[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: should vs must



On Mon, Feb 26, 2001 at 11:25:41PM +0000, Julian Gilbey wrote:
>   MUST   -> RC if it doesn't conform
>   SHOULD -> Normal/RC bug if it doesn't conform, unless there's a
>             really good reason not to conform
>   will-be-MUST -> eventually it will be an RC bug not to conform, but
>             in the meantime, we are allowing a transitional period for
>             packages to conform without submitting bugs, or only
>             submitting wishlist/normal bugs
>   will-be-SHOULD -> similar
> I really like this idea; it will mean that when we discuss new
> proposals, we can talk about whether they ought to be MUSTs or
> SHOULDs, and separately about how long before they become this way.
> For example, source dependencies should probably be MUSTs, but right
> now, we shouldn't file RC bugs because we're nowhere near close.  The
> /usr/doc transition is probably much closer to being a MUST now.

I'm stronly against putting things about the future in policy. That
might not be rational, but we'll see. That said...

One possibility would be to add an extra stage for policy amendments,
something like [BLOCKED], which a proposal can be placed into if it's
objected to, but the objectors and proposers agree that the objection
would be solved if and when something particular and measurable happens.

So the Build-depends proposal (if worded correctly) could be
considered [BLOCKED] pending 95% of packages having fully specified
build dependencies; and the non-US/crypto proposal could be considered
[BLOCKED] pending legal advice.

It might then be appropriate to give rough summaries of BLOCKED proposals
in a policy appendix, or something, maybe.

As far as transitioning though, the minimal common subset is what should
be required (MUST, or even SHOULD), and both the former and new states
should be allowed (SHOULD, or MAY). That was a bit screwed up with the
/usr/doc transition.

Cheers,
aj

-- 
Anthony Towns <aj@humbug.org.au> <http://azure.humbug.org.au/~aj/>
I don't speak for anyone save myself. GPG signed mail preferred.

``_Any_ increase in interface difficulty, in exchange for a benefit you
  do not understand, cannot perceive, or don't care about, is too much.''
                      -- John S. Novak, III (The Humblest Man on the Net)



Reply to: