Bug#850156: Please firmly deprecate vendor-specific series files [and 1 more messages]
Russ Allbery writes ("Bug#850156: Please firmly deprecate vendor-specific series files [and 1 more messages]"):
> Very belatedly, thank you, this makes a lot of sense. I hadn't thought
> about the angle of the most likely vendor to be used in a vendor-specific
> series file deciding whether they want this feature to be used at all.
> When you point that out, it makes perfect sense to me that Ubuntu should
> be able to decide via their own project governance methods if this is a
> feature that Ubuntu doesn't want.
Are you suggesting, therefore, that if Ubuntu say they don't want
this, only series.ubuntu it will be prohibited in Debian ? What about
series.fooix ? Are we to maintain a list of who wants these and who
doesn't ? Or is Ubuntu going to decide for all derivatives ?
My view is that Debian can decide what we will allow in our archive.
That includes deciding not to allow our derivatives to store
series.fooix *in our archive* even if Fooix would like to do so.
Deciding to not allow that does not impede Fooix's freedom to do
whatever they like in their own archive. If Fooix and its derivative
Barix together decide to use series.fooix and series.barix then they
can do that.
I still wish they wouldn't. Users who want to use dgit with Barix
will surely wish they wouldn't. And it is Debian's role as the
standards body for our derivative ecosystem to guide people in the
right direction. If we forbid this in Debian policy then few
derivatives will choose to diverge from it. That is appropriate.
On a purely practical level, I would like to avoid this being derailed
by jursidctional arguments, delayed, and diverted into Ubuntu's
governance processes (which are perhaps less suited to tedhnical
standards development regarding packaging technology, and certainly
less visible to the whole Debian ecosystem).
Ian.
--
Ian Jackson <ijackson@chiark.greenend.org.uk> These opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
Reply to: