Novell Desktop To Standardize On Qt [updated]
Balinares writes "NewsForge reports that Novell has settled for Qt as its Linux desktop development environment, casting more light on their strategy to unify KDE and GNOME. This ought to be interesting. The prospect of using Mono to code against Qt makes me drool in advance. Maybe programming will suck no longer!" Update: 03/30 00:01 GMT by T : Sounds like that story doesn't quite hold water; Nat Friedman writes in this Slashdot comment that "We have not decided that we are standardizing on Qt for the desktop. ... We support development with a variety of toolkits, and our internal development is done using the right tool for the right problem. This includes Qt, Gtk, VCL, XUL and others, depending on the application."
It's good to see these two being merged. I love Linux, but it seems as if some people need to see a common desktop, for them to feel like Linux has arrived. BTW: FP!
- Will they be writing a GTK-on-top-of-QT layer?
- Ditching GTK apps?
- Using GTK apps but not encouraging their creation?
- Just using QT for Novell-specific stuff (like Redhat uses GTK for its configurators, etc)?
*confused*Napster-to-go says "Fill and refill your compatible MP3 player", which is a lie. It's not MP3. It's WMA with DRM.
No way, not cool. I don't want to sell my customers a Linux distro and then tell them that if they choose it for development, they have to pay for dev licenses of QT and an IDE and ... That's bullshit. What's the alternative in my client's eyes? Write internal code against GTK, which will look like a secondary/inferior choice to them, whether it is or isn't. And with Mono inhouse now, how long before Miquel and the Mono group will be forced to use QT#? Wow, now the default toolkit for Mono requires a developer license (which is WAY overpriced). I disagree with this wholeheartedly. My opinion may not amount to anything, but they are wrong in doing this.
the most obvious is that it is a barrier to adoption. if your model is to sell software and make yr money that way, everything's fine... but if you're model is have free wares to drive sales of other commodities (hardware, support, data &c) then the extra pricetag on the software can tank your business.
since you asked....
2 1337 4 u!
Build software that is not sold, but that advances the business goals of a commercial enterprise."
in other words, every company out there that uses KDE legally needs to buy a QT license???
Please someone tell me that I am assuming wrong.
Do not look at laser with remaining good eye.
I can say that there is nothing easier outside of web development. I am an old MFC programmer. I am often lost in developing Qt apps, but I am very comforatable with that because the documentation is copius cnad clear. Whats more is it is soooooooo well thought out. It makes MFC look like the crap that it is (C++ wrappers for C objects). Learning Qt is like learning to walk the right way. It's amazingly simple. I will always request that Qt be used regardless of platform in future jobs.
Now the license is different. I often wish there was a small-business or starting-business license, but this is only pertanant if you are going commercial work. for GPL work it is completely free.
Right now I'm doing some advanced work with QSA (Javascripted Qt apps) It is easy and cross platform. I can now replace a browser (and the rendering issues with a user interface file (loaded at run-time) and ECMA script code (platform indep. cause we run on various architectures with limited space, whose list may change at any time)
The Troll Tech stuff is top notch.
Slashdot's rate-of-post filter: Preventing you from posting too many great ideas at once.
I'm surprised redhat has stuck with gnome this long as their DE of choice. It is more usable, and that's why I am glad personally they have.
With so many of the gtk programs riviling the qt equivalents, I wonder why companies always flock towards Qt.
- tristan
Actually, this may be the event I was waiting for to start writing X desktop apps instead of Windows. Mono is very very powerful and combined with Qt its even better. Who'd have thought that .NET could actually lead developers away from Microsoft..
I personally have been hoping for a while now something like this would happen for the Linux desktop. It's going to take a corporation to step up and unify this effort in order to gain mass acceptance. I'm sure there will be some grumbling in the community, but open source is open so feel free to customize to your hearts content if you don't like it. Most people don't want to have to go to this effor though.
