felicitas

Updated to Firefox 7 final, which now runs at about the same speed as patched Firefox 5/6 on Win98. All old issues remain, especially the bookmarks-/history bug. At least there is a workaround now. Thanks Steven . By the way. Do you guys have any speed issues with patched Firefox 5+? Scrolling looks a bit sluggish to me, especially in comparison to Opera 7.54u2 on 98 or Firefox 5+/Win7. I'd love to hear some reports of running Firefox on real machines. I'm mostly using VMWare at the moment.

Updated to Firefox 6, see first post for details I think I've found the reason for this bug. Expect some information in the next few days on KernelEx' bugtracker. I don't have that much time at the moment, sorry. Short version: UnlockFileEx unimplemented(easy to fix), region locks affect file mappings to everyone, even the creator(in case anyone wants to take a look at it). I haven't looked at the bookmark and history issues yet, but I can absolutely confirm them. Maybe there will be also a Seamonkey 2.2 build for 9x soon if no unexpected problems show up.

Hi everyone I think KernelEx should move towards some kind of community approach, in contrast to the current "The maintainer does everything" one. KernelEx reached a state where many vital things already work. We're able to use modern browsers, modern media players, and various other things. There are some bugs, and surely there are enough applications that do not work. But KernelEx basically does its job. Therefore, there isn't as much motivation left as there was back in the days. There is almost always the possibility to change some settings or use another program to get what you want. In some situations, it even got easier to modify the application itself than improving KernelEx, as with Firefox 4 and higher. That doesn't mean that nobody should work on KernelEx anymore. In fact, the complete opposite is the case. But it has to get much easier to start hacking. Setting up the required tools to build KernelEx isn't easy enough I think. Many old-school developers may still have VS 6 somewhere in their shelves, but most new potential developers have not. Getting the right PSDK 2003 version may also get somewhat tricky. I'm not completely sure about how we can fix that though. Also, some place where users can easily report incompatibilities, give some hints(aceman's report for Firefox was very helpful) and just communicate with developers would be great. There is such a place already, KernelEx' SF bugtracker, but it's not widely used for some reason. That kind of blurs the future perspective. What feature is missing? What application does everybody want to see working after the next release? I don't know. And finally, the biggest problem at the moment: We need much more documentation about KernelEx' internals. It's absolutely possible to get to the bottom if you read the code carefully, and maybe give your debugger a shot sometimes. But it's just unnecessary. It's so much work for new developers to get the big picture about how KernelEx is working, and where they have to change what to even do something without busting their whole system. It takes at least 1-2 days even for skilled developers to get used to the code. That's a lot of time that has to be spent before they can even do anything about the actual bug. It would be so much easier if there was a simple document somewhere in KernelEx' sourcecode archive explaining the general concepts and ideas, together with some details about the concrete implementation, as you suggested. Once we fix at least 2 and 3, I think many part-time developers will gather around KernelEx and do many small changes. And that's the future perspective in my opinion. I can't imagine anyone who dedicates so much time to KernelEx as you did in the past(though I'm very happy you did!). But we have to realize that this may slow down development. It will keep the project alive at least. What do you think? What does the community think? Is it really too hard for all those of you how want to participate to actually do so? That doesn't necessarily mean writing code, but also reporting bugs and proposing new features/applications. Or maybe there is a skilled developer out there who is willing to continue KernelEx in Xeno's fashion? Raise your voices. They will keep a great project alive

Firefox 5.0 is the latest release for Windows. 5.0.1 just contains some fixes for Mac OS X users, so using it wouldn't make any difference. See the changelog at http://www.mozilla.com/en-US/firefox/5.0.1/releasenotes/. There isn't any reason for not updating to 5.0.1 though. I guess I'll update it in the next few days. In general, I will try to update the binaries to the latest stable release as soon as possible, which is 5/5.0.1 at the moment. 6.0 will be released in a 2-3 weeks if I remember correctly. I would prefer fixing KernelEx much more, enabling us to use official builds again. I'm digging into the code right now, to some degree at least. Hunting down the exact reason isn't as easy. Let's see how it turns out.

