Shining more light on the problem
Posted Aug 23, 2006 0:51 UTC (Wed) by dowdle
Parent article: Who maintains RPM?
Thanks for the article. While the problem was public and there for all to see, by bringing it out as a feature article of LWN, chances that the situation will change for the better have increased.
It wouldn't be a bad idea to mention that rpm works well for most everyone most of the time... and that in the corner cases where something does go wrong, there is usually a lot of info to be found on how to fix the problem and/or avoid it in the future. Those admissions aren't an attempt to avoid resolving the situation.
I credit Red Hat for acknowledging the problem with a named individual... rather than the typical corporate anonymous insider comment.
Now we see the problem. We know the solution... we just have to make it happen.
I would like to comment in defense of Jeff Johnson. I think he was unnecessarily vilified... not really here but in various posts linked to in this article. Now, that doesn't mean I agreed with all of his comments, reasoning, or tactics... but what he didn't say was obvious... that with regards to that bug... he really didn't want to fix the issue and didn't want to admit it. If he wasn't being paid to fix it, and it worked well enough for him, and he was happy with it, he isn't a horrible person for not fixing the problem. It was just another opportunity for someone else to say that's not good enough for me, I'll fix it. It is sad that so much effort was put into nagging at him rather than someone just fixing it and submitting a patch. Sure, he egged people on by pretending there wasn't a problem... the but the axiom of "show me the code" always applies. Of course, that is easy for me to say since I'm not a programmer. :)
I would also like clarify that perhaps there should be some sort of group formed by the various distros that use rpm (not just Red Hat and/or Fedora) to maintain it. I would imagine that all of the various companies using rpm have someone assigned to work on rpm when needed... so having all of those people, across distros, work together seems like a no-brainer. Nothing too elaborate... just a commitment to work together, find problems... and hey... maybe even set a direction for the future... unless everyone just wants to adopt that new management system that was created by all of the former Red Hat folks. What was the name of that one again? I read up on it a while ago and it had so much lingo and so many layers and features, it was really above my head. How about an article about that one?? I ask the question rather than googling it, to give others an opportunity to add to this discussion.
Yes, Red Hat created RPM and it has worked well for a long time... but Red Hat set it free some time ago... and I think it should stay that way if at all possible... as long as it finds the way to continuing to work well for years to come... or a path to something better is forged for those using it.
This is just more evidence that Red Hat is suffering from PR shell shock... and they don't want to accidentally piss off the Slashdot hordes again by grabbing control back of something that they really wanted to free... like what happened with Fedora. Yeah, some people just can't grasp reality... and I think most of those folks go into politics.
Now off the beaten path... IMNSHO, Red Hat needs to merge Cluster Server, GFS, Directory Server, and Certificate Server... into RHEL AS/ES... and then reduce the price of RHEL... to somewhere around the cost of Mac OS X Server Unlimited... and/or SUSE Linux Enterprise Server. Really they do. Then I'd use RHEL everywhere... and CentOS wouldn't even cross my mind in some situations. Yes, all of their top 100 customers have deep, deep pockets and can afford their fairly close to reasonable existing price structure... but just think how many people would buy it if it was more affordable? Volume could make up for reduced margins. Really it could.
to post comments)