Qt - LGPL?

Posted by George Wright Wed, 14 Jan 2009 09:02:00 GMT

Looks like Nokia are planning to release Qt under the LGPL licence. I wonder what impact this will have on the free software world?

Posted in  | 47 comments

Comments

Leave a response

  1. Troll? said about 2 hours later:

    All reasons for starting/choosing/developing GTK (apps) will be gone in the not too distant future :P

  2. Cypher said about 2 hours later:

    Hope this means the end of GTK ! GNOME can now switch to Qt.

  3. z said about 2 hours later:

    The death of the worthless GTK toolkit is coming. Get ready.

  4. Artem Vakhitov said about 2 hours later:

    Gnome won’t ever switch to Qt, it’s just infeasible. OTOH, a lighter Qt-based DE similar to LXDE, but more powerful, would be much welcome.

  5. Cypher said about 2 hours later:

    It is not infeasible, except if you do not plan it properly. Of course it will take time, one bit at a time, but it is feasible for GNOME 3.0. KDE4 has been rewritten from scratch, GNOME can be too. It’ll just be a mixed system in the beginning, with both Qt and GTK.

  6. Markus said about 2 hours later:

    GNOME probably won’t switch, but GNOME could allow GNOME apps to be written in Qt. Apps following the GNOME HIG are perfectly possible with Qt GtkStyle.

  7. Jamie said about 2 hours later:

    Now is the perfect time for GNOME to switch toolkit. It would finally lessen the eternal feud that has been going on on the Free Desktop.

    A GNOME desktop with GNOME HIG and GNOME look should be possible with Qt. It would save the community the work of maintaining Gtk+, gtkmm, remove all the compatibility discussions and allow developers to focus on creating good applications.

    For us in the GNOME community this is great opportunity. Just ask yourself, what would you do if such a complete toolkit became available under the LGPL and was not called Qt? We’d jump at the opportunity to make GNOME 3 rock.

  8. Jef said about 2 hours later:

    Well, gnomes don’t like C++ (look at how many gtkmm applications in gnome) so I don’t think swithing to Qt will be a popular idea.

  9. Troll2? said about 3 hours later:

    If and when Qt goes LGPL, Gtk will have no reason to continue to exist, and Gnome and Kde can finally merge, ending the duplication of effort on Gnome and Kde caps.

    Qt is far nicer toolkit to work with. Gtk should now retire.

  10. C++ is no longer a big issue said about 3 hours later:

    GCC used to have issues with c++, but that is mostly history now.

    LGPL Qt would be great for Gnome3. It will push the free / open source / gnu / unix / linux desktop much further.

    The most important in consolidation and standardization. Qt has never looked this promising.

    This is my first positive comment on this issue since Gnome was started.

  11. SW said about 3 hours later:

    “For us in the GNOME community this is great opportunity. Just ask yourself, what would you do if such a complete toolkit became available under the LGPL and was not called Qt?”

    I’d say “oh look, another toolkit like the dozens of other toolkits out there”, and keep on doing what I was doing. Life’s too short to keep rewriting code.

  12. Cynic said about 3 hours later:

    In a perfect world the plans for Gnome 3.0 would change now .. but I don’t think that will happen :(

    People are not realistic .. it does not matter how much better Qt 4.5 will be in EVERY respect compared to GTK.

    Makes me sad.

  13. Zeesahn Ali said about 3 hours later:

    http://mail.gnome.org/archives/gnome-devel-list/2008-July/msg00011.html

  14. Simon said about 3 hours later:

    I think Qt is great, and certainly this license change will get rid of some people’s objections to using it.

    But for all the people saying GNOME should use it instead of GTK, the licensing issue was not the only problem. Some people prefer C to C++ and might therefor prefer a C toolkit.

    And whether you agree with that or not, surely it would be better for the free software community to have two free, high quality toolkits.

  15. Cypher said about 4 hours later:

    –

    And whether you agree with that or not, surely it would be better for the free software community to have two free, high quality toolkits.

    I do not agree, when we are talking about a toolkit, which is very low-level in the development stack, I think we should use the same common base.

    Regarding the use of C instead of C++, you can always develop bindings. There are a lot of bindings for Qt, for Python, Java, Mono,… And the less reluctant might even start using C++ instead of C, and discover that it is easier to use for high-level applications.

  16. oliver said about 4 hours later:

    Please don’t make Gnome3 a rewrite similar to KDE4… From my experience, the KDE4 desktop has lots of bugs so far and not much promising stuff, and frankly I think that way too much time has been spent with too little return. Prove me wrong; but KDE4 in Kubuntu Intrepid did not really make hope IMO.