Sure. I've added Megaupload, just have a look at the first post. Hope that works better. To be honest, I've never tried importing bookmarks =D. May be another KernelEx bug, or maybe something is strange with my version. Maybe I'll get some time to take a look at this.

What's this? This is a build of Firefox 7.0 optimized for running on Windows 98 and higher using KernelEx 4.5.1. Why not use the regular build? Mozilla's official version, or rather their integrated SQLite3, triggers a bug in KernelEx when using Write-Ahead Logging, which causes massive slowdowns, rendering Firefox useless. This version features a workaround for that problem. It just disables use of WAL. This bug was discovered and reported by aceman, see http://sourceforge.net/tracker/?func=detail&aid=3284681&group_id=255038&atid=1126916. Legal As this is a modification of Firefox, it is licensed under the very same terms, which means MPL(Mozilla Public License)1.1/GPLv2, LGPLv2.1. These licenses basically allow you to use, modify and share the software freely, as long as you also redistribute any source code modifications. See the respective licenses for more details. Download Aurora/Firefox 7 http://www.mediafire.com/?w7fsncgptmtv5e7 MD5: f574e3100861ea51c7dee55a22e0a479 SHA1: a7183c4e4609856721f0daad6a64a7c5a6dc24fb Aurora/Firefox 6 http://www.mediafire.com/?tqqh85lmyi8cmqi MD5: f07db34d8dce1c033fccd2e259070db5 SHA1: ad3ba4cad8861f88f1f30db3877230a22c84ed88 Aurora/Firefox 5 Uploaded.to: http://ul.to/o72uofpc Megaupload: http://www.megaupload.com/?d=YTLXAPE7 MD5: 05491aad4361f2c2a729de911e663d52 SHA1: 4f767ebb19237dd8e35bf5f712e7a0c2660c4524 In case you want to compile Firefox yourself, download the following patch. http://www.msfn.org/board/index.php?app=core&module=attach§ion=attach&attach_id=32876 It contains all changes made. Applies to Firefox 5.0 source. Feel free to mirror everything. All binaries provided here are compiled using Microsoft Visual C++ 2008 SP1. Build instructions(https://developer.mozilla.org/en/Build_Documentation). You will need a recent version of Windows to compile it. The binaries above were created on Windows 7 SP1. Tested on Microsoft Windows 98 FE/Gold(4.10.1998) and KernelEx 4.5.1. Remaining bugs 1. Bookmarks and History do not work at all. This will wipe out all your bookmarks of all older Firefox versions! Steven W found a way around that using PlainOldFavorites Addon. See . 2. This build was compiled using Microsoft Visual C++ 2008 SP1, which also means you will need the corresponding runtime files to run it. Just download Visual C 2008 SP1 runtime, and extract msvcr90.dll and msvcp90.dll using WinRAR or something similar. I will maybe download VS 2005 Express some day… 3. Menu fonts look somewhat strange. Is there a missing font? Or maybe it’s related to that new font rendering stuff that everyone was complaining about? I don’t know. 4. WebGL might not work. I’m currently unable to test 3D functionality. Firefox was compiled without DirectX SDK being installed though. Why Aurora? While Firefox itself is open-source, the name “Firefox”, the Firefox logo and related stuff are not. Mozilla prohibits the use of the name “Firefox” as soon as the source code gets changed to ensure consistency for users. Aurora is just the default name for unofficial builds. See http://www.mozilla.org/foundation/licensing.html for more information. Will this fix get into official builds? Hopefully not. It is just a workaround, or a hack. The problem does not lie in Firefox or SQLite, but rather KernelEx. We should fix KernelEx to finally get rid of that bug. It might also affect other applications. The reason for this build is simple: It’s just easier in this case to modify Firefox than debugging KernelEx. firefox-5.0_win98sqliteworkaround.patch.txt