Slashdot Mirror
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."
But I am *not* a license expert. This is pure hearsay.
The World Wide Web is dying. Soon, we shall have only the Internet.
Not anymore, most of them have been resolved some time ago.
Besides, whats wrong with software you have to pay for?
---- Booth was a patriot ----
This does make a lot of sense, actually. And it might actually be early enough in the game for it to work.
.NET API's. .NET might, in fact, actually be the fastest route there (aside from Java, which I think people ought to be using for this purpose, but let's set that aside for a moment). Perhaps if, boosted by a Novell push, developers begin writing and publishing .NET code that uses Qt libraries, there will be that many more cross-platform desktop apps available that won't be bound to Windows.Forms, Avalon, or whatever other Windows-bound API's Microsoft wants everyone to use.
.NET is, on its face, a good design, but that we have to worry about Microsoft using it as a cudgel to beat back its competition.
Right now, there is still the opportunity to attract developers to cross-platform
It would be a double-bonus if Novell could make Mono a unified framework for writing apps that can be backended by KDE, GNOME, or Microsoft Windows without a rewrite. Let's see what happens. What's really a shame is that
Tired of FB/Google censorship? Visit UNCENSORED!
Besides, whats wrong with software you have to pay for?.
Inappropriate for basic foundations of Linux software.
The old Microsoft approach is to Embrace an Extend. I often hear people say that Mono is going to die because MS has the staff to write more and more APIs and Mono will not be able to keep up. But my thoughts are that Mono and other Open Source projects should be able to implement some impressive and highly usable C# and .NET implementations completely independent of MS influence and support. Essentially the community will have the ability to Embrace and Extend a MS created technology with just the ECMA standard C# and CLI recommendation.
.NET framework are very robust when compared to traditional network communications such as CGI or OLE. It is clearly an ideal glue language for a diverse Linux desktop which often mixes many programming and scripting languages together.
I have been impressed with the CLI implementation and the SOAP and Web Services technologies that are a part of the
Brennan Stehling - http://brennan.offwhite.net/blog/
Hmm. I hope you mean free as in speech not as in beer because otherwise that's not much of an argument.
The point of "free" software is that it is open for perusal, poking, sharing, etc. not that it's gratis so you don't have to pay for it. Granted because of the first the gratis part is often the case but the mentality that free speech software MUST mean free beer software is just wrong.
Sure information wants to be free, but how much are you willing to pay for the packaging?
I think that many commercial development companies would consider the GPL to be "annoying license restrictions" if all they want to do is write a GUI app.
Besides, whats wrong with software you have to pay for?
Nothing if you aren't using it as a core library for a free desktop system
Hey, if programming were easy, people would do it for free.
I don't see why this is funny. It is clear that if programming were easy then people would do it for free (like everything else that is easy). On the other hand it dosen't mean that because people program for free, that programming is easy (it's not!). That's why we should all celebrate all those programmers that program for free and share their work with the rest of the world!.
keeping kde and gnome separate for a little while longer, or possibly forever, might be a good plan. There's one catch though, they need to be completely interoperable. I want to be able to install the same program on both desktops and have it work seamlessly. I also want an icon put on the desktop or the taskbar menus. Is this possible? I don't have the expertise but I'm sure it can be done if these programs are on top of a standard api.
Why do I think this is a good idea? Keeping them separate but equal promotes competition between them. It makes them work harder to fix those little glitches that annoy users. KDE is looking kind of like windows xp, and gnome mac os X. It'd be nice if they took on more of an original look, but hey rome wasn't built in a day. I think that having the choice between both desktops knowing that most applications will work the same without needing a hundred different rpms for each thing is what's needed. I also think that libraries need to be consolidated. Basically I guess I'm saying this: the ease of installing programs on windows, but the ability to run them on a mac. Sound difficult? eh, look how far things have come.
Novell is trying to create a professional desktop environment for Linux. To this end, choosing the highest quality toolkit is much more important than having to pay a small fee for the development of non-GPL applications.
Gtk's licensing only has an advantage over Qt's for those shops that want to create commercial non-GPL applications yet are too small to afford Qt's relatively cheap license. Novell have obviously decided that those shops aren't such an important market to sway their choice.
-- Repeat with me: "There is no right to profits".
It should be possible to make a run anywhere application on linux that will run on windows-- for the most part. Of course, details like file path formats are different between the platforms, so in certain cases a little mojo might be needed to work. Actually, it is mostly a case of 'best practices' and assuming nothing about the user's configuration (i.e. don't guess whether they have windows installed on C:\, find out. Don't guess that their home directory is /home/blah, use the objects given to you and find out).
