Quotes of the week

[Posted February 16, 2011 by corbet]

So never _ever_ mark anything "deprecated". If you want to get rid of something, get rid of it and fix the callers. Don't say "somebody else should get rid of it, because it's deprecated".

And yes, next time this discussion comes up, I _will_ remove that piece-of-sh*t. It's a disease. It's just a stupid way to say "somebody else should deal with this problem". It's a way to make excuses. It's crap. It was a mistake to ever take any of that to begin with.

-- Linus Torvalds

Hey, if that's what it takes to get __deprecated removed i'll bring it up tomorrow!!

-- Ingo Molnar

Quotes of the week

Posted Feb 17, 2011 12:18 UTC (Thu) by dgm (subscriber, #49227) [Link] (3 responses)

Linus is in fact meta-marking __deprecated as deprecated, instead of following it's own advice and removing the sucker himself.

Oh, the irony!

Quotes of the week

Posted Feb 17, 2011 22:00 UTC (Thu) by proski (subscriber, #104) [Link] (2 responses)

In other words, Linus is contradicting himself on the meta level. I wonder what Douglas Hofstadter would say.

Quotes of the week

Posted Feb 20, 2011 12:51 UTC (Sun) by smitty_one_each (subscriber, #28989) [Link] (1 responses)

I'll speculate: "Oh my Gödel, I wish somebody would Escher these guys off my Bach and out the door!"

Quotes of the week

Posted Feb 21, 2011 14:51 UTC (Mon) by dmk (guest, #50141) [Link]

Oh my god... I'm rolling on the floor laughing...

Quotes of the week

Posted Feb 21, 2011 3:59 UTC (Mon) by Baylink (guest, #755) [Link] (2 responses)

I think that what Linus is *actually* doing is misunderstanding how deprecation is usually (recommended to be) implemented (and, of course, with my track record, he'll come along and correct me :-) --

When you deprecate something, you're supposed to modify it, in the first deprecated release, so that it works, nominally at least, but *drops big turds in a log or on a command line -- or in a dialog window -- that say that it's deprecated, and that either a user should upgrade their app to the newer version, or that their developer should do so.

You don't just tell people, you make it *annoying enough that they'll fix it.* And then, one or two releases later, you take it out. (My favorite approach is

.1 - be annoying
.2 - crash the app
.3 - go away entirely to save space

Quotes of the week

Posted Feb 21, 2011 4:57 UTC (Mon) by neilbrown (subscriber, #359) [Link] (1 responses)

Nah ... this isn't about apps. This is about in-kernel interfaces. i.e. interface between different parts of the kernel.

In this case there is no point marking anything deprecated. You just remove it and fix up everything that you broke. There is no need for any transition period.

Now if we wanted to make life easier for people who maintain external modules, then it might be good to keep the "deprecated" marking. But we don't. So we don't.

Note that all this has nothing to do with deprecating user->kernel interface. It is certainly still possible to mark them "deprecated. But you don't do that with a __deprecated compiler tag.

Quotes of the week

Posted Feb 21, 2011 8:35 UTC (Mon) by nix (subscriber, #2304) [Link]

There is no need for any transition period.

Is this pre-git tradition, perhaps, dating from the days when carrying forward great big fixup patches was really, really, unpleasant, so you had a reason to push as much as possible upstream even before the API conversion work was done?