    Secondly, I don’t think Gnome and KDE will merge in near- or mid-future… There are fundamental differences in the ideas of the two desktops, and both ideas have a a place and IMO are not mergeable without loosing the advantages of both sides.

    Thirdly, it seems that so far KDE has made a rewrite each time Trolltech has made a new major release (as I said above, in the case of KDE4 this appears to have been very expensive with low return). If Gnome switches to Qt, are Gnoem devs forced to do such rewrites as well whenever Trolltech/Nokia think it’s time for a new major release?

  17. myie said about 4 hours later:

    I think C is far too hard and lame as the ground language for a DE.

    GNOME should rather move to D than Qt.

  18. Anon said about 4 hours later:

    @myie:

    D is a programming language, not a toolkit.

  19. MyOP said about 4 hours later:

    “Prove me wrong; but KDE4 in Kubuntu Intrepid did not really make hope IMO.”

    Found your problem, you used crappy *buntu what is totally made for Gnome. You should try a distribution what cares about KDE people and does not abandon the Gnome users either with that action.

  20. Artem Vakhitov said about 5 hours later:

    The reasons why rewriting Gnome in Qt is infeasible are not technical but social. It’s just that nobody will undertake such a huge task for no clear gain. It’s just far easier to correct the present deficiencies (which are many) in the existing code, if you suddenly have much free time. KDE 4.2 may be cool but honestly, what REAL advantages does it have over KDE 3.5?

  21. Cynic said about 5 hours later:

    @Artem:

    No clear gain??? ” What are the possible advantages?

    * Less confusion for the non-insiders on what to learn and program for
* We can get rid of the whole Freedesktop common widgets talk (e.g. Print and File dialogs) which is nonsense and a distraction anyway
* Hopefully less bickering in the community meaning a more unified and focused front against larger threats
* Focus can move from the lower toolkit layers to the higher level desktop layer which really define the value differences between GNOME and KDE
* Easier technology sharing

    “

    Add to that all the nice features Qt has that GTK is many YEARS short off.

  22. Artem Vakhitov said about 5 hours later:

    @Cynic: You listed 5 points, and not a single one of them really means anything for users (it is for them, after all, that Gnome and KDE exist.) Meanwhile, the hypothetical gain for developers is vastly offset by the effort that would have to be undertaken to even bring Gnome where it is today.

    KDE 4 offers nothing compared to KDE 3.5 that would justify such a massive rewrite. I have nothing against developers constantly rewriting their apps if they so desire, it’s just that it really brings little to users.

  23. oliver said about 5 hours later:

    @myie: yes, D as “blessed” language for building GTK apps would be cool… I begin to understand that people want Mono really just for technical reasons - apart from C (which many people don’t quite like, me included) and Python (which has its own shortcomings, like missing type safety) there just is no well-supported GTK language. Maybe Vala will become one, but who would want to learn a new language just for GTK? I’d much rather learn a language that will also be useful for non-GTK apps, but I don’t see Vala getting even nearly as much traction as D has acquired over the years.

  24. Cypher said about 5 hours later:

    @Artem

    Nothing for users ? Semantic desktop ? Task-oriented ? More beautiful ? Easier to use ? Maybe YOU do not use the new functionalities, but as a user, I do… a lot !

    KDE 4.2 is on par with 3.5 function-wise. The real development of the new desktop paradigm is about to start now.

  25. Micro said about 5 hours later:

    Time to retire Gtk! So long!

  26. kiberlynx said about 6 hours later:

    Diversity is very important for the resilience and survivability of every ecosystem,

    so altough QT allows for way cooler and more efficient software than GTK+, it is good for everybody’s future that there still exists an alternative to it,

    because you never now what the future holds. So you better have free alternatives.

    Thats my take on it.

  27. Alexis said about 6 hours later:

    @Artem Vakhitov KDE 4 offers nothing compared to KDE 3.5 that would justify such a massive rewrite. I have nothing against developers constantly rewriting their apps if they so desire, it’s just that it really brings little to users.

    disagree, you have many new features but not much of them are not yet fully visible. You should run KDE. I just take one example, the task bar in plasma that is much more powerfull than before, you can have several taks bar and much more you are completly free to set the content. I answer to the question you will ask, yes it is usefull see gnome have sometimes two and you will be able to find screenshot on internet of pretty cool usage. This feature was not in in 3.5. This is one example, just one. There are many more. And for this example KDE needed a major rewrite of the desktop shell. The old was crap. Other point the class behind Plasma is QGraphicsView, not in Qt 3.x. KDE needed Qt 4.x for that, so rewrite of KDE because it was needed.

