Slashdot Mirror
'Gnome Foundation' Takes Aim at MS Office
Spasemunki writes: "The NYTimes has a piece today on an agreement reached among I.B.M., Compaq, Hewlett-Packard, Sun Microsystems and several other developers to create the Gnome Foundation, a developer consortium that will undertake, among other things, the creation of a standardized desktop interface for Linux, and a suite of productivity programs designed to compete with MS office. As the name might imply, their efforts will center around the Gnome desktop manager, with Sun moving to adopt Gnome as the GUI for Solaris. Looks like some big names are getting interested in putting Linux on the desktop."
For example, I work for a major futures and options exchange, and a increasing amount of our back-end stuff is running on UNIX (in our case, Solaris). All the new applications we write are UNIX-only on the backend.
The only major thing we use the mainframe for is for Clearing. The next version of our Clearing software is planned to be running on a UNIX cluster.
We use a Tandem for matching all the trades that come into the exchange. That system can be (and is planned to be) ported to UNIX!
Even though its been around longer than I have been alive, UNIX is the future. :-)
In an unrelated sidenote, I told my coworkers 2 months ago that within 2-3 years Sun would dump CDE in favor of GNOME as its desktop environment. You should have heard me shout in excitement when I saw this on /. this morning ;-)
siri
Until one can guarantee 100% bug-for-bug compatibility with MS Office - including templates, macro functionality, and UI, and keep up with Microsoft's gratuitous changes every few years, it's dommed to fail.
I dunno about that. Even Microsoft doesn't rise to the standard of 100% compatibility with itself -- I find that office routinely alters the rendering of rather complex documents, which basically serves me right for confusing content with presentation. At least for word processing, competing suites need to do what MS itself does -- maintain the text and render the majority of documents reasonably consistently, without sweating too much about pathological cases.
I sincerely hope the Gnome office effort gets lots of dough to play with, but I don't think this is exactly earth shattering news. One way or another pretty soon we'll have quite a choice of Office suites which will probably have good enough MS Office import. I tend to view this development as some rather belated bandwagon hopping. If they had done this a year ago they'd have been visionaries.
I think that fonts and font rendering are much bigger stumbling blocks. Star Office has every feature 99% of people need and good enough import for most MS documents, but after installing, it by default produces documents that look hideous on screen because of font scaling issues. That is ridiculous in this day and age.
I'd be much more excited to see the money spent (if there is any money) on Berlin, or on producing a really great set of free typefaces, or on improving X. If there were Aqua quality graphics on Linux, then we'd really be ready for a fight.
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
Several things need to be said (and then moderated up), because no one else is saying them.
- from gtk import *
- win = GtkWindow()
- win.set_title( "Hello, World" )
- win.show()
- win.connect( "destroy", mainquit )
- mainloop()
Go ahead, try it right now. Oh yes, another thing; the GTK+ library is LGPL'ed. As in Unambiguously Free Software.Qt licensing. Although Kde's use of Qt has been generally accepted as license compatible with GPL and LGPL (the naysayers have clearly been over-ruled here) Qt remains the property of TrollTech. It can't be forked. This is simply not acceptable to commercial backers of a linux desktop standard, because THEY may want to work on the underlying gui code as well, independently of Qt. Why is Java not more widely used on the Linux desktop? Because of Sun's "ownership" of the standard. Nobody "owns" gtk and gnome, and this philosophy is agreeable to all commercial parties involved because they are all competing on a level playing field with Gnome/Gtk. While in my opinion, at least, Gtk is a vastly inferior toolkit it is far more maleable because there are no such restrictions on redesign and extension. No commercial entity has veto power as TrollTech has over Qt. TrollTech has made the decision not to GPL Qt, and it was a bad decision. Therefore, TrollTech has lost. It's over, folks, unless TrollTech GPL's Qt and does it VERY quickly.
Kde is not a superior desktop to Gnome any longer. Sure, it was superior a year ago or even 6 months ago. What many people fail to realize is that the new Kde 2 is far from stable. It's shaping up, but the underlying architecture will continue to be changed right up to the date of the final release. There are many problems with the current implementation. Read the Kde developer lists.
