Mozilla Foundation Holds Future Strategy Talks with GNOME FoundationMonday April 26th, 2004Slashdot is reporting that the Mozilla Foundation has held talks with the GNOME Foundation to discuss future collaboration. The minutes of the meeting, which took place via teleconference last Wednesday, reveal that the discussions covered the integration of Mozilla with GNOME, strategies to combat Microsoft Longhorn technologies (especially XAML), duplication of technology between the two projects and the default Linux browser. The meeting follows Brendan Eich's earlier comments about how the Mozilla project can move forward with GNOME. Exactly how closely the two projects will work together is still being debated. The Mozilla Foundation is keen to form an official alliance with GNOME, while founder of the Galeon and Epiphany GNOME browser projects Marco Pesenti Gritti is pushing for a full merger. Galeon and Epiphany, though both based on Mozilla, hardly get any coverage on Mozillazine. Firebird gets all the coverage. Konqueror, KDE's file browser/web browser is widely regarded as a great file manager, and a usable, though not great web browser by most users. I really liked Konqueror, but even on a 166Mhz computer found Firebird faster, and more able to take advantage of the fonts on my system when rendering web pages. I fully agree that a combined product would be good, esp. for Firebird's evolution--FTP, WebDAV, etc. But I like Linux specifically because the default file manager does filemanagement better than the explorer "browser" on windows. (I can't open an image file over a network because Windows tries to grab it for its left-hand column. | pretty picture (requires resources on network)| photo a..b.. c..d (tiff files imagine..) 5 minutes to load photo because I selected the wrong one.. or worse.. am moving through with arrow keys. So, a full merger is not a good idea. If GNOME supports XUL and extensions, that's great. (so if someone wanted the feature I described, they could extend the file manager.) Otherwise, no thanks. *unless* the entire interface was xulized.. like OEone. --Sam I think it's a really bad idea to integrate any mozilla (Fx, TB) too much into a desktop (gnome or kde). I for myself am already really bothered that Fx has the "Save/Cancel" buttons reversed while running in linux. I want it to behave the same way in windows and linux. Mozilla should stay really platform independent. I have no problem in Mozilla requiring gtk. But I would have a problem if mozilla would require and gnome libraries to run. There is nore than gnome, and with the huge steps that kde has made, I wonder if it's a good idea at all to base Mozilla on gtk... To the first poster: konqueror, and opera, both are faster than Fx on Linux. > To the first poster: konqueror, and opera, both are faster than Fx on Linux. Can you back that up with actual numbers? If so, I'd be interested to see them. >> To the first poster: konqueror, and opera, both are faster than Fx on Linux. > >Can you back that up with actual numbers? If so, I'd be interested to see them. No... but it feels like the rendering is more snappy, and opera and konq feel much lighter on tab-switching. Of course this is a subjective perception. In windows I use Fx exclusively, and will never stop to do so. I did a series of tests and I found page load times to be practically identical between Firefox and Konqueror (measured on the basis of when the progress bars got to 100%). Konqueror had a slight edge on some sites (the BBC main page) whereas Firefox had an edge on others (redhat.com), but the difference was no more than a second or so. Where Konqueror does have definite speed benefits is in the initial start-up time, and also when the browser hasn't been used for awhile - if I haven't used Firefox for an extended period of time, maybe half an hour or more, switching to Firefox takes forever (it takes about as long to revive as it does to start). By contrast, Konqueror is live within a second or so. Switching to "stale" tabs also seems faster in Konqueror. "I think it's a really bad idea to integrate any mozilla (Fx, TB) too much into a desktop (gnome or kde). I for myself am already really bothered that Fx has the "Save/Cancel" buttons reversed while running in linux. I want it to behave the same way in windows and linux." Ya, and now we're flipped with respect to KDE too... "Mozilla should stay really platform independent. I have no problem in Mozilla requiring gtk. But I would have a problem if mozilla would require and gnome libraries to run. There is nore than gnome, and with the huge steps that kde has made, I wonder if it's a good idea at all to base Mozilla on gtk..." I agree - I'm at a loss to understand Mozilla's apparent fascination with GNOME and near disdain for anything QT/KDE (cue the toss of QT Mozilla). But on the other hand, there is an opportunity here for greater integration with both desktop environments if done correctly. Right now Mozilla looks more like a GTK app but it doesn't actually integrate at all well - the file picker is a GTK-style file picker but it's part of Mozilla, unlike