  28. jay said about 7 hours later:

    This has serious implications. Nothing more needs to be said.

  29. Bruce Cowan said about 9 hours later:

    Firstly, for GNOME to use Qt, it would require the abandonment of everything they have done so far. GNOME 3.0 is supposed to be evolutionary, so this would be impossible unless they continue using GTK+. Someone here pointed out that GNOME have control of GTK+ themselves, something that KDE doesn’t have over Qt.

    Secondly, Qt would mean a change of language. C has been used for years by GNOME, so suddenly changing everything to C++ would be unpopular.

    Thirdly, it’s not nice to see many people here saying “GTK+’s crap” without having anything to constructive to say.

    GTK+’s here to stay, get used to it. You don’t have to like it, but don’t make a fuss if others do.

  30. Daniel Williams said about 9 hours later:

    All this talk of “kill this [DE/toolkit]” was ridiculous both before and after Qt’s relicensing. IMO, the only way either one of them is going to really fall in favor as the de facto standard Linux interface is if one distro really takes off in usage as a preloaded OS, and that DE happens to be its default. Until then, I think resources would be better spent on continuing to develop and standardize both DEs than pointlessly bicker about which one should exist or use which toolkit. Chances are very low you’re going to change any minds here.

  31. ulf said about 10 hours later:

    GTK’s death. Finally.

  32. Roman said about 10 hours later:

    Yes! This is time to show qt’s misrable in the real appliance, not about license

  33. Cypher said about 10 hours later:

    @Roman

    You’ve never used Qt, have you ? Because you wouldn’t say that if you had. Qt is superior in every aspect to GTK+. That’s a fact that even experienced GTK+ programmers would agree on.

  34. Daniele said about 10 hours later:

    I just think that qt have a single advantage over gtk, it has a better learning curve than gtk for new developers.

    I love gnome HIG, and i will love a DE using qt as toolkit and gnome HIG.

    If after this move such a desktop will exist, i’ll say thanks to nokia.

    these are my 2 cents.

  35. Craig said about 11 hours later:

    Various articles I’ve read in an attempt to educate myself have seemed to suggest that GTK+ has focused so much on stability that it is hobbled by old code and lacks facilities to easily skin new apps and easily create new interaction models.

    Clutter and other suck efforts are advancing the world, but I am so surprised a how far behind GTK+ is. What - GTK3 is over a year away???

    Windows, Mac, and QT-based advancements that are maturing don’t seem to have been enough to spur developers to drive forward with a solid new toolkit as they rather continue to squabble about why their solution is superior.

    I’m not saying there are not good things here (I’ve heard great things about Cairo, Pango, etc.), it’s just the attitudes, lack of solid road map, and apparent lack of motivation to move forward don’t make me confident that this is the basis for any future platform.

    This isn’t a flame, just my sadness over a great project stalling.

  36. Noah said about 12 hours later:

    If this had happened a few years ago, it might have killed the Gnome project, which was largely driven by licensing concerns with Qt.

    Now Gnome has gained enough momentum that predictions of it’s death are quite premature. If Gtk is still not as nice as Qt, it is quite usable, and there is not a strong motivation to port all the current Gtk apps (Firefox, OpenOffice, the Gimp).

    Where this news really makes an impact is in the embedded space. By doing this, Nokia is answering the competition from Google’s Android. It will be interesting to see if Qtopia grabs a larger share of the handset market.

  37. Tim said about 13 hours later:

    CDE had the same problem by not moving forward and/or adopting to changes and got replaced. I fear if we don’t try to move forward here then the same will happen to GNOME in the long run.

  38. Cypher said about 14 hours later:

    –

    If Gtk is still not as nice as Qt, it is quite usable, and there is not a strong motivation to port all the current Gtk apps (Firefox, OpenOffice, the Gimp).

    Firefox is being ported to Qt, and will become as official and supported as the current version, I am pretty sure that OpenOffice will be ported one day, and The Gimp is not needed when you have Krita 2.0 (and further).

  39. Martin said about 20 hours later:

    Can Gnome3 be a evolutionary version with focus on cleanup and Qt integration?

    I guess it is true that the end user do not care much what tool kit they are using, but they do care about the general impression of the whole.

    A GTK based Gnome3 with focus on Qt integration can open up for seamless use of both tool kits. This might be archived trough development collaboration between Gnome, KDE, Qt, and GTK.

    This will make the complete change to QT less important on short to mid term. It can be done in steps.

    The Gnome community have used incremental steps to change from ORBit, GnomeVFS, and other technologies in the past.

