[Gnu-arch-users] Re: named patches, patch order, patch queue manager

[Miles Bader]

"Stephen J. Turnbull" <address@hidden> writes:
> While AFAIK no arch users branch for every clean changeset yet, I
> suspect that this would make a lot of sense for the lk environment,
> where a single patch may live for many months before being blessed.

Using branches like this seems right to me, and it's basically as cheap
as any other changeset (basically one extra directory and one extra file
in the archive; I'll live!).

The one thing that bothers me is namespace pollution -- sure I can use
naming conventions to avoid gross conflicts (e.g., call all my `named
patches' something like CAT--patch-NAME--0), but having a flat namespace
for all branch names might be a bit annoying after a while, e.g., when
looking at abrowse output, you'd see every single named patch,
regardless of whether obsolete or not, and for some of the major linux
hackers, there'd be _many_ such patches.

Perhaps it could be as simple as having an `hidden' (or `closed') flag
in archive branches/versions/something, so that `tla abrowse' and `tla
branches' wouldn't show them by default (say unless given given a
--hidden option); of course branch creation &c would still see them.
Then after a patch was merged into Linus's tree, you could just mark it
as hidden.

This might also be useful in general, since probably people have certain
branches/categories that are basically obsolete, but which following
arch practic, probably shouldn't be deleted from the archive.

-Miles
-- 
Is it true that nothing can be known?  If so how do we know this?  -Woody Allen