in Windows where Mozilla uses the native filepicker. If Mozilla heads towards GNOME integration the code will have to be changed to allow for use of native GNOME tools, like the file picker. Well if that's done then the same can be done for KDE integration *at the same time* by way of qtgtk library developed by KDE: http://developer.kde.org/documentation/tutorials/qtgtk/main.html Mozilla could then be compiled with some KDE support by those who wanted it, and distributions like Debian and Red Hat could issue 2 or 3 versions of Mozilla depending on the user's preferred desktop environment. And an RFE I filed for just that with Firefox: http://bugzilla.mozilla.org/show_bug.cgi?id=237288 The Qt port was killed because it had bitrotted, not because of any sort of distain for Qt/KDE (interesting question - how does the Qt license affect Mozilla? Anything that links to Qt must AFAICT be GPL-compatible or must have a special license fromTrolltech - the MPL has been designed to avoid such restrictions. Although I guess it's not a problem, I would be interested to know what would happen if a company released a closed source product atop a Qt port of Mozilla - that would be OK from a MPL point of view but would they be viiolating the QT license?). Qt integration will happen just as soon as someone cares enough to write the code to make it happen (like, I guess the OS/2 port would be dead by now if IBM weren't spending a lot of money ensuring it survives). The QtGTK thing looks interesting but it will *only* work for the native dialogs (which aren't even implemented fore gnome yet). If you want proper integration, you'll need lots of XUL and nsITheme changes. Personally, I would consider this a very low priority in the grand scheme of things but thehn I'm not producing any code anyway, so I'm not so sure my opinion is relevent :) #21 Re: Re: Please don't tie any Mozilla product to anby AlexBishop Tuesday April 27th, 2004 12:56 PM "Right now Mozilla looks more like a GTK app but it doesn't actually integrate at all well - the file picker is a GTK-style file picker but it's part of Mozilla, unlike in Windows where Mozilla uses the native filepicker." The only reason for that is because, prior to GNOME 2.6, the GTK filepicker sucked like a tornado. Alex After taking a look at the preview version of Longhorn for beta testers, it makes a lot of sense to me that possibly Gnome, Mozilla and KDE work together to come up with a similar model that reduces redundant code and establishes API's for inter-application communication. I personally like KDE over Gnome since I find its design and implementation of modules a littler cleaner the KDE UI out of the boxes just has a cooler feel to it that surpasses OS X and Windows XP, in my opinion. I'm a huge fan of Mozilla (App.Suite on Windows and Linux) Firefox on MacOS. I don't see any details about the others in this message. I personally like the idea of XUL on the (Linux) desktop, but isn't it possible to get this working in BOTH GNOME and KDE??? I that case the KDE users can use Mozilla and XUL without the need to switch to an alien desktop environment. I know that KDE already has Konquerer, but this shouldn't be a problem in my opinion. Just colaborate with each other instead of fighting each other. And whats the status of Windows and MacOS? I don't see any mention of those OS'es in this message, when the Mozilla Foundation is merged with the GNOME foundation this OS'es still need the Mozilla applications... I think they're just trying to combine effords on some technologies to work against MS's longhorn. That doesn't mean they're not going to support other OS. Some technologies are cross platform. Even GDK itself can be used on Windows. What I think they are getting at it using the Mozilla rendering engine (gekko) to render the Gnome Desktop using XUL description language. I don't think it means using Gnome libraries to help write the Mozilla Application Suit thus tying Mozilla to Gnome. I think it means the opposite - using Mozilla libraries to write Gnome's user interfaces. Sounds to me a bit like what OEOne have been doing. Actually, I've a thought about using XUL (or something like XUL) to descripe interface for both GNOME and KDE, then that means we can take advantage of the cross platform nature of XUL itself to decrease the variation between KDE and GNOME, at least in a GUI programmer's perspective? With so much engergy going into a potential alliance with gnome (and for good reasons), I hope mozilla foundation's commitment to cross platform support will be not de-emphasized. If the target is xaml/longhorn, using a platform dependent alternative ain't going to be enough. I think the fight on the desktop is not going to be longhorn vs. linux, we know which is going to win, unfortunately. The fight will more likely be win 9x/nt/2k/xp + linux vs. longhorn. There are 300 million+ 9x/nt/2k/xp desktops out there, and many companies don't necessarily like to "upgrade" to longhorn if they can. Providing an alternative to xaml that works on all these platforms will definately help further slow down the move to