by definition. Perhaps that's not an argument for C++ code (which uses a rather ancient kind of development workflow, what with the absence of incremental compilers), but as Gtk+ tries to cater to the needs of users of many languages (I'll leave aside how well they succeed at it, that's an entirely different question), it's no surprise to me that they're so heavily reflection-oriented - languages like Python, Ruby, or Lisp essentially *thrive* on this kind of binding library implementations.

Re:Just for a browser?

[Zukix]

font rendering issues on Windows

I will point out that it appears work is in progress

It has been "in progress" for so long that stationary might be a better term.

They start adding animations to html elements you can't restyle with CSS

Got a link to more information? I'm not sure what you're referring to.

http://stackoverflow.com/quest...

There were wide-spread issues on their recent releases. You can only auto-update if you are rock-solid.

Link? I certainly never noticed any issues

Read the chrome product support forums or comment rolls on release announcements - its pretty depressing. Countless posts referring to the same set issues crashes, scrolling bugs, inability to pick items from select boxes, flash crashes, modal dialog faults etc.

http://frankcode.wordpress.com...

They already implemented it. It's been used in ChromeOS for a while.

I see. I wouldn't be surprised if this is the root of the problem. They have conflated a web-browser with an OS. No good will come from it except an unfocussed bloated browser and an anaemic OS.

Re: Just for a browser?

[serviscope_minor]

I'm assuming webkit and blink will diverge. And isn't trident what IE uses?

Re:Just for a browser?

[hairyfeet]

How many times is it gonna be pointed out that Google IS MSFT of the 90s before folks notice its the same dance? EEE? Webkit for Blink, taking more and more of Android behind the proprietary wall (look up Android on Ars for a great set of articles on the behind the scenes of android) its the same fricking dance folks, we've been here and done that and just because a company says "Nooo, we aren't evil, nope, nuh uh" don't make it so. Hell you watch the ads for Halliburton you'd think its a company made of kittens and sunshine, don't make it so.

Maybe I'm wrong but I think its like a law of nature, company starts out cool, gets popular, and then when they reach #1 they get paranoid and start caring more about staying big cheese than on what got them there and turn nasty. After all once upon a time MSFT was just a software house that would happily sell to anybody, even had their own UNIX for a time, but they got to the top and went nasty. Google USED to be this cool mad scientist "throw it at the wall and see what happens" kind of company, now they actively ignore their users (How many have said "do not want" to G+ yet they keep bitchslapping you with it?) and pull douchey moves like this with pretty much zero advantage to anybody but themselves.

History is repeating folks, and i only hope enough notice it and take action before we end up with a half a decade of "This page requires Chrome x.x" because as someone who survived the IE era I can say without hesitation that having one company control the web? REALLY sucks.

Re: Just for a browser?

[Anonymous Coward]

Sure, if you're happy to be forced to recompile all binding libraries

Someone has leaky abstractions.

generate new packages and force hundreds of thousands of people to upgrade them whenever the interfaces change instead of just upgrading Gtk+.

Or you just use symbol versioning as supported by dynamic libraries for ages. Also if I understand package management correctly it is an either or situation, if you create a new package it should not conflict with the old package and if it replaces the old package it should stay compatible.

r. Of course, in that case your brain is already so horribly damaged that you can't perceive the comparatively

While I miss automatic reflection support in some C++ projects that is an awful way present your opinion as superior. In fact it makes you look like a troll.

Re:Just for a browser?

[HeckRuler]

oh shit dude, and you trusted them? You already dun fucked up.

I just made a POS boring as hell business app for family*. I slapped a gui over a simple console application because ohbytheway while you're working on it we want the user to be able to redefine the sorting ruleset at runtime.

It has labels, text entry boxes, and comboboxes, and to pack it I used vboxes, hboxes, and tables.
THAT'S IT.

Here's the libraries I needed:

freetype6.dll
intl.dll
libatk-1.0-0.dll
libcairo-2.dll
libcairo-gobject-2.dll
libexpat-1.dll
libfontconfig-1.dll
libgdk-win32-2.0-0.dll
libgdk_pixbuf-2.0-0.dll
libgio-2.0-0.dll
libglib-2.0-0.dll
libgmodule-2.0-0.dll
libgobject-2.0-0.dll
libgthread-2.0-0.dll
libgtk-win32-2.0-0.dll
libpango-1.0-0.dll
libpangocairo-1.0-0.dll
libpangoft2-1.0-0.dll
libpangowin32-1.0-0.dll
libpng14-14.dll
zlib1.dll

Yes... yes... I developed a windows program with the GTK. Shut up.

*NEVER get roped into working for family by the way.

Re:Just for a browser?

[kthreadd]

I don't see anything Gnome-related in that list that I didn't mention.

Re:Just for a browser?

[swillden]

It has been "in progress" for so long that stationary might be a better term.

Did you look at the link? The last update was in January, and described the current status as depending only on one other piece to be ready for launch.

I see. I wouldn't be surprised if this is the root of the problem. They have conflated a web-browser with an OS. No good will come from it except an unfocussed bloated browser and an anaemic OS.

I can see you've never used it.

Re:Just for a browser?

[Arker]

A lot of truth there but not this:

"Well, it used to be that Android was basically a linux distro."

No, it really never was. Because when you say a linux distro what you are actually thinking of is a GNU distro, and android never attempted that at all. It's always been it's own wierd little proprietary linux offshoot and I guess it's nice that people are starting to see it that way, but it's silly to pretend it's a change, it was this way from the start.

Re:Just for a browser?

[Aighearach]

Qt is superior to Gtk in almost every way! Oh, right. Except for the lack of actual ways that it is better. Or even substantially different other than in style. It is clearly more cross-fanboy compatible.

My theory is that we hated on Qt and the authors years ago, but then there was hate for Gnome3 more recently, and some of the hate is falling on Gtk by people who don't use it, and don't realize that Gtk is just an upstream library and looks just like it did in the past and Gnome has little to nothing to do with it.

Re: Just for a browser?

[serviscope_minor]

Oh wait, you're C++ programmer. Of course, in that case your brain is already so horribly damaged that you can't perceive the comparatively mild benefits this would bring to a C++ programmer (removing only one cut from the thousand ones already killing you).

This is the problem with almost everyone who denigrates C++ and trys to hawk a better language. Their minds and outlook are simply too small and limited to see or understand the vast number of utterly different application domains spanned by C++.

You see reflection as a benefit because you work in a limited application area where reflection is useful (and where C++ would benefit). All those don't care types don't not care because they are brain damaged, they don't care because reflection would not have any benefit to them.

I do most of my coding around heavily algorithmic, scientific computing style stuff, with a side order of deep embedded. Reflection would not help me one bit day to day.

Yeah, I can see where it would be useful, but I don't write code in those areas, so it's quite low on my list of features I want.

But since I'm clearly so brain damaged, perhaps you'd like to enlighten me with what I should be using instead and why reflection would help me day to day.

Re: Just for a browser?

[H0p313ss]

The great wiki in the cloud says it is: Trident (layout engine)

All hail the great wiki!

Re:Just for a browser?

[Anonymous Coward]

standard most-recently-used order

What exactly is standard about going through tabs in a most-recently-used order? The browsers I have been using go to the "next" tab to the right when you hit Ctrl-Tab. If you want these damn things to be "Applications" put your browser tab in a new window. Also: one second you say a web browser is basically text in windows then you want to call it an application?

Re:Just for a browser?

[morgauxo]

Didn't Microsoft start when Bill Gates took a bunch of freeware code from a computer club, ported it to a new platform, copyrighted it all and called it his own? That's not quite how I remember Google starting out.

Re:Just for a browser?

[Anonymous Coward]

I thought that the GTK+ and QT did their own object models since they were done with either for C or early versions C++. Why should a toolkit be designed and implemented with any imaginable language in mind? Just do a library that works well with one language, and ignore the others. FLTK seems to be a toolkit done in this way.

Re:Just for a browser?

[shutdown -p now]

This has nothing to do with the original point of the thread, since you're showing a dependencies of a Gtk 2.x app (and doesn't involve anything from Gnome). OP was claiming that Gtk 3.x introduced dependencies on Gnome.