The only potential problem I see is Trolltech's insistance on license fees for commercial development. Not that this is any different in the Windows world, but it'd be nice to give ISVs a completely royalty free solution. I'd like to see Novell take that $50 million that IBM gave them and purchase TT outright and put Qt under a more liberal open license. The wording on the KDE Free Qt clause seems a bit unclear to me. Does Qt get automatically BSD'ed when any company buys Trolltech. What if the purchasing company doesn't make the license any more closed that currently, does that have an effect.
Anyway, I've never understood the reasons people chose to write a whole desktop environment in straight C. C++ just seems a far more natural fit. I've looked over both GNOME and KDE fairly extensively, and there is no doubt in my mind that KDE has a cleaner code base and architecture. With all the "higher level language" rumblings going on in the GNOME community, I suspect that those developers are hitting a brick wall in terms of where they want to go and what the current code is capable of becoming. That sort of thing isn't happening in the KDE world, so I think that speaks volumes.
Good luck Novell, you've got at least one supporter here.
-- Fighting mediocrity one bad post at a time.
The only source of this so far has been Heise. In their original article, they word this quite unclear (I'm german) basically just mentioning that this information "slipped through during the day", without any source or confirmation.
However, derStandard.at asked for confirmation at Novell and they assured that no decision would have been fallen yet.
They (Novell) also demoed quite a few new Gtk applications at the keynotes, most notably the interesting iFolders (using Gtk on Linux), while I don't think there has been shown any _new_ Qt program.
So take this all with a grain of salt, I can't really imagine that this is a clear decision yet.
Linux has been free for years and yet only accounts for 5% of the desktop max. So what is the real barrier to people adopting Linux then?
Support the First Amendment. Read at -1
Give it up man. Everytime I see Qt mentioned, someone posts this FUD. Yeap, you're uding your own FUD against Linux.
I D=60
I'll admit it, the first time I read this conspiracy thoery I was worried. I looked into it. Trust me there are no black helicopters being dispatched from SCO.
KDE Myths #60: http://kdemyths.urbanlizard.com/viewMyth.php?myth
"According to http://www.trolltech.com/newsroom/investors.html the Canopy Group only has 5.7% shares of Trolltech while 64.7% are in posession of Trolltech employees with an additional 5% controlled by the Trolltech founders. One can hardly say that the Canopy Group owns or controls Trolltech."
Don't forget Canopy is just an investent group. They'd probably like to see TrollTech get this boost because SCO is a loss at the moment, and they aren't going to get better.
It is like saying that you own a mutual fun that invests in two competitors, and that you are going to devalue one so the other can rise. The problem is that you want BOTH to perform.
Slashdot's rate-of-post filter: Preventing you from posting too many great ideas at once.
Your company's stance on the GUI toolkits available for Linux is a good one, and makes a lot of sense - as you said, Novell should not standardise on one particular toolkit.
So how could the keynote speech have been misconstrued?
SCREW THE ADS! http://adblock.mozdev.org/ Proud user of teh Fox of Fire - Registered Linux User #289618
What are the legal ramifications of using something like Mono? I'm guessing there isn't, as I'm sure Novell would've taken it into consideration any possibility of Microsoft trying to patent .net technology.
.net and totally screw over any open source projects that use Mono?
I don't know if it's just heresay, but could MS patent
We have secretly replaced these Slashdot mods' sense of humor with a rusty nail. Let's see if they notice!!
Apparently Chris Stone said it in the press gaggle following the talk.
I see you are *still* presenting yourself as the desktop lead, Nat. Interesting since from all accounts, Chris Stone occupies that position.
Why don't you set the record straight as to the organizational heirarchy of the company and where you and Miguel sit in the organizational tree? Do you answer to Chris Stone or Richard Seibt or do they answer to you? Or are you all just a bunch of middle management?
