Re: Must and should: new proposal (was: Re: Must and should again)
>>"Julian" == Julian Gilbey <J.D.Gilbey@qmw.ac.uk> writes:
Julian> - MUST and SHOULD change to the universally-recognised IETF meanings
Julian> - the distinction between RC and non-RC bugs is retained clearly
Julian> - it's clear what one ought to do to create a "good" Debian package
Julian> - there's no time component involved
All the above sound good.
Julian> - there's no longer a suggestion of using policy as anything other
Julian> than a set of guidelines
Is that really the case? I certainly do not find that I treat
Policy as a guideline, to be followed or violated at my whimsy; I do
think of it as a set of rules, and as all rules, there may be
exceptions. I would hesitate to declare such an exception myself,
though, and I'd ask for a ruling on an exception to policy rather than
deciding that on my own.
And when I slip up, and no longer follow policy, there are all
these people helpfully pointing that out to me in bug reports.
Perhaps I am getting bogged down in semantics here. But part
of the value of policy is that I can rely on packages in Debian to
follow the rules (and not just when the maintainer chose to follow
the guideline), so that my packages can add value that would not be
possible in an anarchic environment.
To be wise, the only thing you really need to know is when to say "I
Manoj Srivastava <email@example.com> <http://www.debian.org/%7Esrivasta/>
1024R/C7261095 print CB D9 F4 12 68 07 E4 05 CC 2D 27 12 1D F5 E8 6E
1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C