Bringing you the latest news from the Linux World.
Dedicated to keeping Linux users up-to-date, with concise news for all interests
Linux in the news
All in one big page
Here is the permanent site for this page.
Linux-Mandrake is on a roll. The announcement for the very first version (5.1) of Linux-Mandrake came out just over a year ago - July 23, 1998. It was, at the beginning, simply a rework of Red Hat 5.1 with the KDE desktop integrated into it. People liked the result, and Linux-Mandrake began to attract a sizeable user base.
Consider, one year later, some events from the last month:
Obviously Linux-Mandrake is going somewhere. They now have twenty employees, offices in numerous countries, and the ability to fund open-source development projects. Even a year ago, the distribution market was looking crowded; how can it be that such a newcomer can have such success?
Mandrake is a clear example of both the benefits and the pitfalls of being in the open source software business. MandrakeSoft was able to start with a complete distribution - Red Hat 5.1 - and create a new, marketable product with some comparatively minor additions. There is an old joke that, while most scientific fields make progress by standing on the shoulders of those who came before, computer scientists stand on each others' feet. Thanks to free software, MandrakeSoft was able to stand on Red Hat's shoulders.
Was this fair to Red Hat, which might not have wanted to lend out its shoulders in this manner? In a sense, it doesn't matter. Those are the rules of the game that Red Hat chose to play. Remember also that Red Hat has always been free to take MandrakeSoft's additions and fold them back into their own distribution; if they have not taken advantage of this resource, they certainly can not complain if Linux-Mandrake starts to take some sales from them. (And, in fact, if Red Hat has complained we certainly have not seen it).
MandrakeSoft's real advantage would appear to be a strong emphasis on ease of use and integration of the desktop tools. By creating a distribution that people want to use, they have created a market for themselves. Linux-Mandrake is successful because it has added real value.
Linux-Mandrake is now moving away from direct use of the Red Hat code base; they have sufficient development resources that they can do things their way. They intend, however, to maintain their Red Hat compatibility - RPM packages for Red Hat will always install on Mandrake as well. This is a smart move, which will help to guarantee the availability of large amounts of contributed software for their system.
Please see our interview with MandrakeSoft's Gael Duval for details on the above and much more: MandrakeSoft's future plans, and the origin of the name.
Caldera OpenLinux 2.2 and 2.3, and Red Hat 5.2 and 6.0 are year-2000 compliant, according to an independent analysis done by Key Labs (Caldera) and The Software Laboratory Limited (Red Hat). Both tests looked at a variety of commands to insure their proper behavior in the next century. While a few small bugs were turned up, they had little to do with the year 2000 itself. Both OpenLinux and Red Hat have a clean bill of Y2K health.
One could certainly find faults with the methodology of these surveys (the full reports are available from KeyLabs' Caldera page and Red Hat's Y2K page). There was no complete survey of the code done. Other possibly time-sensitive parts of the system (PAM modules, PostgreSQL, sendmail, etc.) were not tested (the Caldera survey was rather more comprehensive than Red Hat's in this regard). Testing of applications was done, with little attention paid to the kernel. They are far from a complete survey of the Y2K readiness of Linux.
Still, it is a step beyond anything we have had before. The Linux community has, as a whole, not been all that concerned about the year 2000 problem; Y2K is a problem that other systems have. And, to a great extent, that is true. The internal Unix time format has no Y2K problem (though 2038 is another story). But that does not mean that libraries and applications have not introduced problems of their own. There must be at least one lurking out there somewhere still.
So it is good to see that a serious attempt has been made to insure that things will keep working after the end of this year. Even with a system as good as Linux, the absolute absence of Y2K problems can not simply be assumed. (See also: SuSE's Y2K page, Debian's Y2K page, and the GNU software Y2K page. We were unable to find Y2K pages for Linux-Mandrake, or TurboLinux; MandrakeSoft tells us that they have a Y2K study underway currently. Slackware has a brief mention on its FAQ page saying "The responsibility for ensuring that Linux is y2k compliant enough rests with the end user.").
Suggestions for the future of the Linux Documentation Project. Matt Welsh, co-founder of the Linux Documentation Project, has dropped us a notewith his thoughts on where the project should go from here. "I think that the only thing the LDP needs to do to get on track is to retain the essential structure it has had for the last five years. Making things any more complicated will only raise the barrier of entry to new authors, which will eventually cause the project to die out." Worth a read.
Linux fragmenting? No way, says Eric Raymond. "Instead, [Linux is] cheerfully absorbing its competition. And the fact that it is `absorbing' rather than `destroying' is key; vendors are belatedly figuring out that the value proposition in the OS business doesn't really depend on code secrecy at all, but instead hinges on smarts and service and features and responsiveness."
This Week's LWN was brought to you by:
August 26, 1999