I don't rate $1000 per seat as a reasonable price when I could give each developer Windows XP and bulk-licensed Visual Studio for a much lower price. I think QT is a great toolkit and am glad it's licensed under the GPL, however their closed-source version is much too expensive and I have to wonder whether many people actually buy it when there are equally as valid business models that involve GPL'd products these days anyway.
Please specify a single popular commercial app that uses GTK...
People fail to understand that you don't pay only for the right for developing commercial programs when you buy a QT licence. You are also paying for SUPPORT.
There exist lot of free software for Linux, thus if you want to SELL your software, you need to invest lot of money in any case to create something much, much better.
Last time I checked, there were lot of undocumented widgets in the GTK. How are you supposed to figure them out? $1000-$2000 QT licence pays itself quickly back with faster development times.
Heh heh... Even now I can picture Balmer rubbing his fat little hands together and giggling gleefully. Micorsoft's fondest hope is that a good chunk of open source software will come to depend on mono before they play the patent card. Assurances that Microsoft would "just never do such a thing", somehow leave me unconvinced. ...and I'm sure someone will tempted to respond with the "but, but C# is a standard!" line too - don't bother, just submitting the "standard" in no way prevents Microsoft from enforcing related patents anytime they choose.
Regardless of NET's good or bad points, it's a potential legal land mine for open source that could make the SCO fiasco look postiviely quaint. Unlike SCO, Microsoft would have an actual case.
I see you are *still* presenting yourself as the desktop lead, Nat. Interesting since from all accounts, Chris Stone occupies that position.
From the Novell web site I would have guessed that it is Markus Rex:
"He is also responsible for Novell's Linux desktop activities."
Because Qt is a better, faster, and more well-documented toolkit than GTK+? Its got many more commercial users (including a lot of big-name companies) and a dedicated team of full-time developers.
A deep unwavering belief is a sure sign you're missing something...
Why is the open source community interested in taking sucky .Net and making it suck on Linux too?
Well, it appears that it was actually SuSE, with their Trolltech connections, that took over Novell, and not the other way around.
And if we are not careful, Trolltech is going to end up controlling Linux. [We also need to consider the possibility of a backroom deal between Trolltech and Microsoft, similar to the deal between Microsoft and SCO.]
The danger, of course, is not the GPL'd version of Qt. KDE is also not a problem. Both of those are Open Source, so we don't have to worry about them, and I have no reason to talk about them.
The real danger of Trolltech is the proprietary version of Qt and the applications that depend on it.
Or, in more general terms, the danger is proprietary middleware.
Let's look at an example from history. Remember when the PC platform was open, from bottom to top? Remember when you could not only buy the hardware from any number of vendors, but there were also competing vendors for the BIOS, there was more than one windowing architecture (Gem, Geoworks, Borland's GUI libs,...), and there were dozens of development environments, and thousands of code libraries? Remember when standards were simple, when it was easy to write code to interface directly with printers, with video cards, and so on?
What happened? Microsoft introduced Windows.
After a while, Windows became an indespensible component of the PC. Applications depended on it, and hardware manufacturers had to support it. And there was only one supplier of the Windows component, namely, Microsoft.
Soon, Microsoft started using its control of Windows to make itself the winner in the applications market. And today, Microsoft (with a few accomplices) also dictates the standards for PC hardware, its BIOS, its peripherals, and its network protocols.
What was special about Windows? It was proprietary middleware!
Windows sat in the middle, in between the applications and the PC. Any applications that wanted to access PC hardware used (and became dependent on) Windows APIs. And any PC manufacturer that wanted applications had to support Windows APIs. Bill Gates once said that by controlling the APIs, he controlled the industry.
Likewise, PC users became locked in. They could replace their hardware. They could replace an application. But they couldn't replace Windows. As the PC became cheaper, Windows became more expensive, as did the Microsoft applications that Windows "encouraged" users to use.
The proprietary version of Qt is just like Windows. It is proprietary middleware. It sits between the applications and Linux. The applications that use Qt are dependent on its APIs. And if Linux wants to be able to run those applications, then it has to support Qt.
