actuality.log


Tuesday, November the 4th, 2003

A couple of people have asked me about this and I’ve essentially typed up the same thing (for two moderately different questions). I paste it here because I’m a lazy slacker and I don’t see why a good 10 minutes of typing should be wasted, if people really care about my thoughts on the issue.

I just came to know about Redhat’s plan to stop support for the 7.x – 9 line and start commercial enterprise Linux! I trust you must be knowing about the Fedora project too. I have a question?

Given the uncertainity that distros typically exhibit (are exhibiting), does this mean LFS is the future?

Firstly, they are not “starting” their EL line of products just now. They have had it for longer than you know. That, and their other services like training and so on is what makes them money.

There will always be uncertainity about distributions until they figure out how to make money, essentially packaging someone else’s work. And there are other distributions like Debian whose fate is not “uncertain”. RedHat is just trying to switch to the development model followed by them. (Again, personally, I feel this might do a lot of good to their product. I know people who maintain rpms for software that is not part of the standard distribution. Now, people like them will have a place to officially “put in” their work, and many more will get to enjoy it.)

The community wants frequently updated and freely downloadable software. They don’t (I don’t) necessarily care if some odd package is broken. The companies want slowly changing, robust software, and don’t care if it’s not bleeding edge.

For right now anyway, Fedora IS RedHat. Just unsupported. (But it’s not like yout current free download of RedHat was “supported” by them, so it is not an issue.) The only thing different will be the lack of free up2date and other such services that lose them money (bandwidth, having to backport and maintain all this for free).

Like I said, it is worth their while only if their “commercial” OS and their “consumer” OS are close to each other. They will be doing it for the commercial one anyway so they can release patches to the public. Now the two products are diverging because the consumer one is being released more often, and more radical changes are going into each release.

So they let the “public who want it that way” handle that and support it. While they work on the slower (older) product for making money, which is what companies try to do.

I don’t think it’s necessary for everyone to abandon them and rush to something like LFS. Firstly, that’s fun as a learning tool, but not terribly productive unless your job or aim of the exercise IS installing the system. Secondly, like I started, there are a lot of other distributions, and there always will be as long as people tend to think they can make money off packaging someone else’s work.

And personally, I’ve seen Fedora and it’s very polished. I don’t see the need to jump ship yet. Long story short, they might be behaving “evilly” now. But they have done a lot more good than I care to detail, and until all that built up positive karma is exhausted, I am not going to “abandon” the distribution.

Yes, it’s all just religion, and I think they know it and are counting on it.

This is a printer-friendly version of the journal entry “My thoughts on Fedora” from actuality.log. Visit http://emphaticallystatic.org/earlier/my-thoughts-on-fedora/ to read the original entry and follow any responses to it.
»

3 Responses to “My thoughts on Fedora”

  1. Puli says:

    ok. just one clarification, what do you mean by “package is broken”? broken as in “not in development” or broken as in “buggy”?

  2. wahgnube says:

    (I normally use the work “broken” for code that’s suddenly failing to build during recent checkouts. But that’s not the sense I meant it in here.)

    Firstly, inherently most bleeding edge stuff is buggy to a certain extent, so I am not talking about that sort of buggy. And we are not talking about the development of the software over here, just the packaging (into RPM or what not). If that stops, the project is effectively dead.

    I meant broken, here, in the sense of “messed up” packages made by people who didn’t necessarily go through the same level of quality control (I hope) RedHat goes through (in the sense of, even if the original software was fine) before releasing a package.

    And by “messed up”, I mean anything from packages not being updated fast enough to match fairly recent software releases, and things that are plain.. broken. Fail to install, mess up your system on install and what not.

  3. Puli says:

    that’s more clear… thanks!


1 people conned into wasting their bandwidth.