Guest post by “I’m happy to announce that the Qt Project officially went live today. Starting today, development of Qt will be governed as a true open source project. We now have qt-project.org – a website where all development of Qt will be centered, providing the same infrastructure and processes for everybody that wants to contribute to Qt.”
I could have sworn the project was always open source.
Has been open source for quite a while now. It is now open governance as well.
Open governance means that a meritocracy, a consensus amongst developers, decides directions for the project, not just Nokia any longer.
It was an open source project with commercial backup, with all good and bad things such deal brings us.
Frankly speaking I’m worried that with more freedom and less resources Qt will enter a permanent redesign cycle. Let’s wait and see how it works out – this is one of the biggest challenges for the OS community so far and, who knows, it may become one of its greatest achievements.
KDE Applauds Qt’s Move to Open Governance
http://dot.kde.org/2011/10/21/kde-applauds-qts-move-open-governance
“We fully support the work being done with the Qt Project. An openly governed Qt is in the best interests of all Qt developers. The Open Governance structure of the Qt Project empowers developers to influence the direction and the pace of Qt development. Stakeholders in the future of Qt, such as KDE, can now contribute according to their own priorities and take ownership over areas of Qt that are of particular importance to them.”
Edited 2011-10-25 00:06 UTC
Sorry to say, but there is quite a quality gap between Qt and KDE.
I am not sure if I would want to see the KDE community taking Qt their direction.
It needs to do some fitness before it gets really fat
I’ve been playing with the QtCreator recently. It’s probably the best open source IDE for GUI applications I’ve seen so far. Here’s hoping this new move for Qt continues to deliver such high quality tools.
I agree with you.
Qt Creator is slick, slim, fast and very objective IDE, without the bloat.
have just been invalidated
I’ve been hearing the same thing for the last 5 years, and GTK is still alive and widely used.
Yea man, this is hurting me too.
Seriously, GTK+ used to be the best full LGPL toolkit around in late 90ties, but still not enough to push Linux desktop, esp. with meagre developer experience, slow performance, excessive, unnecessary complexity. It was better that Motif, but well, what wasn’t?
Now it’s just a legacy burden, and (imo) the single most important reason for failure Nokia mobile linux efforts.
The funny thing here, is that Nokia Linux efforts went down the drain when they bought Qt and pushed it on MeeGo, ditching GTK+.
Nokia sold many more GTK+ Maemo based devices than Qt MeeGo based ones.
And for the “last reason for keeping GTK around” I think you forget Gnome is still alive and the default desktop on most Linux distros. There are few KDE based distros and, honestly, I have to see one enterprise Linux deployment based on KDE yet.
GTK+ is here to stay and in the technically is as good as Qt4, in fact it uses less memory than Qt4 in my experience. The only thing I see it’s better in Qt4 is the IDE – Qt creator – that the GTK+ team does not ships.
What it seems from your words about “excessive, unnecessary complexity” and “meagre developer experience” is you are not a GTK+ developer and know little about it, so it seems complex and difficult to you. May be you are not proficient in C either.
If you had even a little experience with GTK+ and C in no way you would see it complex or difficult.
I actively work with both platforms, both are really great. I have ported GTK+ apps to Windows as easily as Qt4 ones.
GTK+ is very very good, Qt4 is great and I am very happy of both being around.
If it hurts to you that GTK+ is alive … I am sorry for you, because it is going to itch for a long long time.
That’s the point, they followed the community and then had to turn the ship when it was too late, bc GTK+ turned to be no go from the commercial/independent APP dev POV. It may be acceptable from seasoned UNIX dev who was C savvy from the primary school but not so for remaining 99% of developers.
The quality of (not numerous) apps created with GTK+ is more the testament of their developer than the toolkit itselt. The uptake of GTK+ compared to QT in the independent sw. development (outside of core linux desktop community) speaks for itself.
And don’t get me started start with complexity. Ever saw a gtk component, drowned in boilerplate macros and stuff.
Even old Windows COM headers looked elegant next to that.
The only upside GTK+ ever had compared to QT (besides the licence) was a better ABI stability. The reason for their engagement with GTK+ has just been finally invalidated, but as Linux desktop is no longer fancy , they will probably maintain whatever is there.
Who ever could accept the idea to base most important app framework on “emulate OO in C” skunkworks?
Edited 2011-10-28 09:53 UTC
And yet, new GTK+ projects get started; quite young ones get successful. The most trending up options of desktop environments, which give a practical alternative to GNOME, seem to be fast & lean XFCE or LXDE (particularly the second one: rather young and curiously solid & dependable already; going quite mainstream, as far as one can use this term with Linux; while not a totally reliable indicator, the Distrowatch ranking position held – for some time now – by Lubuntu is nothing to be scoffed at)
Being such practical (you know, seen as most suitable, most chosen in practice) alternative is something with which Qt-focused KDE struggles. And where KDE was chosen (the choices made mostly during 3.x era?…) for mass deployments, it seemed to be more about “alternative to Windows” – in the sense that KDE was then outwardly much more similar to it (that was actually a justification at my non-IT/CS/etc. uni department few years back, for some number of “browsing & OOo boxes for layman”) – but LXDE seems to be cornering this one now… (bonuses: more suitable for postponing hardware obsolescence, and gets less in the way)
Why don’t you accuse the Qt desktop of “not enough to push Linux desktop”? KDE was even earlier on the stage… (yes, licensing issues in its early years – but shouldn’t they barely impact the quality of work done, giving “ready” product – and a nice platform to jump onto, for previously distanced devs – once licensing issues were mostly resolved? And now, as one comment above points out, KDE wishes to – shudder – party take over the direction of Qt; BTW, all this written by me, somebody who for quite a few years was pushing KDE, mostly on the “niceness” of its foundations …but that’s not everything / enough was enough)
Perhaps GTK+ gives ultimately better results in, supposedly, “developer satisfaction driven” OSS world (with Qt maybe better suited for commercial style of development).
Perhaps it’s a case of… what clearly strong advocates of OSS toolkits (Qt too, in a story essentially about it) also tend to claim here (for example http://www.osnews.com/thread?493982 ) – that more hand-holding tools ultimately attract worse coders? (even somewhat more strict entrance filters probably visibly lessen this problem)
(heck, I know I’m quite comfortable thinking about PyQT or PySide, but for some reason I mostly wouldn’t dare to touch serious GTK+ dev; and I can’t be described even as a code monkey)
Right beside, you essentially write it yourself:
Well, for core / big / important projects, those which drive whole thing forward, I very much prefer this 1% of devs to be on board, instead of people like me.
(but I’m not sure what “not numerous […] uptake” is supposed to mean – it’s not like there isn’t enough of them, or did you fall into recent appstores-driven mindsets of ~”the more shitty fillers, the better”?)
But anyway, OTOH, how is supposedly Qt-preferring OVI Store going along? And before you claim it’s a case of “turn the ship when it was too late” – remember that Qt is, for over a year, the officially promoted way of development also on Symbian (with the compatibility going back to S60v3 FP1, I believe – so 5 years, with most likely the biggest installed base worldwide); and available for 1.5 or 2 years.
How many of the top OVI Store apps it spawned?
Edited 2011-10-29 12:21 UTC