Gnome, on the other hand, has its architecutre pretty much in place for the 1.x series which is very competetive with Kde 2. Notice that Gnome updates (at Helix) are incremental. A few packages are updated every week or so, instead of a major new release every 6 months. The continuity for developers is much smoother. New Kde apps are NOT being ported from Kde 1.x to Kde 2 in great numbers because the architecture is so different and Kde 2 is a moving target and an unstable one at that. Frequent crashes of the core components are very discouraging to developers (outside the core Kde team) who are considering porting their old Kde 1 apps or developing new apps for Kde 2. Gnome, on the other hand, is rock solid in my opinon. Sure the clunky gtk+ objects using C are less than sexy for C++ programmers and others, but they work with the current Gnome quite nicely and developers are developing for Gnome in far greater numbers than for Kde.
Regarding esthetics, that is a matter of opinion. Kde and Gnome have very different looks, although with the ability to import some themes from Gtk Kde can overcome the difference somewhat. I think the Kde 2 desktop is beautiful, but the new Gnome using Sawfish as the "standard" window manager may be more 'commercial' looking. The new Kde has a sort of a classic Egyptian look but the Tigert artwork has its own kind of appeal as well. Both Gnome and Kde are far more visually appealing and customizable than Windows. It's a tossup in the artistic area.
In summary, TrollTech has blown it by refusing to GPL Qt. Nothing else is really relevant here. TrollTech could have had it all, but instead chose short term profits from exhorbitant licensing for its "professional" edition over having its toolkit adopted as the standard for linix and unix as it should have been. Troll Tech still has a small window of opportunity in which to rectify its mistake and give Kde a fighting chance. Let's hope they do.
Login: free-news Password: slashdot
I don't think that this is a death kneel for KDE. Will it mean that more and more standard installations will have Gnome as the primary desktop? Maybe. Just because several major companies have said that they are adopting Gnome as their standard does not necessarly mean that it will happen. First, we have to see Gnome distributed as a standard. Second, see Gnome actually implemented on those distributions. If those two things actually occur, then there is a higher chance for Gnome to take a more high profile existance in the different Linux distros. But just because the major UNICES are using it doesn't not necessarly imply that the other Linux distros will.
Are the big corps trying to 'Bogard' the Linux marketplace? I'd say no. All of the companies listed above realize that UNIX is not ready for the home user. They also seem to understand that they have a worse chance of taking on MS in the low-mid size server market that Linux now has 24% in. By supporting a common user interface between the different UNICES and Linux they stand to gain in many ways, all of which include keeping Linux.
Linux is starting to eat away at MS's marketshare. By supporting a common GUI between Linux and UNIX, there becomes more incentive for a corperation to use Linux on their workstations and have their huge UNIX box in the background with seamless connectivity, no more WinFrame. As people get used to Linux, it will gain more market share in the home market, people don't like to learn more than one system. By making MS not the standard they can sell more of their product.
Are these companies after Linux's marketshare? Not really. The big corps listed in the article know that their product is not a good product for home use. In some of cases it is TOO powerful for workstation use, which is where Linux comes in. Linux can scale downwards very easily, which is something the brand name UNICES cannot do. At the same time the big corps are not worried about Linux taking their market share either, they know that Linux does not scale upwards at the level any of the brand name UNICES can (we all know that it is true, deal with it). Anyway, companies like Sun are more conserned with selling hardware and services anyway, why do you think they released Solaris 8 for free (8 processors or less). In the case of Sun, if pushing Linux means selling more hardware then they are all for it.
This is good for the open-source movement. The support of those big corps will only give more good press to Linux, which can only serve to increase the user base. A large user base means more testers and more programmers. It means that more programs will be written for Linux. This is a good thing for Linux.
Disclamer - Opinion of Person
1. TrollTech has already said it would GPL Qt, if the GPL v3 prevented proprietary development. Currently it does not, in this respect a lot of the trouble is caused because the FSFs PR engine refuses to admit this. See Eric's piece on FreshMeat a few months back (And Matthias E's comments expanding on it).
2. It is true that KDE 1.x is no longer better than Gnome 1.2, but KDE 2.0 blows Gnome 1.2 away. You are wrong that the underlying framework for KDE 2.0 will keep changing - it won't. There aren't 'many problems' with it as you claim, there are some bugs sure, but they are getting fixed extremely rapidly.
