Bug#727708: init system coupling draft CFV

2014-02-21 Thread Andreas Barth
* Russ Allbery (r...@debian.org) [140221 06:15]: This includes the change I proposed to Andreas, although unfortunately Andreas hasn't had a chance to respond on whether that addressed his objection. It also makes it clearer that the point about not offering advice past jessie only applies to

Bug#727708: init system coupling draft CFV

2014-02-21 Thread Ian Jackson
Andreas Barth writes (Bug#727708: init system coupling draft CFV): * Russ Allbery (r...@debian.org) [140221 06:15]: This includes the change I proposed to Andreas, although unfortunately Andreas hasn't had a chance to respond on whether that addressed his objection. It also makes

Bug#727708: init system coupling draft CFV

2014-02-21 Thread Ian Jackson
Russ Allbery writes (Bug#727708: init system coupling draft CFV): Here is the new draft of my ballot option. I think this addresses all of the issues that were raised. I didn't rewrite the whole thing to avoid all the shoulds; I thought about it, but it felt like it made it a bit too hard

Bug#727708: init system coupling draft CFV

2014-02-21 Thread Russ Allbery
Ian Jackson ijack...@chiark.greenend.org.uk writes: I don't see it. Perhaps you failed to push. I have transposed it into git myself and pushed. Indeed, I forgot to push. I'm sorry about that. :/ I'm clearly not at my best at the moment. I can confirm that Git found your change to be

Bug#727708: init system coupling draft CFV

2014-02-20 Thread Ian Jackson
Ian Jackson writes (Bug#727708: init system coupling draft CFV): Here are the options which have been proposed so far, with the proposed amendments incorporated as applicable. You can find the history (with messageids) in the tc git repo. There are curently four options: Ian mark 2

Bug#727708: init system coupling draft CFV

2014-02-20 Thread Ian Jackson
Ian Jackson writes (Bug#727708: init system coupling draft CFV): There are curently four options: Ian mark 2 (inclues amendments from Colin and Ian) Keith Russ (includes one thing that loos like an amendment) These descriptions of the options are obviously suboptimal. We should

Bug#727708: init system coupling draft CFV

2014-02-20 Thread Russ Allbery
Ian Jackson ijack...@chiark.greenend.org.uk writes: I suggest (for the options on the ballot right now): L Software may not depend on a specific init system (Ian mk2) A Provide advice; software may so require (Russ) A Advise sysvinit compatibility through jessie, multiple init

Bug#727708: init system coupling draft CFV

2014-02-20 Thread Ian Jackson
Ian Jackson writes (Re: Bug#727708: init system coupling draft CFV): Russ Allbery writes (Bug#727708: init system coupling draft CFV): is the closest I can get (at least right at the moment) to capturing the nuance of my proposal in one line. How about A Advice: sysvinit

Bug#727708: init system coupling draft CFV

2014-02-20 Thread Russ Allbery
Ian Jackson ijack...@chiark.greenend.org.uk writes: How about A Advice: sysvinit compatibility in jessie and multiple init support which makes it clear that it's all advice. Replacing through with in gives us more space and sounds less corporate :-). Sounds great here. Thanks! --

Bug#727708: init system coupling draft CFV

2014-02-20 Thread Russ Allbery
Here is the new draft of my ballot option. I think this addresses all of the issues that were raised. I didn't rewrite the whole thing to avoid all the shoulds; I thought about it, but it felt like it made it a bit too hard to read. I did try a few different wordings of the bit about upgrades,

Bug#727708: init system coupling draft CFV

2014-02-19 Thread Ian Jackson
Here are the options which have been proposed so far, with the proposed amendments incorporated as applicable. You can find the history (with messageids) in the tc git repo. There are curently four options: Ian mark 2 (inclues amendments from Colin and Ian) Keith Russ (includes one thing