-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Alejandro Forero Cuervo wrote: ...
I believe having a lot of packages (without allowing redundant packages) would be good, even if some of them are not as properly maintained as they could be. I would rather have 3 ISOs of very properly maintained software + 4 ISOs of averagely maintained packages + 4 ISOs of sparringly maintained software than just 3 ISOs of very properly maintained software. I feel that allowing relatively unused
I wonder why you are mentioning ISOs ? I think the top priority should be to discuss how we could integrate 3rd party package repositories (like Packman, James Ogley's, various suser-* on gwdg.de, mine, ...) into the distribution, /not/ into ISOs but e.g. offer them as a choice list of installation sources in YaST2, aggregate information about them in the opensuse.org site, through a common build infrastructure, common policies, etc... I think that putting them into ISOs is definately not a good idea, as that isn't flexible enough. Furthermore, the added value of our 3rd party repositories is two-fold: 1) offer packages that are not included in SUSE Linux (although it already includes a lot of packages) 2) offer newer releases of packages that are included in SUSE Linux, as SUSE/Novell's policy is to stick with the version that's shipped with a release and only provide security fixes (that are usually backported) Although openSUSE is the effort to create a strong community around the SUSE Linux distribution and that we're all looking forward to work together on that, the ISOs and the boxed set are still "SUSE Linux" (OSS or not) and that's Novell/SUSE's reputation. Like Scott already said, Novell should not have to be liable (in terms of support) or have its name put on the packages made by others. Besides that, as you have read the threads about it, I suppose that you also noticed the complexity involved with how to best implement that process. We still have to discuss a lot to get there, we'll eventually do, and then maybe one could think about a process to integrate some of those packages into SUSE Linux. But that decision must be left to Novell alone. Even if you name the ISOs "unsupported", most people will say that SUSE Linux s**ks because there were broken packages on it. Let's first try to find the best ways of integrating 3rd party repositories.
packages into the distribution would be benefical, at least better than not including them at all. I don't see any reasons why the quality of the "3 ISOs of very properly maintained software" would diminish by allowing the rest of the packages into the distribution. In this case, if a package is poorly maintained, anyone should be allowed to contribute and help improve its quality.
Sure. We can do that with 3rd party repositories, but not with ISOs.
Oh, and we could use tools similar to Debian's popularity contest to decide how to place our RPMs in our ISOs (probably marking the less important ISOs as "additional" or "optional").
No, you can't compare Debian with Novell here. The decision of what makes it into the SUSE Linux distribution should (and shall) remain with Novell. Is Debian offering installation support ? They're not.
Certainly, I'm not advocating letting anyone put up random crap and making it part of the distribution! I think there are many alternatives less extreme than the "Novell remains in control" approach.
Novell remains in control of the SUSE Linux distribution. That's fine and has already been discussed a little. Let them do (almost) whatever it takes to keep on providing us with the solid distribution SUSE Linux has always been. And let us as a community build (almost) whatever we want on top of and around that. You could make your /own/ distribution based on SUSE Linux OSS + integrate packages from other repositories (like Packman or mine) into ISOs. That's fine, nothing wrong with that.
Althoug I use SLES, SUSE Pro and NLD almost every day, I've been a long time Debian user: my perception is that most of their packages are of *very* *good* quality (it should be said that I have very little familiarity with Fedora). In their case, allowing anyone to create packages that are official part of the distribution has most certainly not decreased their overall quality.
It's true that the Debian packages are very high quality. But AFAIK that's because they're doing the exact opposite of what you're saying: they have *official maintainers* that are in charge of packages. They are *not* allowing "anyone" to create packages that make it into the "official" Debian distribution. It's quite a long and complex process of becoming an entitled official Debian package maintainer, they raise the bar quite high for someone to get there, and that's why they have good packages.
In my case I would package the Chicken Scheme compiler (along with many extensions) and the Ion window manager. Sure, they are relatively unpopular, but having them part of the distribution would surely make the life of those who depend on them easier.
And you're going to do the installation support for those, right ? You're going to do all the tests to see whether they work properly in the distribution, right ? Make those packages, put them online and available to everyone in your own package repository. Then join whatever the openSUSE packager aggregation process will be. That would be just fine, but including them into the SUSE Linux distribution ISOs involves a lot more than that.
With the current approach, I can't see any reasons why third-party developers would prefer to participate in (i.e. create packages for, promote, etc.) OpenSuSE rather than a distribution such as Debian, where they can become directly involved (as long as the specified procedures are followed).
Like becoming an official package maintainer, which is quite tedious. You're making wrong assumptions here. The reason for third-party packagers to participate in openSUSE should be that they like the distribution and because we (as a community, together with SUSE) develop a good model on how to integrate them into the process, but _not_ into the core distribution ISOs.
The reason I ask this is because, being a Novell employee, I am genuinously concerned with the future of OpenSUSE and I think its chances of success would be greater with a different approach. I know most of the Novell guys at SuSE are currently occupied with getting the 10.0 release out, but I find this issue far more important. It's a shame I couldn't jump into the discussion earlier.
The top priority now is to release 10.0, definately.
Let's wait a little for the SUSE staff to get finished with that and I'm sure quite a few of them
will jump actively into the discussion, and have their minds free enough to quickly get forward with
that very critical topic.
cheers
- --
-o) Pascal Bleser http://linux01.gwdg.de/~pbleser/
/\\