3. You are so wrong about the difficulty of porting from KDE 1.x to KDE 2.0. Most simple apps can be ported in the most minimal sense of the word in less than half an hour. More complex apps might take a week or so. If you actually look around you'll see that the process of moving over to KDE 2.0 has already begun with many of the non-CVS apps starting to develop 2.0 branches. This will become more and more common as we approach closer to the release.
I also doubt your comment that more people are developing for Gnome than for KDE. For one thing a big question is what are they writing? In many cases they are still playing catch up for apps KDE had a year ago.
It is really not an issue if Sun and Gnome try to do this - for one thing the effort involved will be huge, and I seriously doubt if the results will justify it. Star Office has some good points, but there's no getting away from the fact that it is a pretty hairy piece of software. By the time such a port becomes usable we will already have KDE 2.1, and maybe even 2.2.
One thing a lot of people has missed is that the plan for KDE 2 is to have a much more rapid release cycle than that between 1.1 and 2.0. We have spent a lot of time making sure that not only will 2.0 be good in itself, but also that it is an API that is well documented and will remain usable for a long time. KDE 2.1 and 2.2 at the least will remain source and binary compatible.
Here is another story about this over at The Register. My question through this whole thing (since the announced GPLing of StarOffice) is, what is going to happen to the Gnome Office? I personally like Gnumeric much better than the Star Office Spreadsheet.....is there space for two office suites in the world of Gnome?
I read a piece 3-4 years ago that predicted that in 5-6 years (from the date of the article) there would be 2 main OS's in the market: Microsoft's offerings (Windows and derrived), and Linux. We are progressing rapidly in that direction. Not to belittle Solaris, Apple, BSD, etc., but Linux is gaining ground rapidly at the expense of everyone else.
With the opening of StarOffice, the use of Linux everywhere (embedded, desktop, server, super computer), high-powered desktops like Gnome and KDE, Linux is coming of age!
AC because I have to be.... Pls. moderate up!
Couldn't they concentrate effords?
With all respect to the Koffice team (I'm very fond of KDE), they should do with Koffice what the GIMP guys are doing for GIMP 2.0: making it toolkit/environment independant. The new GIMP should allow both KDE and Gnome front-ends with the same engine. Why not follow this approach for the Office suite?
*NIX would be even better if some of these common modules would seperate engine from interface, just like the OS is seperated from the GUI.
A few existing examples: Kicq has done this by creating ICQlib. XMMS uses mpg-123, aKtion uses xanim..
This would be harder for Koffice of course, because of the KParts integration within KDE, but there is no reason to have a unified rendering and file format (including filters) engine for office tasks such as word processing and presentation.
Look at Windows NT and 98 -- they start off with the easier, consumer-grade stuff, and move on to NT relatively painlessly. Yes, the underlying system is completely different, (and no, I'm not really trying to compare Linux to Windows 98 -- it's just for the purpose of illustrating this admittedly limited point) but because the widgets and desktop layout are similar, you could sit any relatively competent Windows user in front of any Windows machine, and they could run Office, surf the web, or copy some files.
IBM, HP, and Compaq get similar benefits -- their in-house UNIXes can have a familiar interface, and they can train an entire generation of UNIX users by cloning a dominant Linux UI. Personally, I think it could be great to be able to sit down at basically any decent UNIX box running a current OS, log in, and have the option of dropping into a familiar, high-quality GUI. No one will be locked into it -- hell, even if you're stuck with GNOME, you still have a lot of options for window managers, desktop customization, application choice, etc.
Q: Why are these companies focusing on Linux as a "Windows killer"?
A: Because it is already a Windows killer. They want to ride the wave.
Q: And, up to this point, have any of the "Windows killer" features been GUI related?
A: No.
Q: What have they been related to?
A: Stability, performance, reliability, flexibility, scalability, cost and of course freedom.
Q: What are these same companies contributing to these "core values" of Linux?
A: With the partial exception of IBM, nothing.
Blah. I've seen a lot of "consortiums" and "joint ventures"--few (if any) of them produce anything of real value. Instead of hyping up how much "contributing to the community" they are going to do (jam tomorrow) why not just produce some code and release it (jam today)?
--
Linux MAPI Server!
http://www.openone.com/software/MailOne/
(Exchange Migration HOWTO coming soon)
I can't believe they chose GNOME when KDE is superior is every way.