What applications are we talking about? Applications like Kylix, Hancom Office, Opera, the professional versions of Quanta and Rekall and various other TheKompany products, ATI's setup utility, Quasar Accounting, Adobe Photoshop Album, and so on. If a Linux distribution wants to be able to run those applications, then it must support the proprietary version of Qt.
So how bad is it? Has Linux been taken over by proprietary middleware, the way the PC was taken over by Windows?
In other words, Have we reached the point where the proprietary version of Qt, with its single supplier, is an indedpensible component of Linux?
If this article is correct, and Novell is standardizing on Qt as the foundation of its desktop and development environment, then things are very bad.
Others have also argued that it has become almost impossible to have Linux without proprietary Qt...
In their paper Conquering the Enterprise Desktop, a group of developers argued that Bruce Perens' UserLinux would have trouble succeeding, unless it included the Qt Library in its basic install. Were those developers just talking about the GPL'd version of Qt? No, as shown by these statements:
I've used Swing on small projects, but never used GTK, Qt or GNUStep.
I had a job for part of which I spent about four or five months writing WxWindows. It seemed fairly easy to pick up, entirely reasonable as a development platform, and the quibbles I had with the functionality and documentation only came up a couple of times (one of the class docs was misleadingly written at one point, and one of the classes, WxListBox or something, turned out to only work on the Windows platform). Is Qt "better" than it? I wouldn't know. All I know is I had no objections to WxWindows whatsoever.
Most of the stuff I do is not GUI. However what I do all of my personal GUI development in is Cocoa, one of the two Mac OS X GUI libraries, which I consider as elegant, complete and (usually) well-documented as I could possibly imagine a GUI library to be. From my perspective, it is generally a joy to use. And, um, well, it doesn't cost me anything to use. The development tools came free with the OS. Now, given, I had to pay money for OS X itself, and anyone who uses my programs will have to have paid money for the OS as well. But, um, you know what? I don't really mind paying for the OS so much. Personally. I feel like I'm getting what I paid for there. And I CERTAINLY prefer paying money to use the OS and then getting dev tools and a lovely dev library that I can do anything I want with to either YOU WILL USE THE GPL PERIOD or having to pay, um, what appears to be a per-year, per-developer, per-platform fee of at least either $500 or $1500, i can't tell which, to develop software. What I'm currently getting just sounds like a better deal to me, so personally I'm going to stick with it.
(And since Cocoa is what GNUStep aims toward, who knows, if I really wanted to I think I could probably port my software to GNUStep without *too* much trouble. I know GNUStep is woefully incomplete, but this is why I mentioned it.)
But this is of course totally just my opinion. And it isn't entirely germane. But hey, you asked.
Irritable, left-wing and possibly humorous bumper stickers and t-shirts
My first thought was that the reporter had misunderstood something Chris Stone had said, perhaps because of language issues. This seems especially plausible if the reporter had any sort of pro-Qt or pro-KDE bias. People tend to hear what they want to hear.
Maybe programming will suck no longer!
At first I read this as 'Maybe programs will suck no longer!"
Good luck with that one, given the direction of the current corporate development climate.
Probably applies to programming also, same reason.
Glonoinha the MebiByte Slayer
You almost have it.
I haven't actually been arguing about profit in a monetary sense at all. What I'm trying to argue is that users of freedom software must be conditioned to know that the expectation is that they give back something. I don't care what it is, coding, documentation, word of mouth of how freedom helps them or money.
In your case I've been trying to condition you to react by both giving back software and evangelizing freedom NOT free beer.(since you mentioned your a coder)
This started because it seemed to me that you expect that software must be free of cost in a monetary sense. Monetary free is only a possibility from freedom source not a mandatory requirement.
