Re: Optional and conflicting packages.
On Thu, 28 Jan 1999, Richard Braakman wrote:
> Santiago Vila wrote:
> > Which do you think it is the purpose of downgrading a package to extra
> > when it does conflict with a package of "higher priority" if it is not to
> > make required+important+standard+optional a self-consistent set of
> > packages, then?
>
> I think it's because removing a standard package to install another
> package is a bit unusual, and shouldn't be done without some prior
> thought.
Good point.
I was talking in terms of "global consistency".
If we had no rules about conflicts (not even the one in the definition of
"extra"), I think we would have at least the common-sense rule that the
"default" packages (required+important+standard) may not conflict between
them.
With your interpretation, you can install all the default
(required+important+standard) packages, plus any other *single* optional
package at your choice, but if you choose two optional packages at random
then you can found a conflict between them.
This is not a great gain with respect to the common-sense rule, it
does only allow you to install just a *single* optional package in
addition to the default ones without conflicts.
With my interpretation (that the higher priorities refers to the
priorities that have been presented in the previous paragraphs), you could
install all the default packages, plus any given set of optional packages
chosen at random (which is what a clueless user would do).
Is it not this interpretation more meaningful than the other one?
Do you think Ian Jackson would write a policy rule which does only avoids
conflicts if you only install a single optional package?
BTW: Should not we ask him about the real meaning, instead of saying how
we do interpret the definition of "extra"? I'm not a native english
speaker, but I think that the phrase is, at least, ambiguous, and my
interpretation may be at least as right as the yours one.
Thanks.
--
"86622529da6713af92eff573d53b5170" (a truly random sig)
Reply to: