There were a lot of things that I could have written about this week. Like, for example, Red Hat getting p0wned. Or how freetards in Quebec want their goverment to be even more inefficient. (Does this mean now they have to come to the US to buy their copy of Windows, along with their MRI's?)
No, the fruit ripest for the picking this week is Red Hat's perl performance bug.* Long story short, it turns out that the version of Perl that RH shipped has a huge performance bug that a whole bunch of people have just been working around. There's also apparently a bug report open about this that's been around for the better part of a year. Nicolas Clark, a core perl developer, also chimes in on how Red Hat fucked up.
The bug itself is not that interesting. It's just a performance bug. Whatever, it happens. But it's everything around it that has the ever so recognizable luser stench.
First there's the bug report it self. Go ahead, read through the comments. It's full of a bunch of whiners being actively hostile in an attempt to get RH to fix the bug. Way to set an example guys. To all the companies who have yet to go deploy a public-facing bugzilla, you're effectively saying, "do this, and we will shit all over you whenever you have a bad bug." It's a really great way to show that open bug DB's can be good for companies. **
Second, it's yet another example of where access to the source code causes stupid things to happen. RH released a version of the source code that upstream never released. I guess they have the freedom to do so, in this case, this freedom wasted a bunch of time for a bunch of people.
Third, is something that the upstream developer notes this himself:
RedHat seem to have an aggressive policy of incorporating pre-release changes in their released production code. This would not be so bad if they actually communicated back with upstream (i.e. me and the other people on the perl5-porters mailing list), or demonstrated that they had sufficient in-house knowledge that they didn't need to. But evidence suggests that neither is true, certainly for
Just because you have the source code, doesn't mean you have the expertise to maintain it, modify it, and release it. This is one of the biggest fallacies of the open source model. Lusers always like to say that as long as they have the code, they can make it work and fix things themselves. Never do they say that as long as they have the code, they have way more opportunities to fuck things up, because they often have no idea how or why the code was written as it was. This is just another example of this. *cough* Debian *cough* SSL *cough*
* Bad week for them, I suppose.
** Not that people don't do this already on various forums and bulletin boards. But when the nastiness appears on the same company-hosted page that everyone else is going to be looking at to find a solution, it looks especially bad.