"The only way to make money with information is to either generate new information or to provide a more efficient way to spread,gather or copy information"
There you go you got it. Well at least you have part of it. Where in your above statement does it say anything about the nature of the license attached to the information? It doesn't matter if the information is licensed under the GPL, if I make my product more attractive, easier to use, more worth while, more...something. Than people will pay, either through their own hard work, word of mouth or with money. In fact it should be expected that people do pay in one of these manners. Anyone not paying somehow, not necessarily with money, is a free loader pure and simple. In the long run enough people will see the value of paying for freedom software to make it profitable.
As to the music,movie, television industries. Their problem isn't that they are losing control of distribution it's that they are LAZY. They refuse to charge the correct amount for the information or they refuse to provide sufficient value-added to make buying the information desirable.
I will pay a reasonable amount to have a CD,book,DVD or easy cable access. Currently all of these are too high for the product that is sold. I care not a wit for "Top 40" music it's not worth even a penny to me. If a music exec asked himself why than maybe they'd understand. The reason is there is absolutely no value in that information to me. I won't even waste bandwidth on it. However, if you somehow got George Clinton and Parliament back together, put together a good CD of funk, with bitchin' liner notes and maybe throw in the possibility of tickets to see them at a concert in New York(e.g. put a coupon in random jewel cases) I might pay $30,$40 or even more dollars for that. In both cases the information is free(the music) but only in the latter is there a value added service that I'd pay for.
As a physicist by training and a coder by profession I KNOW information wants to be free. The question is what are you willing to pay for sharing in it?(Again pay is not just in a monetary sense.)
Sure information wants to be free, but how much are you willing to pay for the packaging?
Anyway, I wonder what in the world would make Novell pick QT? I don't have a problem with the QT license since it is free for GPL'ed code and if you want to do closed source, you pay, just like the rest of the world. I am fine with that. What I don't understand is what is Novell's game plan concerning all the Ximinan "IP" that they bought? There is no acceptable QT based groupware application that is ready _NOW_, while there is GTK+ based Ximian Evolution that even works very well with MS Exchange. There is Mozilla that is GTK+ based. Novell is not going to get all those users to switch to konq when, IMO Konq is not standards compliant enough yet and the rendering engine is not up to snuff with Gecko. What about OOo? Again, it uses GTK+, I cannot see Novell porting that beast to QT. These three apps are critical to the Linux desktop, everything else is fluff in comparision.
Maybe they plan to do some good QT/GTK+ integration? I hope so, since the two tool-kits have issues with simple interoperation such as drag-n-drop and copy-n-paste, not to mention how different the widgets look. I personally don't want a Linux desktop that has no consistency between the widgets. This is the main reason I always look for Gnome apps over KDE apps to keep my desktop consistent (with the exception of K3b, since it is one great app).
Oh, there is also the issues of the different technologies that the two desktops use. What sound server are they going to use? arts or esd? Are they going to make the coding changes to to the all the other apps or do they expect a user to have both running? DCOP or Orbit? etc vs etc. IMO, Novell should have picked one desktop and then ported any of the very good apps from the other desktop over. It would take a long time to port Evolution, Mozilla/Firefox and OOo to QT and have it be stable.
Since Redhat dropped interest in the desktop, I was hoping for Novell to help push Desktop Linux, though this move makes me less optimistic.
If Tyranny and Oppression come to this land,
it will be in the guise of fighting a foreign enemy. -James Madison
impossible to have Linux without proprietary Qt...
that a Linux distribution must include support for proprietary Qt
In effect, they are saying that proprietary Qt is to Linux, as Windows has been to the PC.
etc.
There's a BIG difference between windows and Proprietary QT here, and the difference is that Windows costs money per-user. So to use the windows middleware, your *users* have to pay Microsoft.
However, with Proprietary QT the only cost is per-developer. So to install opera, hancom office, etc, I don't have to pay a penny to Trolltech. Only the companies developing these proprietary apps need to.
For that reason alone, trolltech is not going to fall into the control of trolltech.