Re:Just for a browser?

[shutdown -p now]

Don't forget high DPI issues on Windows. It seems like Chrome is still the only app that I actually use that's not high-DPI-aware and so gets bitmap upscaling.

Re:Just for a browser?

[shutdown -p now]

The Chrome tab ordering is better. most-recently-used sucks when you have 20 tabs and have bounced around between them somewhat randomly (as is normal). It makes ctrl-tab completely unpredictable unless you're just jumping back one or two levels. The Chrome way is better.

The point is that Alt+Tab is most-recently-used-app pretty much everywhere, and Ctrl+Tab is most-recently-used-tab (or document) in all multitab apps. This actually makes a lot of sense when you have many tabs open, but are going back and forth between a few of them, e.g. copy pasting text or some such - as GP rightly notes, this is especially common once your tabs run "web apps" rather than just websites.

If you just want to switch to next/prev tab, there's already a shortcut for that, Ctrl+PageUp/Down. Ctrl+Tab is really meant to be the "switch to whatever I was using before" shortcut, fulfilling a different but also very common need.

Re:Just for a browser?

[Kremmy]

In my experience, I have an awful lot of tabs open. If my ctrl-tab/ctrl-shift-tab jumped to the last used tab, it would be a disorienting mess. Makes sense to alt-tab back and forth, but I remember when alt-tab used the old way of it, where we didn't have the most recently used application popping up. Sure made it a lot less frustrating back when we were using multiple applications do things and needed to switch between them without worrying about them randomly changing order in the task switcher. I swear kids these days don't know how to tell their computers what they want and expect them to just know.

Re:Just for a browser?

[hairyfeet]

Nooo..Bill Gates and Paul Allen wrote a BASIC interpreter on the school's PDP11 which they sold to Altair, which gave us the famous "computer club piracy" speech. Now one could argue whether or not the computer time was "stealing" but they got their BASIC commands from the same place everybody else did, Knuth and his books of the early 70s.

Re:Just for a browser?

[Kremmy]

This completely overlooks a very basic incompatibility that Qt has with C. The fact that it's written in C++, and while there are ways to bind C++ libraries to C, it's not exactly what one might call elegant. Qt has a lot of merits - it is absolutely not a replacement for a GUI toolkit used from C, which is what GIMP needed. Must remember that C++ was quite a bit less stable at the time. Make sure your .NET runtimes are up to date so you can draw colored rectangles into the Metro overlay.

Re:Just for a browser?

[Anonymous Coward]

Every time I've seen reflection used, it was by some terrible programmer who'd just learned about reflection and was looking for an excuse to use it.

This needs to be said more often. Reflection IS NOT an appropriate solution in almost every case.

Re:Just for a browser?

[allo]

> vanishingly small number of people
Seems like someone underestimates the gimp users.

Re:Just for a browser?

[Anonymous Coward]

> .dll
you are doing it wrong.

Re:Just for a browser?

[hobarrera]

Well a great truth is, we used to have issues with qt due to it's former license. IIRC, it used to be dual commercial/GPL. It isn't anymore.

I'm with Google...

[evilviper]

Porting a huge GTK-2 application to QT is substantially more work than just forking GTK-2 with a new name, and maybe some small modifications.

And adopting and modifying GTK-3 seems like a battered woman going back to her abusive husband... The FSF has a sorry track record of destroying everything they touch, and the GPLv3 getting forced on all their software is making individuals and companies run for the hills at a record pace.

Hey, if Google makes Aurora a distributable package, I'll compile all the GTK-2 apps I use against it, and just keep using it. I even still depend on a few GTK-1 apps...

Re:I'm with Google...

[Anonymous Coward]

And adopting and modifying GTK-3 seems like a battered woman going back to her abusive husband... The FSF has a sorry track record of destroying everything they touch

These things are related how exactly?

Pro-tip: Don't make grand statements from a position of ignorance.

GTK+ is managed by the GNOME Project which is independent of the FSF.

Re: I'm with Google...

[Anonymous Coward]

For starters, Aura isn't a fork of GTK.

Re:I'm with Google...

[phantomfive]

Reading through the documents, it doesn't look like a trivial task to recompile all your GTK-2 apps against it. From the UI Toolkit standpoint, it looks like a combination of NextStep and Swing.

AFAIKT Aura is a more than just a UI Toolkit, it's a complete Window Manager. A replacement for Gnome (wow! I hope that takes off!) Apparently it's been running on the Chromebooks. Here is Linus' take on the topic.

The main reason I would be reticent to use it is because Google doesn't always have a strong commitment to backwards compatibility. So you may end up having to rewrite pieces of your code, just to keep them compiling. If you're ok with that though, go for it.

Re:I'm with Google...

[peppepz]

GTK+ 3 is LGPLv2, not GPLv3; it is not developed by the FSF, and never has been. And the GPLv3 is arguably more friendly for businesses than the GPLv2, with its explicit patent provisions, the lack of the termination provision, and the explicit system libraries exception.

Re: I'm with Google...

[Anonymous Coward]

Oooh, shots fired.

Re:I'm with Google...

[Anonymous Coward]

The main reason I would be reticent to use it is because Google doesn't always have a strong commitment to backwards compatibility. So you may end up having to rewrite pieces of your code, just to keep them compiling. If you're ok with that though, go for it.

GTK isn't better in that regard.

Re:I'm with Google...

[K. S. Kyosuke]

AFAIKT Aura is a more than just a UI Toolkit, it's a complete Window Manager.

You are aware, that - at least in sane computer system designs - this is mostly the same thing?

Re:I'm with Google...

[David Jao]

If you like the GPLv3 so much, no problem. Any copy of LGPLv2 software can be converted to GPLv3 at will. The text of the LGPL itself states:

3. You may opt to apply the terms of the ordinary GNU General Public License instead of this License to a given copy of the Library. To do this, you must alter all the notices that refer to this License, so that they refer to the ordinary GNU General Public License, version 2, instead of to this License. (If a newer version than version 2 of the ordinary GNU General Public License has appeared, then you can specify that version instead if you wish.)

Re:I'm with Google...

[Anonymous Coward]

GTK+ is part of GNOME, which is part of GNU, which is managed by FSF.

Re:I'm with Google...

[phantomfive]

Not at all, there is no system where everyone wants to use the same UI toolkit (mainly the API, but also look and feel), for various reasons good and bad. Mostly though, every program on a system will want to use the same WindowManager; that is, you normally don't want more than one WM running at the same time.

Re:I'm with Google...

[phantomfive]

Code that has already been released as LGPLv2 can always be used as LGPLv2. No one can stop you from using it under LGPLv2 once it's been released under that license (or any other license; note that there was a change in the licensing of QT, for example).

Re:I'm with Google...

[David Jao]

If you invoke GPLv3 for your copy of the code, then you can stop other people from using your copy under LGPLv2. If you made modifications to your copy, then GPLv3 applies to your modifications. Of course, other copies of the code are licensed independently of your copy. I never claimed otherwise.

Re:I'm with Google...

[K. S. Kyosuke]

I didn't say that Aura will (or should) force you to do that, I'm just saying that the conceptual distance between the two is very small to begin with.

Re:I'm with Google...

[evilviper]

The GPLv3 is "arguably more friendly for businesses" in the same way the Earth is argueably flat... You COULD argue that, yes, but overwhelming evidence points to the contrary, and no sane person will agree with you...

Re:I'm with Google...

[MrEricSir]

And adopting and modifying GTK-3 seems like a battered woman going back to her abusive husband... The FSF has a sorry track record of destroying everything they touch, and the GPLv3 getting forced on all their software is making individuals and companies run for the hills at a record pace.

You can't blame FSF for what happened to GTK3. Besides, GTK3 is licensed under LGPL 2.

No, if you want to point fingers at GTK3, blame the developers. Also note that GTK's developers do all their work from the command line -- not exactly the people you want designing a GUI toolkit.

Re:I'm with Google...

[phantomfive]

Yes, but there's also a very good reason to not have all the functionality of a windowManager within every UI toolkit (mainly because there's no need to duplicate all that functionality)

Re:I'm with Google...