As far as the UI libs go, a lot of the time, they are going to have bindings into native code-- but that is something the libraries handle for you. So assuming your qt library or whatever has linux and windows support, it should work on both platform. Obviously, if a coder screws up one side or the other, there will be issues, but that can be said of any kind of multiplatform development-- or any development at all.
====
Crudely Drawn Games
No not really. You can licence it through GPL, if you do open source work, or you can buy licences from Trolltech at a reasonalble price if you want to do closed source work.
The fact that you have to buy a licence for closed source may be a problem for hobby developers that have more time than money. To companies like Novell the quality is more important than the price . To them, time to market is everything. And they will spend far more money on developer saleries and marketing than they will spend on QT licences. So they will go for the best tool. And QT is one of the best there is. (Cross platform, good development tools, well designed framework...)
And if they want to release free software they have less risk of being hurt by some competitor if they release it under GPL than under LGPL or BSD type licences. I guess this was the reason MySQL AB changed the licence on their client libs to GPL.
So, QT would be a very good choise.
God is REAL! Unless explicitly declared INTEGER
First off, the cost of the software was only listed as a barrier to adoption, not the only one. As for what the additional barriers there are, two that I can think of quickly are learning curve and OEM buy-in. It is human nature to be reluctant to learn to use product B when you already are comfortable using product A. It takes a pretty good incentive to push people to expend the necessary energy. As for the OEMs, if they aren't already locked into a restrictive exclusive contract with Microsoft, they are probably reluctant to spend the resources to support installation of an alternative operating system on the machines they sell unless they see a sizable market. Given that it's a chicken and egg argument, adoption of Linux on the desktop will be slow until the market gathers enough mass. If and when that point is reached, you'll probably see very quick gains in market share. Getting large companies like IBM and Novell behind the push will go a long way in helping adoption.
GreyPoopon
--
Why is it I can write insightful comments but can't come up with a clever signature?
Wrong. http://dot.kde.org/1073668213/
How can I say that? Well let's see, there are many boxed versions of Redhat & Suse that have been sold and I don't think anyone was ripped off.
While it is probably impossible for Redhat not to create a free as in beer version of their free as in speech software, the fact that it can and is paid for is very telling.
The main advantage of Open Source to most people is absolutely NOT the free binary whether they know it or not. It is not a "few geeks such as yourself" there is a very large world of geeks out here that value the source to their programs for many reasons other than that it may be free as in beer as well.
If in fact you value the monitary freeness of your software over the freedom of the source than I would argue that your not a geek at all but just someone who wants to ride a gravy train.
Furthermore the dichotomy you used between KDE & gnome is false because both are GPL'd. Would you rather pay for an Open Source KDE or get a closed source Windows for free? If you choose the Windows option than you definitely don't understand freedom and your no geek I want to hang out with.
The fact is that people have to stop with this idea that "free" in the GPL'd way means "free beer". Free software is a costly thing,I have spent untold hours of my valuable time helping to code free software, but that's fine I don't expect payment for that work because I was paid in kind by other coders. But users who just take and never give back must understand exactly what they are doing. I can't force anyone to pay for software they can get for "free" but your mentality that software MUST be free as in beer is totally antithema to the actual freedom that is espoused by free(dom) software.
Freedom software is about having the source, never looking over your shoulder for the thought police, adding to something and feeling you've contributed, learning, sharing. "Free beer" software is about being a leech!
Sure information wants to be free, but how much are you willing to pay for the packaging?
Use GNOME/Gtk, because you can USE Gtk as much as you want for COMMERCIAL development without paying anything.
Don't use KDE/Qt, because you can only develop FREE software using it, otherwise it costs money.
So.. NOW KDE/Qt is the champion of free software, whereas GNOME/Gtk is for the COMMERCIAL (and apparently not so evil after all) PROPRIETARY closed source solutions.
You make me laugh!
If GNOME/Gtk is REALLY a friend, let's see them place everything under GPL (for true software protection) rather than the LGPL.
What's the big deal about support Qt is you use the toolkit? Yep.. it's commercial...and if you use it for commercial development, it costs money... so?? Is someone suggesting their software business plan is only to sell like 10 copies of their software, so they can't afford to by a real development license?? Just seems weird.
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.
There's no reason a small business can't do commercial work and license it under the GPL. It's done all the time by the Linux distributors both large and small, and many other development groups.
Anyway, I see this misconception about QT's licensing all the time, and I think part of it stems from Trolltech's own use of the word commercial to describe their other license. The word they (and you) are looking for is proprietary, not commercial. If you don't want to use the GPL (usually because you're paranoid about people stealing the ideas, or because you want to link with some other proprietary code), then you have to pay for commercial (i.e. proprietary) Qt licensing. However, you're free to use the free (GPL) version of Qt commercially, as long as you follow the GPL.