Gnome is a bloated, memory hungry moster, while KDE actually has a negative memory footprint. That's right, when I run KDE, it magically *reduces* my memory load!
Gnome is also coded horribly. Miguel "buffer overflow" de Icaza realizes this - this is why he's getting all these other big companies (I.B.M., Compaq, Hewlett-Packard, Sun Microsystems , even Eazel) to come in and fix it for him. If coders were U.S. presidents, the KDE programmers would be Abraham Lincoln, and the Gnome coders would be Warren G. Harding.
Gnome is ugly. Especially the one's by Tigert (or "mspaint.exe", his alter-identity on IRC.) KDE is, on the other hand, so beautiful that everytime I log in, I collase to the floor in a state of blissful shock for the sheer beauty of my desktop.
So please, people, stop the madness - drop Gnome and start using KDE instead!
(P.S. I wonder how many people will take this seriously...?)
It's not illegal (or wrong) to control 100% of the market because you have a good (or popular) product. It is illegal (and wrong) to use your large market share to kill companies you don't like.
-- Erich
Slashdot reader since 1997
Im sure this will get the usual share of cliched responses, but lets' forget that for the moment. The interesting thing here is what the 'big boys' consider is needed for a commercially viable desktop-oriented Linux, why they're doing it, and what they're prepared to do about it.
You might not think you need a fancy GUI, especially if it looks like Windows, but there are an awful lot of consumers out there who just wantr to click on buttons, not memorise a new and unique argument set for every action they want to undertake. Hence an elegant GUI is a Good Thing for a certain group of users. A unified GUI is a Good Thing for developers, and thus commercial interests. A unified GUI with the features that simplify cross-porting of established software (ie drag and drop, unified colour printing, clipboarding, all that sorta stuff) that's a Good Thing for those developers tired of doing it all again from scratch. Whether KDE or Gnome is better is moot. For the desktop, those features can be incredibly useful, even if you can get along without them
Very little of Linux, per se, is desktop oriented. That means whatever solutions are developed, are probably pretty much cross-platform. For the other Unix vendors, even those competing against Linux for share, that makes assisting in the development of the solution also a Good Thing. If you're paranoid that Sun might hijack it, thats dumb. Gnome is still GPL'd, so they can't. But if they use it as a stepping stone toeards features they want, then that benefits us in the long term, if only to spur KDE to compete.
Whatever happens, this is a Good Thing. Don't let the zealots and bigots persuade you otherwise. Any expansion or improvement of the available feature set is a good thing, because Open Source is a Darwinian process. If its cruft, it'll fall by the wayside. And it doesnt matter if Sun et.c have selfish motives; the process is one they can only partake in; they can't, ultimately, control it.
Do we need another office suite? Maybe not, but is it a problem if we get one? At the end of the day, its all down to choice, and more available choices is never a bad thing. At least we do get the choice.
Pax,
White Rabbit +++ Divide by Cucumber Error ++
free experimental electronic music netlabel at www.viablehybrid.com
This is really it. The major corporate players are now USING open source to achieve that which their business models could not.
They are using it to break the Office monopoly. They want to be able, in a year's time, to go to a CTO and sell them on linux, GNOME, and StarOffice. With players like Sun, HP, and Dell, and the bottom line (the price), this strategy will work.
Once StarOffice makes inroads in Office Suites against MS Office, Microsoft will be forced to compete on quality and service with a free adversary that already kicks Microsoft's butt in service. Make no mistake about it. IF StarOffice meets corporate expectations, it will rapidly grow in its user base. As will linux as a desktop OS.
The open source model will have broken a very strong monopoly being maintained by anticompetitive tactics.
Expect lots and lots of FUD coming out of Redmond. This threatens their living more than the antitrust suit.
Never knock on Death's door:
The Anti-Blog
Q: And, up to this point, have any of the "Windows killer" features been GUI related?
A: No.
Q: Is Linux hindered in killing Windows in mainstream markets by its lack of simple, high-quality GUI features?
A: Yes.
Q: Are companies used to producing commercial software likely to be able to help Linux develop such features?
A: Yes.
Q: Then why complain?
- Michael Cohn
-----
Go ahead, blame me... I voted for Nader!