John DiMarco wrote:
> Unfortunately, it's a fundamental problem with the Linux development
> style. People tend to develop what they're interested in, not what's
> needed by the people who use it for real work. As more and more
> companies jump onto the Linux bandwagon, hopefully, some of the less
> sexy stuff that "needs to work" will get done. This is already
> happening, but needs to happen more.
I agree, the sexy stuff tends to get done first by the open source
community. However RedHat itself doesn't have this excuse anymore since
they're selling their OS as a business ready OS. It is improving as you
said though.
> As for RedHat in particular, I was troubled by the fact that "we need
> this to work and here's the RFC that says it ought to work" wasn't
> very important to the developer(s) I communicated with.
Yes, the BugTrack for this gave me pause. The Bugzilla log for this is
particularly ugly in the way it plays out. I thought (though I can't say
why) that RedHat would be more immune to these sorts of arbitrary
decisions. It's becoming more clear that this is not the case as they
change (or don't) high visibility items in ways that don't seem to
necessarily reflect general industry consensus or logic. Another example
was rewriting "dhcpcd" with "pump" rather than fixing the open source
ISC version. That was a real "huh?" to me.
- Matt
_______________________________________________
Redhat-devel-list mailing list
[EMAIL PROTECTED]
https://listman.redhat.com/mailman/listinfo/redhat-devel-list
