Mozilla Drops glibc 2.0 Support - For NowTuesday August 10th, 1999Andrew Niese writes in with this news: "Daniel 'leaf' Nunes, mozilla.org release engineer, today brought the bad news that the Mozilla project will cease focusing development efforts on glibc versions lower than 2.1. glibc, the GNU C compiler runtime library, appears to have a bug in it that often blocks the progress of the Mozilla project on some Unix platforms, and 'seems to have no workaround from the browser side,' Nunes says. A BugZilla thread has been started to help figure out the whole situation. Worst-case scenario, builds of Seamonkey (the Mozilla.org 5.0 browser project) could eventually become unsupported on distributions that use the older version of glibc, such as RedHat 5.2. Many Netscape engineers agree that the final 5.0 release of Seamonkey should support glibc 2.0, but for now the future looks cloudy. Leaf tells us that 'there are still top men working on an eventual solution.'" Well, if it will decrease the amount of time it takes to get to market, I'm all for it. Most of the older glibc installations will be servers IMHO anyway. So, it won't be that big of a deal for the machines that are likely to have someone sitting at the console everyday anyway...or the average linux user. If mozilla only supports glibc 2.1 then for glibc2.0 systems you can have a statically linked mozilla. or include the libraries in with the rest of the distribution. Of course I wouldn't be saying that if I had glibc 2.0 but then again if I did have 2.0 I would upgrade. I'm not a linux tech-head, I do use it, but am not all that knowledgeable about it. I do know however that after I installed RH5.2, I needed to upgrade a lot of the libraries before I could install much. I wouldn't hesitate to upgrade a library if it allowed me to run Moz. I'm sure most people would be the same, unless there's some technical reason why you can't upgrade... (?) I agree. glibc 2.1 has lots of improvements over 2.0 anyway, and anything that could encourage users to upgrade is good. Upgrading to glibc 2.1 was completely painless on my system. i'm in complete agreement. most linux distros (where "most" refers to redhat, debian, and probably slackware although i know less about it) make it surprisingly painless to upgrade things like this. from both a software engineering perspective, and a business model perspective, it is obviously desirable to support as many platforms out of the box as possible. however, i wouldn't sweat too hard over it, since the core of browser users on linux are either going to be gearheads (who will know how to deal with this sort of thing) or less savvy users of newer distros (which will already have glibc2.1 installed). Top men. But _who_? Top... men... Then why even bother to look for workarounds to support glibc 2.0? Backwards bug compatibility is half the problem with Windows nowadays? I understand the desire to be backward compatible, but sometimes you just have to draw the line... Don't waste effort on that glibc mess. If someone _really_ feels the need to make it work, let 'em. Otherwise, just distribute a 'glibc2.0' version (a.k.a static build) and be done with it. If it means getting a USABLE milestone -- by all means ignore 2.0. Check the report in Linux Today about Burlington Coat Factory's Linux rollout. Biggest concern with regard to desktops? That people start using IE without a comparable Netscape browser available, thus throwing a wrench into Linux on desktop plans. |