KS2008: Bootstrap code

Posted Sep 18, 2008 16:33 UTC (Thu) by vmole (guest, #111) [Link]

That's what Dave Jones is doing. If you read the rest of the article, you see there are some other approaches being considered :-).

I think the whole thing is going to be a tough sell. The first time a customer's system won't boot because of the new init code, there will be strong temptation to revert.

Posted Sep 19, 2008 2:13 UTC (Fri) by roelofs (guest, #2599) [Link]

That's sociology, and I claim it's the fundamental problem. The rest of the article mostly talks about technology and implies that it will somehow address the sociological issues, but it seems clear to me that that's not the case. (Or at least it's clear to me that the argument has not been successfully made.)

Better to spend some time looking at what the different big distributions do (because they probably have hit the most weird situations), pick one, and ask the other distributions "Why won't this work for you."

As you can see from the rest of the article. this is precisely what is being done. Using Fedora's as a starting point.

To restate it slightly: a Red Hat/Fedora guy is starting with the Red Hat/Fedora solution and hoping/expecting that it will work for everybody after a bit (or even a lot) of tinkering.

That, too, is sociology, but of a different sort. Assuming we can agree that the fundamental problem is sociological--and I'm under no illusions that we'll even agree on that, but bear with me ;-) --then it would seem that the approach Dave has taken could do with a fair bit of improvement. In particular, if you really want to convince others that "we can all just get along," maybe the starting point should be someone else's solution (perhaps Debian's or SuSE's or whatever). That would make a strong statement, and it would become an even stronger statement if/when Fedora and Red Hat officially adopted the new initramfs. (Of course, if they never did, that would also make a strong statement, albeit of a different flavor. :-) )

Greg