I was going to moderate, but this is driving me insane. There are just too many posts like this. How is $1000 "locking out" single developers who intend to sell closed-source licenses? Even for sole proprietorships $1K is nothing special. Besides, you are talking about selling closed-source software, which by definition requires other people to pay you money per license, but somehow you think TrollTech is a fiend for wanting to do the same thing. If you want to be open, TrollTech is right there with you.
Our intelligent designer has never created an animal that we couldn't improve by strapping a bomb to it.
After 7 years of CORBA push, GNOME uses CORBA for pretty much nothing.
Try to get a GNOME guy to give you five real, practical examples of how they use CORBA. He won't.
He will mention that some apps can embed a gnumeric sheet, though.
I'm perennially amazed that the GNOME zealots, who started out from the GNU "all software must be free" zealot camp, now argue that Qt is bad because it doesn't allow proprietary applications. (Actually, it does, if you buy a licence from Qt. Unlike, say, GNU's readline library, which was deliberately GPL'd and not LGPL'd by Stallman, who will not issue you a commercial licence.)
In fact, RMS even wrote an article on why you shouldn't use the LGPL for your next library. Without the backing RMS gave GNOME in its early days, when it was an unusable piece of crap and KDE had already hit a high-quality 1.0, it would never have got off the ground. (Remember GNOME 1.0? *shudder*) And yet the selling point now is that GNOME is more suitable to proprietary apps? I just can't figure out where all this is coming from.
If anything, Qt is a shining success story on how to make money with GPL'd software using a dual-licensing strategy. Far from continuing to vilify Troll Tech, the GNU/GNOME zealots ought to trumpet this story as a way to encourage more proprietary software companies to play nice with the linux world. (Peter Deutsch did the dual-licence thing long back with ghostscript, but he only released year-old versions of ghostscript under the GPL, and that's still the practice. Troll Tech releases current versions of Qt under GPL as well as their commercial licence.)
Yes, but Gtk+ doesn't restrict you to C, whereas Qt is almost entirely C++ (except for some scripting language bindings, like Ruby, Perl, and Python).
I despise C++. Hate it. It is an abomination both to C, and to OO programming. The syntax extensions to C are complex, as is the OO model. The only thing I think C++ did right is the templating system.
I prefer Objective C, which takes a minimalist approach, and the late dynamic binding makes templating irrelevent. It's a much cleeaner language.
So, the issue comes down to choice. I know a lot of people prefer C++ to Objective C. You can't account for taste-- it seems more people prefer Budweiser to any other beer, and McDonald's to any other food source, and MS-Windows to any other desktop OS. But the point is, there is choice. I know I don't want to declare a single standard language for all programmers to use, and I distrust anyone who does.
If they standardize on any one toolkit, we are screwed as developers. Instead, they should concentrate on standardizing protocols. And, from a comment Nat Friedman made somewhere along the line in this topic, that's exactly what Novell is doing.
Anyway, just my two bits.
- Tony
Microsoft is to software what Budweiser is to beer.
If you're going to write software either write it as a hobby and make it free-beer, or go whole hog and be a real professional. Shareware is facing the end of its existance. Once Free Software catches on in the Windows world, shareware is dead.
Don't blame me, I didn't vote for either of them!
I am really missing the point of all of this. We have a link to a German article that is uninteligble to the majority of posters here, and a lot of people talking absolute nonsense. Why?
When the previous article about Novell merging KDE and Gnome I immediately thought, which platform will they use as the base. I guessed KDE since the KDE people are already doing all the work whereas the Gnome people don't seem so interested. Nobody asked that fundamental question. Why?
Now, assuming that article says what it is claimed, I am left wondering why people think Gtk is being dropped, and scratching there heads over the Ximian aquisition. Why do people think that Gtk is being dropped, and what does integration mean to you anyway?
When Novell say *they* are standardising on Qt why do people then say that everyone else must now develop their apps in Qt too? How do you explain the fact that Gtk apps have long worked under KDE, and now Novell wants to integrate them more tightly?
To me this all sounds like Novell are going to make more use of Qt for their own future development, are going to use KDE as a base system for their OS, but are going to continue and extend the work being done in the KDE camp so that Gtk based applications work flawlessly in KDE (print dialogs, file dialogs, look 'n feel, font settings, control center options, etc).
The only question I am left wondering is whether they will also offer a Gnome desktop and do the exact same integration the other way round so that KDE applications work flawlessly in Gnome. Anyone care to comment?