The Rise of Red Hat; The Demise of Slackware? My regular readers know that I have been a long time supporter of Patrick Volkerding's popular Slackware Linux distribution. So it may surprise some of you (as it has surprised me) that I now believe Slackware will soon taper off in popularity and use much like its predecessor, SLS. Actually the writing has been on the wall for some time now. ftpd, nfsd, rpc.portmapper are known security holes in Slackware (improved versions have been available for ages, but never find their way into the latest Slackware). Slackware still uses old or buggy versions of init (recent alpha kernels crash due to bugs in the stock Slackware init), talk, and countless others. It is almost impossible to find the source code (and any modifications made to the original source) for most of the included packages. It's virtually impossible to upgrade the basic system tools without breaking the system in fundamental ways. The list of problems goes on and on. Why did such a great distribution fail? I think the problem is that Slackware is no longer well maintained. In it's heyday (from Fall '93 - Spring '94), Volkerding released new versions of Slackware quite often (sometimes monthly); he actively participated in the news groups; and it was relatively bug free. In short, Slackware was well supported by it's maintainer and the Linux community. But when version 1.0 of the kernel was released in March of 1994, Slackware started slacking off. Since Patrick's work was of such high quality, few noticed or cared (hell, we had working systems!). But over the past year as Slackware lagged further and further behind the cutting edge, Red Hat Commercial Linux and Debian have been organizing their coup. Red Hat is now the distribution to beat. At least that's what Matt Welsh implied at his Gene's Books lecture. After re-reading George Evans' review of the first beta release of Red Hat (Data Bus, October 1994, page 30), I'm doubly surprised that Red Hat has managed to win such praise from a widely respected member of the Linux community like Mr. Welsh. It seems that Red Hat must have worked very hard this past year. Red Hat is available from several ftp sites (ftp://ftp.pht.com//pub/linux/redhat, ftp://ftp.cc.gatech.edu//pub/linux/distributions/redhat and others). The 2.0 release of Red Hat seems particularly enticing: it's 100% ELF, any and every piece of the system is upgradeable without fear of breaking your system, and it includes all the latest software (at least this is what their ad in comp.os.linux.announce said). During his lecture Matt Welsh emphasized how Red Hat makes it easy to learn the origin of each package/subsystem and what changes/enhancements the Red Hat developers have made. So it's easy to figure out where to go if there are any problems you need to fix (this particular project is extremely difficult under Slackware). I hope to install Red Hat on a newly ordered 1 Gig drive next weekend. And I will also install Debian (because it's a volunteer effort that I might be able to contribute to!). At next month's Unix SIG meeting, I will discuss Linux Distributions in more detail. In the mean time, if you're thinking about buying a CD, you probably want to make sure it includes Slackware 2.3 (it's still easy to install and for single-user home systems is not too bad), Red Hat 2.0, and Debian 0.93 (to be released Monday September 25). UNIX SIG Notes: We need more UNIX contributors to the Data Bus! I can't do it all by myself --- please help out, guys and gals. There is increasing interest in forming a Philadelphia Linux User's Group. Should this be part of PACS? A separate entity? Who wants to organize it? My only comment: I'll show up to the meetings whenever and wherever they are held. Next month's topic: "Getting Started with Linux: Installing a Distribution." 11 am - 1 pm in room 242.