[Kremmy]

It is also important to note that this is an OPTIONAL clause of the GNU General Public License - it is in fact a clause that many have chosen to leave out, as they do NOT want their code to be grandfathered into GPLv3.

Someone please explain

[LordLimecat]

Why it really matters whether Google uses QT or GTK or their own stack. I mean for a GDE or distro like Ubuntu, I can see that "make another one" matters because it impacts all sorts of other projects. For Chrome, though, it doesnt really affect anyone else that I can see, and its really just Gnome folks being upset that Google didnt want to use their stack. At the end of the day, isnt it just more work for Google? If theyre happy to do it, who cares?

And-- though Im not privy to all of the politics-- Ive gotten the impression that the GTK3 folks werent terribly interested in hearing other people's thoughts.

Re:Someone please explain

[LordLimecat]

Double-post, but why is this in the news now? All of the linked design docs are from Dec 2011. This stuff is 2 years old.

Re:Someone please explain

[Anonymous Coward]

Your impression of the GTK3 folks is a bit misled.

They listened very intently to other people, but those people were the "other other" people in their usability labs. The mere only "other" people were basically all the users that they didn't drag into a usability lab, who decided that they didn't care if one could provide a more consistent interface; they wanted their damn Windows 98 style start bar, and weren't about to spend ten minutes learning anything else.

It is a common problem. Do nothing and you eventually lose the market due to "lack of innovation". Do something and you lose a piece of your market due to lack of user migration.

I say let the dinosaurs rot in their self chosed La Brea tar pits.

Re:Someone please explain

[Anonymous Coward]

Why it really matters whether Google uses QT or GTK or their own stack.

When we're running apps, we inevitably end up with using at least one QT app, at least one GTK app and probably in future at least one Aura app. These libraries have a huge level of duplication (e.g. each one will have a completely separately implemented file dialog). Add to this that each library will be used in several incompatible versions and you end up with serious bloat.

At the end of the day, isnt it just more work for Google? If theyre happy to do it, who cares?

Nobody is happy to do needless work. If Google were happy with the current situation then they would have continued contributing to the existing libraries. This would have made those better. This sounds like a serious warning that something is wrong. Whether it is Google's wish to take control of all free software or it is GTK3's inability to look after their customers. We need to know what the problem is.

Ive gotten the impression that the GTK3 folks werent terribly interested in hearing other people's thoughts.

This sounds like a serious problem; do we have any proper evidence? Given the way that previous Gnome changes have effectively caused the Ubuntu/Mint split and so on, it sounds to me like GTK3 needs to be eliminated. That sounds like serious work and will might put Linux back noticably. On the other hand, if the GTK3 people are responsible for causing things like this (and I'm not saying that they are; this may be Google's fault) then getting rid of them will be a good thing long term.

Re:Someone please explain

[Anonymous Coward]

Double-post, but why is this in the news now? All of the linked design docs are from Dec 2011. This stuff is 2 years old.

It's going live now. The stuff has been experimental for that time, it's just now being pushed into Release builds.

Fun fact is that Aura is already enabled on Windows, this is why scrollbars, buttons, combos and everything else now looks like shit and are missing usability features that every other scrollbar on the system has.

Re:Someone please explain

[peppepz]

When we're running apps, we inevitably end up with using at least one QT app, at least one GTK app and probably in future at least one Aura app. These libraries have a huge level of duplication (e.g. each one will have a completely separately implemented file dialog). Add to this that each library will be used in several incompatible versions and you end up with serious bloat.

That's true, but how much can that bloat amount to? 20 MB? 100 MB? It won't be much relevant for today's standards. Code duplication is what happens regularly in the closed source world, where applications ship with a private version of all the libraries they use, and not only for the UI - with few people complaining.

Ive gotten the impression that the GTK3 folks werent terribly interested in hearing other people's thoughts.

This sounds like a serious problem; do we have any proper evidence?

https://mail.gnome.org/archive... - don't know if things have changed in the last two years.

Re:Someone please explain

[brunes69]

"Looks like shit" is subjective. Personally I feel like Chrome "looks like shit" on Linux compared to Windows and Mac OSX where it has a consistent look. I am looking very much forward to this move. Like many nowadays I live in the browser and spend 95% of my time there. The more it is consistent across platforms the better. It doesn't matter nearly as much that my browser on Linux looks like my Eclipse as it does that my browser on Linux looks and behaves like my browser on Android or Windows or Mac.

Re:Someone please explain

[davester666]

Really?

I think that most people don't care about how Chrome works on multiple OS's because they only use one. And for that one, they prefer that app looks and works like apps for that platform.

And even the people that say, using Windows at work and MacOSX at home, probably would like the two versions to work similarly, but still look and feel like other apps for each specific platform, and not do shit like Adobe and try to make an AdobeOS app that happens to run on Mac or Windows.

Re:Someone please explain

[williamhb]

Why it really matters whether Google uses QT or GTK or their own stack. I mean for a GDE or distro like Ubuntu, I can see that "make another one" matters because it impacts all sorts of other projects. For Chrome, though, it doesnt really affect anyone else that I can see, and its really just Gnome folks being upset that Google didnt want to use their stack. At the end of the day, isnt it just more work for Google?

I guess it depends whether their interest in it is limited to "we need something to write Chrome using, and GTK isn't doing it for us any more" or whether they will later be saying "come write apps for Chrome and ChromeOS using NaCl and Aura". Google has taken on their own UI stack -- is their only interest in it really to write just one application? If it is instead another step in the direction of encouraging developers to write apps that only work in Google's browser, that would be interesting to hear.

But I haven't looked into it closely.

Re:Someone please explain

[Anonymous Coward]

Bloat is often not about megabytes, but about security vulnerabilities. A larger unnecessary code base increases the amount of attack vectors.

Re: Someone please explain

[Anonymous Coward]

"Like many nowadays I live in the browser and spend 95% of my time there."

Really? I live in an apartment.

Or do you mean you spend 95% of your time browsing the web? If so, you must not sleep much.

Or do you mean you browse the web 95% of your time at work? If so, don't you get bored, and what kind of job lets you do that?

Re:Someone please explain

[Anonymous Coward]

Changing things that work is only "innovation" if it actually results in a real, useful improvement, which is frequently not the case. Whether the change is useful should be determined by the users, and not some misguided "usability lab" committee often driven by politics or egoism. People have every right to resist unasked for changes that are useless or even harmful. Those can be forced down the throat of users if you have a monopoly (Microsoft), but for open source software like GNOME, they just contribute (deservedly) to maintaining a very small market share. Want to force crap like GNOME 3 on your users ? No problem, they will just delete Linux and go back to Windows.

Re: Someone please explain

[Anonymous Coward]

Jesus christ, just shut the fuck up.

Re:Someone please explain

[LordLimecat]

To your points, it seems to me that one could argue that its only bloat if youre using QT and GTK and Aura, which many users will not. On the other hand, if someone made One Library to Rule Them All and it hit all of the requirements for all of the projects, that too would be bloat that everyone had to deal with.

And I kind of get the impression that Google is happy to donate to some things where they like the direction, but theyre not about to change their course to cater to or help out another project. If GTK isnt going to do what they need, Google is going to create their own (see also Webkit/blink).

My comment about "not listening" was based on the terrible feedback every GTK3 project seems to have gotten by anyone who cares, and the relative lack of action from the GTK folks. As I say though Im not really following the politics that closely.

Re: Someone please explain

[Anonymous Coward]

What's Windows?

Re:Someone please explain

[spitzak]

The Unix philosophy would help with the file dialogs. They should be a separate executable, at least for simple ones that are in a floating window. Then all the toolkits could share them. More importantly people could write new and better file dialogs. Also the common one could share a daemon that kept track of preview icons, etc.

FLTK was tiny, but more than 50% of the code was for the file chooser dialog.

Re: Someone please explain

[Anonymous Coward]

It does NOT look consistent on Windows.

Re:Someone please explain

[shutdown -p now]

Enforcing modality would be a bit tricky for a dialog created by a different process. And if it's non-modal, then what should be the expected behavior if user switches back to the main app and navigates away from wherever it is that he was when he caused the dialog to appear, then switches to the dialog again and opens a file?