longhorn. And when the day comes that they have to move off 9x/nt/2k/xp for whatever reason, knowing that what they have on their existing platform (xul+svg+xbl+xpcom, etc.) will carry on working on linux should, hopefully, help some make the right decision. George Cao XP is one of Mozilla's key strengths, and we're in no hurry to throw that away, or even diminish it. Wow, a developer that actually responds to user feedback. Ben, you *rock* man. I downloaded a nightly of Firefox and checked out the new extension UI. Very nice work. I appreciate your dedication and passion to this project. Thanks. --Buff Well said. I agree, Ben desirves some huge respect for what he has done to Firefox amongst many complaints/fights/etc. he's kept strong and done what he thinks is best for Firefox. I or others might not agree with some UI decisions or other aspects, but that is not relevent. What is, is that he has "the vision" and has kept track to it despite what others think. He is sure to make an incredible product out of Firefox. You are right, Ben ROCKS! The people in power are definately strongly against a single platform solution. See this http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&oe=UTF-8&threadm=4074FEFC.8000004%40meer.net&rnum=1&prev=/groups%3Fhl%3Den%26lr%3D%26ie%3DUTF-8%26oe%3DUTF-8%26selm%3D4074FEFC.8000004%2540meer.net group of post from the newsgroups. The question for me is, what does a tight co-operation with Gnome mean for other Mozilla platforms. Would there be capabilities which are present in conjunction with Gnome only? Would there be any hidden APIs (probably not possible due to the open source nature of both projects) which expose those additional capabilities? If not, what would be the benefits? More development resources for Mozilla and friends? If the target is MS and Longhorn I believe there needs to be the KDE team heavily involved. Who knows which Linux desktop will prevail, if any? Would Gnome and KDE continue to co-exist? Referring to some Novell comments, there might be a movement towards KDE. I would definitely not like a merger with Gnome since it affects credibility of Mozilla as XP application and even as working with both under Linux, Gnome and KDE. This does not necessarily mean any technical problems but it is a problem of perception. Be careful about that... Two objections: 1. As Windows user I want Mozilla.org to be as focused on Windows versions as they already are. Merging with Linux desktop vendor will not be helpful for this - I guess. 2. GNOME's GTK/GTK+ are HORRIBLE in terms of Windows GUI standarts and usability (BTW QT is better, but also far from perfect). Even with special "Windows-emulation" (Windows "emulation" on Windows??? Doh!!!) themes GTK sucks (and not only visually, but also in the way how controls work, how windows are painted, etc). But honestly - what can you expect from Linux desktop makers? The only "so-so" usable Linux GUIs are the ones that mimic Windows or MacOS. Please do read the links, not just the blurbs. There's a lot more to GNOME than the UI. I know. But the GUI is bad enough for me to rant anyway (unless they will not use it). What? I'm pretty sure that no one is suggestting using GTK on windows... I hope! These Linux fanatics, you'll never know ;-) Well, FWIW, I've never had a problem with GTK on linux (even the old filepicker's alright once you realise it does tab completion *really* nicely). I agree it's out of place on windows though :) GTK on Linux is a lot nicer than GTK on Windows. GNOME is extremely usable if you use it in its native environment. In fact, many people (including me) find it much more intuitive and visually pleasing than Windows. Evaluating GTK apps run on Windows is like evaluating Windows apps run on Linux. It can be done, but it's not pretty. AOL/Netscape are currently reworking their internet strategy. Presently Netscape is hiring programers--web and software and plan to release an updated version of the Netscape suite. The news report (can't find the link at the moment) made it seem with the Netscape hiring of programers that the next version of Netscape would be more than a simple repackaging of Mozilla 1.x (1.7?). Maybe AOL has a strategy up its sleeve. Remember their agreement with MS to use IE gave them the right to use it, but did not state that AOL had to use IE exclusively or even at AOL. AOL, under its agreement with MS could drop IE at anytime. Maybe with the upcomming Longhorn release and the tightening of IE with the OS even more, this integration takes some control of how AOL can market themselves on the net. Longhorn may dictact which media formats can/cannot be used and builds in an MS search engine which must be used. Maybe AOL sees the writing on the wall. Maybe AOL is afraid that if they stay with IE when longhorn is released AOL will feel as if they have been run over with a steamroller by MS. Whatever the motivations behind AOL/Netscape's renewed interest in the suite, they are return to the suite and are hiring programers to work on it. With Netscape once again behind the suite, I don't see the suite going away any time soon. |