XPI infrastructure needs some love

[Lapo Luchini]

Doug Barton wrote:
> If we have maintainers who are willing to keep these things
> up to date and people agree that this is an acceptable [way]

Well, I'm not so sure about the ones I don't use, but the ones I do use
are usualyl well maintained (either that or they get a PR form me), but
there's also the fact that updating them is usually a no-brainer (in
most cases it could well be automated, as Andrew Pantyukhin has
suggested); it seems to me that a maintainer is not the most scarce
resource in updating an XPI port, committer time seems to be a scarcer
resource to me (but that's of course the point of view of a maintainer).

> I'm particularly concerned about the situation where we have them in the
> tree but they are not up to date. That's not only a waste of resources
> it makes us look bad.

Especially so because system-wide packages are then more difficult to
update for the single user; I don't remember at the moment if a stale
system package can be Disabled *and* installed locally by the user or
not (as I usually update the port first without trying to update as
user), but that (if it is the case that it's impossible) would be a
strong additional reason to be sure to have latest version in the ports.

But I'm not really sure about this, I will double check next time an
extension I do use is outdated (before updating the port).

-- 
Lapo Luchini - http://lapo.it/

“The use of COBOL cripples the mind; its teaching should, therefore, be
regarded as a criminal offense.” (Edsger W. Dijkstra)