Re:Someone please explain

[spitzak]

The main app is aware that it is executing the file dialog program, they are talking to each other (probably just by using a pipe). Therefore the app can tell the file dialog that it lost focus, or (more likely) it can just stay in a modal state where the user cannot change things until they dismiss the file dialog.

Re:Someone please explain

[shutdown -p now]

If the user can bring up the main app window such that it overlaps and covers the dialog, but remains in a modal state where it's inactive until the dialog is closed, that can lead to severe user confusion. Basically, for this to work, you need a cooperative WM that understands the notion of window relationships across process boundaries.

Re:Someone please explain

[spitzak]

Transient-for hint works across process boundaries.

Re:Someone please explain

[shutdown -p now]

Ah, I didn't realize that it is already there. Yes, that makes a lot of sense then.

is Google trying to do Canonical?

[Anonymous Coward]

No. Google is much more effective, and evil than canonical.

It's gonna be fine

[jones_supa]

Qt is my golden standard too, but in case of Chrome, it does not matter much. Go with "Aura" if it makes them happy. I mean, how many UI widgets do you see in Chrome anyway? There's the tab bar, pop-up menu, and some little popup thingies here and there. Everything else is a web page, which is rendered with its own engine.

Re:It's gonna be fine

[Anonymous Coward]

It isnt about how its widgets look. It is how Chrome fits to your environment. You want coherency on desktop environments so every application looks and behaves in same manner. That every window follows your window manager settings and behavior.

You are required to use Chrome in KDE with native window manager to get it work like it should with animations, effects, bindings etc. But same time you make Chrome look terrible as it has own style and you can't find a style what still would make it look good.
If you take Chrome own window manager to use, you don't get effects, functions, shortcuts but at least the browse looks good as standalone, but doesn't still fit at all to environment.

Re:It's gonna be fine

[Anonymous Coward]

How can Qt be standard, when C is the absolute standard for Unix, BSD and Linux?

Re:It's gonna be fine

[Anonymous Coward]

Especially considering how much overhead the entire Qt would bring.

Qt is very nice, easily my favorite UI toolkit (although I must admit I have gripes; on OS X, it doesn't properly emulate native behavior, things like Emacs keybindings - which normally work in OS X text widgets - aren't there), but Chrome doesn't need most of its features, and a web browser with separate states per tab with the complexity of todays web pages doesn't need too much extra overhead...

behind the aura, behind the aura, behind the aura

[mtbink.com]

behind the aura, behind the aura, behind the aura

Google Chrome also uses Blink-182.

Do Canonical?

[peppepz]

By saying that, do "the Free Software Community" mean making Linux accessible to many users that wouldn't have dreamt of using it before? Being the first ones to provide a distribution that you can actually recommend to a computer illiterate?

And then again, why should anyone have a say on what toolkit Google decide to use for their own browser? Did "the Free Software Community" have anything to say when it was slang vs ncurses, emacs vs vim, gtk vs qt, gnome vs kde? No, because exploring alternate solutions is good for the whole community in the long run. Please stop this poisonous attitude of finding "enemies of the people" among people who dare write free software.

Re:Do Canonical?

[the_fat_kid]

no, no one has ever had anything to say about emacs vs. vim.

I am all for exploring solutions and like the linux community as a whole but let us not pretend that flame wars in FSF lands haven't started over less.

Re:Do Canonical?

[peppepz]

The emacs vs vim flamewar is about which one is the best (that is vim, of course), not about the legitimacy to exist of either. I've never heard anybody suggest that people should stop developing vim and contribute to emacs instead.

Re:Do Canonical?

[narcc]

pico

Flame on!

Re:Do Canonical?

[Anonymous Coward]

Now your judgement is in question, if you can't recognize the truth and the glory that is emacs then your taint must run deep.

j/k

Re:Do Canonical?

[Anonymous Coward]

both wrong , emacs

Re:Do Canonical?

[LoRdTAW]

I'm writing a text editor, the one text editor to rule them all. It's called femto. Everything you type is so small you can't see it. When you save the file, its size so small you can't see it in the directory listing! Its going to change the way we edit.

Re:Do Canonical?

[thejynxed]

As far as text editing, etc goes... Notepad++

*cough*

Only because I've done hand-editing of web pages, etc for years now using it and would never, ever go back to using emacs, vim, or the rest. It's also highly extensible, which is a big plus.

Re:behind the aura, behind the aura, behind the au

[neiras]

Google Chrome also uses Blink-182.

Woo hoo! So they lie and they're easy?

Why care?

[msh104]

To be honest i see this more as a feature than as a problem.
This will very likely improve the quality of the linux build making it more complete and compatible with the windows build and features.

Just compare the linux and windows versions of firefox for example.
They look far from the same.
And for a big part this is caused by the difference in toolkits used beneath the skin.

Now i am a big fan of QT.
But even if they port their own: one toolkit everywhere can only make things better.

Re:Why care?

[Anonymous Coward]

I think the Linux version of Firefox looks like crap. It's kind of 'fuzzy' and not as sleek as the Windows one.

Re:Why care?

[BitZtream]

Firefox is an app that runs on XULRunner. XULRunner uses no native widgets, it has its own widget set and its own interface definition language, XUL. Firefox's entire user interface is written in XUL. The widget set in code is the same for OSX, Windows, Linux, Solaris, and whatever else it runs on.

The interface only looks different between windows and linux or mac because the default CSS theme for the application is auto detected and selected at startup so the widgets 'look and feel' native, but again - they aren't.

You can make the widgets look like any OS you want, make it act a lot like most other OSes as well, though some functionality is different such as the file open dialogs and such, which I guess you can count as 'widgets', which are actually native depending on OS/features.

At Mozilla, there is only XUL - http://www.mozilla.org/keymast...

Seriously though - http://en.wikipedia.org/wiki/X...

Re:Why care?

[Blaskowicz]

I like it better on linux as I get the full "file edit view history.." menu per default.
It looks good in a GTK2 desktop if the system-wide icon theme is decent.
the worst thing with it is that I left myself open way too many tabs and keeping track of them is near impossible.

Qt?

[unwesen]

"Couldn't Google just switch to Qt, which is becoming an industry standard?"

It is? I haven't seen evidence of that. Trolltech/Digia have been working on that for a long time, and have in fact gained significant market share, but I don't see many projects outside of the KDE sphere of influence or very specific embedded platforms adopting Qt. In fact, the popularity of entirely new mobile platforms that did not adopt Qt is a great counter-argument (i.e. iOS, Android, ChromeOS).

Mind you, that's no argument against using Qt - I just don't see evidence of it becoming an industry standard.

Re:Qt?

[chaboud]

I came here to say this.

I'm quite the fan of Qt, but it's far from an industry standard. HTML5 + wrapper probably has as much, if not more, adoption.

And, once you use iOS or Android to dev GUI, some modern, convenient, and well-crafted patterns begin to emerge. They're not perfect, but they're nice to use. Honestly, if Google wants to use their own toolkit and publish it as open source, why should anyone complain about that? Some very interesting ideas may come out of it and be brought into other projects. Just as Mozilla's XUL was clearly aped for Microsoft's XAML, open source contributes to the field as a whole, not just one particular project. There's no need to lick the pizza with open source.

Only the ever-trolling slashdot community could turn Google releasing and dog-fooding an open source project into a bad thing.

Re:Qt?

[Anonymous Coward]

Go look what industry software is written in, not just mobile crapware.

All the big player proprietary softwares are shipped with QT4 anymore. Automotive is also embracing it finally along with QNX... which I suspect will bleed back into smartphones if the market forces allow.

Re:Qt?

[Anonymous Coward]

HTML5 + wrapper probably has as much, if not more, adoption.

Qt does provide a very convenient HTML5 wrapper, though. So that's not mutually exclusive.

Re:Qt?

[hobarrera]

Perhaps not "industry standard", but it IS used by several proyects that dominate in their area, and several quite large (lage as in "with a large userbase") projects:

Skype, Vlc, Teamspeak, Origin, and lots more:
https://en.wikipedia.org/wiki/...

Re:Qt?

[Thanosius]

Indeed. Google Earth, qBittorrent, VirtualBox in addition to those you mentioned...

Anyone who things Qt isn't incredibly popular with rich applications is fucking blind or biased.

Re:Qt?

[Anonymous Coward]

please , be specific , HTML5 + wrapper ?? which wrapper ?? what wrapper ??? for which platform ?? QML already done great job.(however i am gtk guy)

Re:Qt?

[gbjbaanb]

choice is good, but too much choice is bad - imagine if every project rolled its own GUI toolkit :(

Re:Qt?

[civilizedINTENSITY]

Mathematica use QT.

Re: Qt?

[netsurfer912]

Exactly.

Re:Qt?

[Anonymous Coward]

XUL was released earlier than XAML

Re:Qt?

[T.E.D.]

"Couldn't Google just switch to Qt, which is becoming an industry standard?" It is? I haven't seen evidence of that.

On Windows, yes it is.

My employer, traditionally a very dedicated Microsoft toolchain Windows shop, is currently transitioning to Qt for GUIs. The only option MS really supports these days is .NET based, which means a closed-source vendor is forced to buy some expensive bytecode munging utility if they don't want their code reverse-engineerable back to its sources. Its also damned inconvenient if there's some reason your application needs to be "unmanaged".

Our alternatives are to go with way obsolete MFC, or use some other third-party toolkit. Qt appears to have been judged the best available of these.

As near as I can tell, we are far from alone. This appears to be a common logic path that Windows vendors are following to Qt.

Re:Qt?

[Anonymous Coward]

It is an "industrial standard". As in if you were to write an application for the industry (which in this context means for someone other than the software industry) you would probably use C/C++ and Qt or java+whatever.

  I work for a company that does exactly that : write industrial applications (in france it is even often refered as "industrial software engineering" as if it was somehow a complete different field from other software engineering)

  E.g: right now we are making a Simulator for nuclear power plants control stations. C++ and Qt (and qml Yay !)

  The thing is if you are building some software that will power an industrial system that might be used for decades (average time is circa 30 years for most projects here) then you want it to be vendor independant and cross-platform as much as possible.

  there isn't much other than C++\Qt or java that is mature (but not completely ancient/outdated) and cross platform.

Re:Qt?

[Anonymous Coward]

Nah, choice is never a bad thing.

There's absolutely no way that every project could roll their own toolkit. What would happen in reality is that new/smaller projects would adopt their pick of the older/larger project's toolkits. Then major toolkits get more contributions/bugfixes and end up competing with one another (or falling behind). This is the ever-changing ecosystem of free software. Without this central tenet we'd have lock-in and stagnation.

So I say bring it on Google; lets see what you've got to offer. We don't have to use it, after all.

Re:Qt?

[strikethree]

Only the ever-trolling slashdot community could turn Google releasing and dog-fooding an open source project into a bad thing.

Jesus fucking H. Christ. What the fuck dude? Yes, there are fucking trolls on Slashdot. OOOOO. That is big news. But wait, that is NOT what you said, you said the entire fucking community is against what Google has done.

Fuck off with your generalizations and your bad attitude. Jackass.

GTK+ is for GIMP!!!

[erroneus]

Every use of GTK outside of GIMP is a problem. Try running the latest CentOS with GNOME and see if you can run a newer GIMP. You can't. You will have to do all manner of things and you still will not get 100%.

I have discussed this topic with GTK, GIMP and GNOME projects and at the end of the day it comes back to GIMP/GTK developers. They say GTK is for GIMP. So every developer out there would be well advised not to use GTK any longer.

Re:GTK+ is for GIMP!!!

[kthreadd]

Every use of GTK outside of GIMP is a problem. Try running the latest CentOS with GNOME and see if you can run a newer GIMP. You can't. You will have to do all manner of things and you still will not get 100%.

I don't see how this is specific to GTK. If a program depends on newer versions of libraries, then you obviously need the newer libraries.

I have discussed this topic with GTK, GIMP and GNOME projects and at the end of the day it comes back to GIMP/GTK developers. They say GTK is for GIMP. So every developer out there would be well advised not to use GTK any longer.

But it's also for a lot of other projects. Gnome is largely based on GTK, and it's commonly used outside the Gnome project as well.

Re:GTK+ is for GIMP!!!

[Anonymous Coward]

What I believe he's saying is that the GTK+ devs don't care about anything except supporting GIMP. I don't know if it's true.

If true, it's too bad. There is an important place for a pure C GUI toolkit, because most languages don't integrate well with C++.

Re:GTK+ is for GIMP!!!

[kthreadd]

That is absolutely not the case. GTK+ started out as part of the Gimp project, but is now developed separately under the Gnome project.

Re:GTK+ is for GIMP!!!

[dbIII]

That one is deliberate to try to kill gnome2. Personally I think it should have been implemented differently, and I've got some very rude words and accusations of incompetence ready if I ever meet any of the current gnome team.
Those donkey fuckers managed to create something resembling DLL Hell on linux for the first time and made us all look bad.

Re:GTK+ is for GIMP!!!

[dbIII]

I don't see how this is specific to GTK. If a program depends on newer versions of libraries, then you obviously need the newer libraries.

It's far worse than that because they deliberately broke their own naming convention to prevent the older software from running on a newer system and vice versa. They brought something like DLL Hell to linux for the first time.

Re:GTK+ is for GIMP!!!

[erroneus]

Problem:

GNOME, an environment library is used by an application as well. If an older version of the environment is used, one cannot upgrade the application. Does that sound like a problem to you?

If GIMP were to do it right, it would make calls THROUGH the UI afforded by the OS and environment, not directly to libraries. But that's not what's happening. GNOME says "It's GTK, speak to the GTK people" GTK says "It's GIMP, speak to them." GIMP says "GNOME develops GTK now."

Someone is doing it wrong. And the result of that is Linux GIMP under a common distro is a second-class citizen.

Re:GTK+ is for GIMP!!!

[Anonymous Coward]

exactly , this is bigeast problem with Qt . whole "C++" idea seems is wrong . just look at gobject introspection, how well gtk done binding , without any big effort

Re:GTK+ is for GIMP!!!

[BitZtream]

dllhell has always been a problem, you just don't realize it because you hide the process behind a package manager and you update all of your software at once most of the time. Since the repos at the package manager get all the software working together and built, you don't see the problem. Its there. Someone else does it for you.

Generally doesn't work when you start using closed source apps on Linux, which people in the real world do.

Or when you need to use an old version of some software and the new version of some library that some jackass (like the gtk team) went and changed function names and class names ... because ... thats right ... because ...

Re:GTK+ is for GIMP!!!

[kthreadd]

At some point you just have to pick what you want to use. RHEL and its clones are intended to be a stable environment. 99+ % of users should be ok with the version of Gimp bundled with the system. Should the Gimp developers bend over backwards to support old enterprise systems? Maybe, but there would be a cost.

Again, why does the interface keep getting messed

[Anonymous Coward]

It's gray text on a gray background, this is really annoying to read on. Please fix the site design! Why does the interface keep changing to this? It's terrible.

And then medium gold text on medium gray for the comments verification!!! This is rookie design garbage.

gnustep (cocoa)

[tarzeau]

they'd rather make a gnustep version on par with the google-chrome on osx (isn't that already based on cocoa)?

ofcourse..

[SuperDre]

It's logical that google wants to use Aura, they want about every technology used to be from their labs, and people bitched at MS for straying away from so called 'industry standards' but Google is much worse at that... And yes, they put it into opensource, but just look how they really keep control over the projects, if you deviate as a company you'll get slapped on the wrist (just look at android)..

Re:ofcourse..

[cyber-vandal]

People complain about Microsoft because they used closed proprietary non-standards software to cement a monopoly position that held back the industry for years. This is still causing problems now with the difficulty of ditching IE6/7/8 and XP and the fact that IE isn't compatible with their own software half the time. Google are nowhere near as bad.

Re:ofcourse..

[thejynxed]

I can attest to this, in that I've had pages on Microsoft websites fail to render properly or function properly in Internet Explorer, but worked just fine in Chrome, an older version of Opera, and Firefox. I about pissed myself laughing before letting them know about the issues.