  40. LGPL 2.1, Qt 4.5 and C++ templates said about 20 hours later:

    LGPL 2.1, Qt 4.5 and C++ templates: http://lab.obsethryl.eu/content/lgpl-21-qt-45-and-c-templates

    “There is an old moral here; for what C++ templates are concerned to the point where third party code is making use of the library and instantiating itself such templates, the GPL and LGPLv2.1 licenses are in effect equivalent in GPL behaviour of the template - based code, while LGPLv3 corrects this slight but mephistophelian oversight on behalf of the LGPLv2.1. This has been discussed over and over by several developers and you have the FSF being clear about it as in the libstdc++ case and the corrections made for LGPLv3”

  41. Henrique said 1 day later:

    GNOME is a great DE with a poor toolkit (Gtk).

    KDE is a poor DE with a great toolkit (Qt).

    I see KDE and Qt improving in a faster pace than Gnome can improve Gtk+.

    Either make a plan for improving Gtk+ to attract third-party (read “not only Gnome hackers and freetards”) developers and spread the use of Gtk+, or see the Gnome ecosystem die for the competition.

  42. Stephan said 2 days later:

    @Cypher

    KDE 4.2 is nowhere near KDE 3.5 feature-wise. Maybe it’s got some nice glitz but it’s sparse config dialogs are still far too GNOME-like for me and it has all sorts of little grating oversights. (eg. KEdit removed and no replacement lightweight text editor offered)

    I still think it’s better than GNOME, Xfce, and so on (eg. KParts, an I/O abstraction system that doesn’t crash all the time, a tabbed file manager, etc.), but I’ll stick with 3.5 until they’ve brought 4.x back up to my standards. (eg. They didn’t add auto-hiding panels until 4.2 and I can’t live without them)

    Oh, and to George Wright, allow me to express my deep disappointment in you for your “non-ajax commenting is disabled” message. (If anyone should know how much of a hassle that can be, it should be a KDE user. Javascript whitelisting is a pain on Konqueror without a statubar toggle icon.)

  43. Ivan said 3 days later:

    Personally, I would like to see Gnome switch to Qt simply because that would imply lower overhead (memory…) when I’m running Gimp or other /Gnome/ applications on KDE, and better integration with KDE. (and, vice versa, Gnome users would have a nicely integrated KDE’s killer apps like Amarok and K3B)

    As for Qt vs GTK, the only thing I can say is that the both toolkits have proven themselves and that it is not /that/ easy to ditch one for another (especially with a large existing codebase written in one of them).

    With that said, I simply love Qt, its API and fancy new features they introduce with every 4.x release. And I have never felt that much /at home/ using GTK.

    So, peace to all, and we hope to see soon what Topaz has to offer. (I’m aware it will not be soon in any meaning of that word, but I do hope that it will take you guys lot less than KDE4 development took)

    OFFTOPIC { Why is that everywhere I go, I see posts of KDE 3 vs 4? It is OK (arguably) to see them on KDE forums and related blogs, but bickering how /4/ hasn’t got all the options /3/ had and so on Qt and Gnome only related post is really too much.

    @Stephan - the editor is KWrite, located in kdebase package, so, yes, a lightweight editor still exists.

    @oliver and @Artem Each rewrite of KDE (or to be precise, each major version change) produced a lot healthier and efficient code base, and a lot of new and exciting things.

    Rewrites are not something that should be done often, but having them every couple of years has very positive effects.

    }

  44. Bernhard said 4 days later:

    Interesting how the “GTK applications” listed by people aren’t actually GTK applications. ;)

    OpenOffice is not GTK, there are just some bad distributions that make it link to every gnome library out there to push their agenda. OpenOffice by itself uses its own toolkit (VCL) which can already use Qt to draw the widgets in a native style if you just pass the right build flags.

    Firefox uses XUL, which already has different widget support backends (including Qt, but the Qt port isn’t fully mature yet).

    As for GIMP, yes, that’s a GTK app, but is it really useful anymore? It was a decent application some 10 years ago, but just hasn’t moved ahead with time as its competitors did. Krita (Qt) is doing a better job at almost everything nowadays, and it even has a user interface normal people can figure out.

    The only worthwhile GTK application I’ve seen lately is Inkscape – let’s port that to Qt!

  45. Johnny Soprano said 7 months later:

    I think we should all program in cobol, and use OLWM as window manager…

  46. rolex replica said about 1 year later:

    good articles

  47. tractor sales said about 1 year later:

    I enjoyed reading it. I admiring time and effort you put in your blog, because it is obviously one great place where I can find lot of useful info..

RSS feed for this post

(leave url/email »)

   Comment Markup Help Preview comment