Re:ofcourse..

[hairyfeet]

You DO know its possible, even easy, to pull a "EEE" on a project despite it being open source, yes? That being open source isn't a magical "get out of jail free" card, yes? If you haven't read it you might want to start here and please note that despite the fact there is a FOSS fork (ASOP) its quickly reaching the point where the FOSS part will be about as useful as the FOSS on a TiVo since none of the applications will actually run without major rewrites, which of course nobody is gonna do.

Re:ofcourse..

[cyber-vandal]

I know you don't like Google but comparing any of their behaviour to Microsoft's is just crazy.

Re:ofcourse..

[Anonymous Coward]

Because MS is the "evil empire" and Google is "don't be evil" blahblahblah

Google is trying to push single vendor "standards" like NaCL and Dart, fortunately everyone learned the hard way with Internet Explorer. Nothing to do with "evil", just big company technology politics.

Re:ofcourse..

[Lisias]

its quickly reaching the point where the FOSS part will be about as useful as the FOSS on a TiVo since none of the applications will actually run without major rewrites, which of course nobody is gonna do.

So you're bitching because Google is doing something that nobody else's gonna do?

The Google's problem is on another level (too much control over so many data on too few hands).

The source is there - get it, modify it, ????, profit. You don't want to modify it because you don't know how to do it, and don't wanna to spend money paying someone else to do it? No problem, really. But stop bitching about how Google is not helping you making money for free.

It's their servers you're using when you use Google Applications. You're free to setup your own servers and write your own APPs.

Why not build the Chrome UI as a web page?

[DrR0b]

Why do they need a GUI toolkit at all? Why don't they build the Chrome UI in HTML/JS/CSS?

Re:Why not build the Chrome UI as a web page?

[Chrisq]

Why do they need a GUI toolkit at all? Why don't they build the Chrome UI in HTML/JS/CSS?

I wish I had funny mod points!

Re:Why not build the Chrome UI as a web page?

[Shados]

Yeah.

Really funny

Re:Why not build the Chrome UI as a web page?

[BitZtream]

Most of it is going forward, if you read some of the links you'll see that very thing discussed, converting native widgets to WebUI where possible.

directfb-lite and other webkit ports

[lkcl]

i've worked with webkit a *lot*. for example, i helped denis with the port of webkit to directfb. in doing the python-webkit (direct) bindings http://www.gnu.org/software/py... i covered a *lot* of different ports of webkit. here's the summary:

* when compiling the standard webkit to run on a 400mhz ARM9, the gtk port started up in around... i think it was somewhere around 8 seconds. this was tolerable. it used about 130mb of memory to load a single basic page.

* when compiling the DirectFB port to run on the same system, it started up in about 3 to 4 seconds, and used about 1 or 2mb less memory. this was great!

* when compiling the Qt4 port to run on the same system, it took NINETY SECONDS to start up, consumed DOUBLE the amount of memory, and was basically completely intolerable.

the directfb port basically used an older (revision 1.2) version of the lite toolkit. to say it's light-weight would be an understatement: it's absolutely awesome. qt4 has unfortunately turned into a bit of a monster. gtk by comparison has remained reasonably level-headed, and when it (finally!) has the equivalence of COM's co-classes added to the gobject introspection layer gtk will become highly significant, strategically.

the only thing that the directfb-lite port lacked (at the time i was involved) was a window manager. this basically meant that you could only have one browser window open, and you had a callback for dealing with console alerts, which you had to then deal with yourself. i _thought_ about doing the same trick that mozilla does (which is most clearly demonstrated in b2g) - namely to implement the windowing system *in* webkit itself, in a high-level language: that would be cool. not many people are aware that firefox's menus including the toolbars and tabs are actually implemented *in javascript* (!), and the main browser "window" is merely a (secure) frame. b2g is an extension of that.

so anyway, the point is: there are lots of ways this can be achieved. you can implement the window manager externally and treat the browser as an isolated "component". you can go the other route and implement the window manager *using* the browser engine. but the main point is that either way, gtk and qt4 are to be honest complete overkill. it's only when you have things like co-classes built-in to the underlying infrastructure (like COM has) that you get any _real_ flexible benefit from the widget set, and as neither gtk nor qt4 have those, there's honestly really no point having them around.

Re:directfb-lite and other webkit ports

[Anonymous Coward]

Interesting read!

Re:directfb-lite and other webkit ports

[Anonymous Coward]

If you think Gtk is lightweight then you will love FLTK (stick to the v1.x series though, none of the v2/v3/+ crap).

Re:directfb-lite and other webkit ports

[scorp1us]

Which version of Qt did you use? There were a few releases that focused on load-time speedups.
Have you tried it against Qt5? It should be 99% identical, the only change I know you need is to add widgets to the QT line, as Widgets are no longer the default application environment.

Re:directfb-lite and other webkit ports

[Teunis]

Oh that explains a lot!
Thank you :)

Re:directfb-lite and other webkit ports

[lkcl]

Which version of Qt did you use? There were a few releases that focused on load-time speedups.
Have you tried it against Qt5? It should be 99% identical

it was qt4.3 or thereabouts. the problem is that qt does far too much. when you think that lite 1.2 is around an 86k binary and qt4 and qt5 are several tens of *megabytes* you start to understand the extent of the problem. libQtCore is 3mbytes. libQtGui is 11mbytes.

now bear in mind that when you're doing something like a web browser, all you really need is a font and pixel drawing system (cairo, pango), an input box (liblite), a way to read the keyboard and mouse, and err... it really ain't that complicated, then you start to understand why GTK and QT are complete overkill. only when you need to do things like open a new popup window or open a new browser window that you need something more complex, and heck, those can be done with a bit of X11 or Win32/GDI message handling for goodness sake. in cases where you're doing direct framebuffer writes (such as in chrome os, android, b2g, DirectFB applications and more) then you don't even need _that_, in many cases.

so in effect it doesn't matter how good Qt4, Qt5 or GTK2 or GTK3 are, the fact remains that even the initialisation of the sub-systems that aren't going to be used are all simply too much. the only reason for maintaining those ports (of webkit) is to make it easy for people who wish to integrate webkit into *their* applications that are written in those frameworks.

so the difference is: under the circumstances where you don't need the infrastructure of those frameworks, because you're doing a stand-alone web browser, these heavy-weight frameworks like Qt and Gtk are an exceptionally bad idea.

Re:directfb-lite and other webkit ports

[lkcl]

ls -altr /usr/local/lib/*lite*
lrwxrwxrwx 1 root staff 16 Dec 7 2010 /usr/local/lib/liblite.so.3 -> liblite.so.3.0.5
lrwxrwxrwx 1 root staff 16 Dec 7 2010 /usr/local/lib/liblite.so -> liblite.so.3.0.5
-rwxr-xr-x 1 root staff 928 Dec 7 2010 /usr/local/lib/liblite.la
-rwxr-xr-x 1 root staff 48848 May 3 2011 /usr/local/lib/liblite.so.3.0.5

i'm sorry - that's 48k not 86k!! liblite is *tiny*.

Re:directfb-lite and other webkit ports

[dkf]

Which version of Qt did you use? There were a few releases that focused on load-time speedups.

Compared with the startup time reported (90s!) even a 10x speedup would leave it being SLOW.

Actually, it sounds like something is critically misconfigured; I can't imagine any situation where a 90s startup is acceptable. Well, not in the last 20 years. (I remember using LispWorks when I started waaay back, and that had that sort of startup time. I dropped that commercial system for a lightweight open source solution and never looked back.)

Re:directfb-lite and other webkit ports

[Anonymous Coward]

True, but I somehow doubt that creating Aura is less expensive and "better" than Google just helping Qt out.

Like most of their tech lately this is just a case of "not invented here". NaCl, WebP, SPDY, PPAPI, Aura... the list goes on. If Google hasn't made it, they don't want to work with it. They want the rest of the world to adopt their tech. That puts them in foremost control AND ahead of the competition, while still making them benevolent.

I couldn't care less if Chrome starts up in 8 seconds or 4. I just don't want to live in a web world where Chrome's the only option because nobody else could keep up with their pace and gone the way of Opera. Especially if all they've effectively done is replace existing tech with their own versions because NIH.

Re:directfb-lite and other webkit ports

[Arker]

Pardon what may be a stupid question, but does the FB port need X-Windows at all?

If not just run multiple browser windows using multiple virtual consoles, Alt-F1 Alt-F2 etc switches back and forth. You could have chrome in one terminal, bitchX in the next, editor of choice in the third, and so on.

Re:directfb-lite and other webkit ports

[scorp1us]

Your criticisms are accurate, and they've worked to address them. A significant reason for the '5' in the Qt versioning is they granular-ized the libraries, which is why you need to specify widgets if you want widgets.

Yes, 4.3 was a slow loader. I believe the speed ups came around 4.6.

One of the new Qt5 features is the surfaces are OpenCL/GL. So drawing is not sucking down the CPU. It doesn't do much for loading, but its nice not having to push pixels through a CPU.

Re:directfb-lite and other webkit ports

[scorp1us]

If he's using direct-fb he's doing something special. I assumed it was some embedded device.

But Qt has adjusted to those criticisms since he he last tried it. No idea what it would be now.

If Linus would just endorse a toolkit...

[tlambert]

AFAIKT Aura is a more than just a UI Toolkit, it's a complete Window Manager. A replacement for Gnome (wow! I hope that takes off!) Apparently it's been running on the Chromebooks. Here is Linus' take on the topic.

If Linus would just endorse a toolkit, then there would be One True Toolkit; this would be the most likely thing to drive an actual "Linux desktop revolution". I am not holding my breath.

Re: If Linus would just endorse a toolkit...

[Anonymous Coward]

Linus is an expert at kernels. I wouldn't pay much notice over his opinions on UI toolkits. He'd probably be against anything C++.

Re:If Linus would just endorse a toolkit...

[Simon Brooke]

If Linus would just endorse a toolkit, then there would be One True Toolkit; this would be the most likely thing to drive an actual "Linux desktop revolution". I am not holding my breath.

And that's why he won't. The whole point is to avoid homogeneity, because homogeneity strangles progress and provides a single target for the spread of malware.

Re:If Linus would just endorse a toolkit...

[tlambert]

If Linus would just endorse a toolkit, then there would be One True Toolkit; this would be the most likely thing to drive an actual "Linux desktop revolution". I am not holding my breath.

And that's why he won't. The whole point is to avoid homogeneity, because homogeneity strangles progress and provides a single target for the spread of malware.

I was more concerned with a single target for the spread of commercial applications. Malware is only really a big issue once you make yourself a target by having a large user base. Being a target of malware would be a *good* thing for Linux, since it would indicate a higher market share.

Re: If Linus would just endorse a toolkit...

[tlambert]

Linus is an expert at kernels. I wouldn't pay much notice over his opinions on UI toolkits. He'd probably be against anything C++.

It doesn't matter if *you* personnaly pay attention, it matters whether he's respected enough to play 800 pound gorilla. Also, it doesn't really matter who "wins" as the selected GUI, what matters is that *one* does.

Re:If Linus would just endorse a toolkit...

[Kremmy]

I think the Windows user is spoiled by the fact that he's merely a client in a server reality, he'll learn what it is to be a target when he's hosting the services that everyone is pouring their data into. Browser and e-mail hijacking is kind of the lowest hanging fruit in the technological ecosystem.

Use Elementary/EFL.

[Anonymous Coward]

Using Elementary/EFL might be even better for the resource usage overall...

GTK

[Anonymous Coward]

It's everything that's wrong about FOSS.

GTK+ is for GIMP!!!

[Anonymous Coward]

Every use of GIMP is a problem.
FTFY

whoooshhhes

[Chrisq]

Why do they need a GUI toolkit at all? Why don't they build the Chrome UI in HTML/JS/CSS?

I wish I had funny mod points!

Yeah.

Really funny

Of all the whoooshhhes I've seen ... this is the biggest.

C++

[Anonymous Coward]

If by "anything C++" you mean "anything that forces everyone else to use C++", he'd have very good reasons to be against it.

As far as I know, QT does this. Which is one of the reasons that I (as a C developer) don't use it.

Re:C++

[shutdown -p now]

If by "anything C++" you mean "anything that forces everyone else to use C++", he'd have very good reasons to be against it.

Qt has bindings to half a dozen different languages, from C# to Python.

FWIW, it's fairly easy to take a C++ library and wrap it in C.

Google should take note ...

[Rambo Tribble]

... there's a fine line between going your own way and jumping the shark.

In addition: Re:Just for a browser?

[Anonymous Coward]

To add to the parent's post:

Stroustrup created C++ to be a system's programming language with object oriented features. He succeeded. If you are writing system's level code, business logic code, middle tier code, numerical processing, or basically anything that is engine/under-the-hood then C++ is an acceptable choice of language. It gives you some constructs to be object-oriented to an extent while allowing the ability to achieve needed performance. It may not be the best with respect to RAD style development, but in the right hands it's a good balance.

GUI programming is a different animal altogether. Alan Kay, Dan Ingalls, and crew, designed Smalltalk around the GUI technology that they innovating (or said differently, Smalltalk was designed to help them build the GUI technology that they were innovating). You may think your favorite dynamic programming language innovated some cool little feature but the fact of the matter is that most programming languages today have borrowed from Smalltalk.

Stroustrup cared about systems level programming and was very influenced by Simula. Smalltalk doesn't show itself in C++, which is why some features in C++ are lacking (like RTTI), which is why GUI programming in C++ has always been painful. Don't believe me? Ask yourself the following:

1) Why did OWL/MFC (Borland/Microsoft) invent message maps?
2) Why did C++ Builder (Borland) introduce new keywords into the C++ language for properties and events?
3) Why did Qt invent the signal/slot mechanism?

To me the answer is pretty clear. Plain and simple, C++ is crap for GUI programming. It needs features to allow certain constructs to be simple and elegant. It needs features that Smalltalk, Delphi, and C# have been enjoying for years.

People who argue differently about GUI development and C++ A) doesn't know what they are talking about, or B) has never done serious work in another technology where GUI development is a rather enjoyable experience. Anyone who plays the "but Boost has this template class that solves everything" card definitely falls into category A.

obligatory xkcd

[Anonymous Coward]

http://xkcd.com/927/

Umm what?

[Anonymous Coward]

Qt apps most definitely do not need to be "web apps". What have you been smoking?

About QString, use QString.toStdString and QString.fromStdString - yeah, that's REALLY difficult to integrate... I cannot fathom the complexity... If you still can't figure it out, try asking in Stackoverflow.com, I'm sure someone will give you a helping hand.

The vendor isn't dead either, and the library isn't proprietary.

Is this some heavy case of Not Invented Here?

Qt is, judging with objective technical merits, simply splendid.

Re:Umm what?

[spitzak]

No you cannot use toStdString and fromStdString as they do not use UTF-8 but attempt to use the "locale". Also even if you set it to UTF-8 the throw exceptions on invalid encodings, making them impossible to use safely (since the exceptions are rare enough that code is not written to handle them).

Hilarious

[Anonymous Coward]

Mozilla already did this over 9 years ago and are still dealing with the fallout. Google sure has big balls to think they'll be able to just swoop in and not lose a lot of mindshare. Sometimes I think they're trying to drive people away from Chrome by delivering the most watered-down product where they control everything.

Au contrare, mon frere.

[emil]

The "free software community" had quite a bit to say on gtk vs qt.

The design of KDE was based on a fundamental mistake: use of the Qt library, which at the time was non-free software. Despite the good intentions of the KDE developers, and despite the fact that the code of KDE itself was free software, KDE could never be part of a completely free operating system as long as it needed a non-free program to function.

obviously not

[Anonymous Coward]

QT is owned by Nokia which still can't make up its mind on whether it is a Google partner or competitor or just plain bankrupt.

Soon: 15 competing standards?

[IwantToKeepAnon]

https://xkcd.com/927/

qt is not an option...

[fastflo]

i had a long'ish text here trying to explain my point. i decided not to post it because it would get way too long to explain everything. just some simple keywords:
gtk: opensource!, c-api, glade (did overcome code generation), xml-ui spec files, gobject / glib, gtkmm c++ wrapper, themes//theme-engines, event-loops, very nice handwritten gtk2 pygtk wrapper, gtk3, gobject INTROSPECTION, pango&cario integration
qt: trolltech, nokia, microsoft..., e, CODE GENERATION SIGNAL SLOT CRAP!, trying to make money!, no c-api !?, a bloated API of operating system abstractions --- in a gui toolkit? really? all those socket, filesystem .... crap, it has qmake? don't we have the autotools? and who is too stupid to write a 4-lines makefile to compile a simple 1 or 2 files gui app?

all about money.

[fastflo]

sky: microsoft.
vlc: there is mplayer.
teamspeak: proprietary,
origin: proprietary

i don't care about industry. they have money, they can hire and pay as many developers as they want/need.
once in a time people had the idea to be able to make use out of computers without needing those big companies to support them.
to me it seems qt is only taking care of how to best adjust to the needs of even those companies!

Re:all about money.

[hobarrera]

vlc isn't backed by any company.

And most large software IS backed by companies. Care to mention some big GUI proyect not-backed by some company OTHER than firefox? Because I can't even think of one. KDE uses qt.

compatible with windows?

[fastflo]

who wants to be compatible with windows? why even trying!?! we all should try to do better!

Most uninformed post in this thread award?

[Anonymous Coward]

>Qt wants all apps to be web apps
What the hell are you talking about? Qt renders either native-appearance widgets and forms with QtWidgets or an OpenGL driven UI specified through QtQuick. If you want to draw stuff in HTML, you use WebKit. What does that have to do with Qt?
-
>web browsers don't need that or the overhead it brings
All UI toolkits have overhead. Qt is modular enough that it'll only pull in what it needs for a simple UI (not a lot)
-
>Qt also pointlessly reinvented lots of the C++ standard
Because at the time, the Qt devs needed to get around a lack of language features and a lack of cross platform compiler support. It's not 'pointless' when it served an important purpose. Even now Qt5 can be built and runs fine on non-C++11 compilers. It even builds and deploys to embedded systems which often use crappy STL implementations. As Qt5 evolves, the legacy Qt replacement functions are being phased out. qSort no longer exists, etc.
-
>QString and all their container classes
QString is UTF-16. Whatever your stance on UTF-8 v UTF-16, at least its strongly defined in Qt. In the STL, string is a just a char container with no encoding information.
-
>People who use Qt are mostly allowing themselves to be locked in to a dead vendor's proprietary library
1. Most of Qt is open source and available under the LGPL.
2. "Digia" isn't dead (why would you think this?), and actively update the library. They offer commercial licensing and support for people who aren't okay with LGPL or GPL licensing.
3. There's a KDE clause that comes into effect with Qt if a company that owns the Qt codebase decides to make it proprietary which will force the release of most of the codebase under a BSD license.
-
Why do people insist on spreading FUD when they have no idea what they're talking about?

Above is incorrect in this situation

[dbIII]

DLL hell is where different library versions can not co-exist to save you so is something different to the mere dependency problem mentioned above. The gnome situation is effectively DLL hell with background applications since the new ones have the same names (with no version number to let them co-exist, you have one or the other and not both) as the old ones but operate differently.
That's why having multiple versions of files does not solve the problem with the new gnome - you need to muck about with chroot, virtual machines or running the stuff remotely from another host if you want to have new gnome and old gnome applications on your screen at the same time.
That's why there was so much anger about it a year or two back.


So readers, I suggest you read and understand posts and the context they come from before getting condescending and then mentioning the "real world" after reading two words and going off on an irrelevant tangent. It's very insulting.

aftleittle2580

[brenda4302]

However, the desire for exclusivity unavoidably tempts imitators. In fact many of Louis vuitton Outlet Online Monogrammed Canvas bags nike hyperdunk 2012 low offered either internet or offline are eliminate. In some cases it ares challenging to find a well-made duplicate, not to discuss the genuine version.Having another drive on the same cable can significantly decrease your primary hard disk performance. Use the latest DirectX version as this will improve gaming performance. Needless to say also, the best video card you can get, the best would be your 3D gaming performance.2006 WK6--DEN 13 OAK 3 (OAK +15) W 2006 WK6--TB 14 CIN 13 (TB +4. 2005 WK8--HOU 19 CLE 16 (HOU -2) W ''Look, go to the Universal Studios lot near North Hollywood, California. They've just built a new ride there called Back to the Future. It cost $60 million.Think I am crazy? JUST TRY ITMay 6, 2011 at 12:21 |I said I had bra scars and rather dry elbows! For me, this was my chance to mark a milestone in my life, like my tattoos, everytime I view these pics they would take me back to this day, a couple of weeks before my wedding to the most awesome man, (sorry other men!) a time in my life when I am extremely happy being who I am, in a place surrounded by my friends and family who also accept me for me.No, I usually just look like the local, mental asshole. Honestly I could go on for to long about anger; don't worry if your interested one day you can read my book.What's important, is I'm going to change my inner motivation with innovation. In late 1999/very early 2000, the very last original remaining MBTA-owned FP10 contraptions were being being presented along with still have handled when Maryland, Spanking New Orleans, Georgia, and thus Idaho about various sorts of holidaymaker train engines.Wikimedia Commons comes armed with movie associated with: EMD FP9 enginesPinkepank, Jerry The New.Virtually every customer we have that uses our accounting or payroll system use only less than half of the features. Every street corner we see cars that can deliver twice or thrice the speed than the owner is really legally allowed to go, or will ever go.How many of us get Ipods that can store 600 hours of music, when all we have in our collection Nike KD 5 is just less than 100 hours? Just before you reach your imaginary seat, stand back up. Do this for 30 seconds or for a full minute.If you are more advanced in your fitness, add a jump at the top of your squat and be sure to land with soft knees.FINL'S rise in sales was largely driven by an increase in store traffic, additional days of sale and a rise in comparable-store sales. The company reported an EPS of $1.59 in FY2012, an increase of around 26% from the previous year.Ain that a stinker?" Now, what if this happened every six months or so?Such is the landscape of search engine optimization, where the search engines are constantly changing the game, and the advice you find online, dating all the way back to 2005, may no longer be relevant. But even without an official release date, the http://www.nba2014seasons.com/... shoe seems to have leaked out into the marketplace. Ostensibly authentic versions are posted on EBay with a $400 price tag -- more than double the expected retail price of $175, and employees at one local boutique -- Flight Club Los Angeles on Fairfax Avenue -- claim to have already received (and sold) the store's entire allotment of the 2009 reissued Space Jams for $350 to $450 a pair.So what goes into making the clump of patent leather, plastic and fabric known as "Style Code:136046-041" so hotly The Football Association Challenge Cup, commonly known as the FA Cup, was founded in 1871 and is the oldest football knockout competition in the world. It includes the English Football League and Clubs as well as the six Welsh clubs. The first competition, season 1871-1872, had fifteen entries, more than 75

Field of use restrictions are not FOSS

[marxmarv]

"You must pay us upfront before you even start if you even think you might commercialize your software" is the acme of desperate opportunism (aka shark-jumping) and of an inflated sense of self-importance, and probably entirely unenforceable as a matter of copyright.

A BSD-licensed toolkit of usable quality would, indeed, correct Qt's attitude problem.

As to quality, have you ever built and ran gimp-1 prereleases? "Interesting" times....

Just sent you an email man... apk

[Anonymous Coward]

One you will REALLY like (firewall rules table vs. monitoring HERE and other spots online)...

APK

P.S.=> Reply to me in email - I truly *think* (heck, I KNOW) you'll LOVE this one... apk

Google is Evil, Closing Markets.

[bbsalem]

None of this surprises me. It seems that every effort by Google is to create a captive market, even one that looks like it is using open standards, but really ins't. Google Docs comes to mind. In fact I think that Google Docs is managed by Microsoft or by Microsoft alumni around Office 2000 for market capture with a dumbed down legacy platform that only partly supports decade old standards. This is intentional, so it is no surprise that Google would try to roll it own GUI standard, even one that is based on X11 or QT.