On Sat, 2003-12-20 at 20:26, George Shapovalov wrote:
> Well, the thing is that I *do not want* to maintain these packages. Over these 
> years (sounds funny when applied to Gentoo, doesn't it? :) . But it really 
> was like two years for me already) I processed all kinds of packages 
> semirandomly assigned to me. So now, with the push for organization, I would 
> like to loose some of them :).
> 
> Adding a temporary metadata does not work, because in reality it is not going 
> to be temporary. In my experiense things tend to stick, even if there is a 
> reasonably high traffic in them. With low maintaince packages their metadata 
> are going to be like set in stone.

Well, you do have to actively look for a maintainer. Anyway, you can
usually go with the category herd, only you do have to inform them and
make it 'official' theirs.

> Besides, on that "noherd" thing - I despise it. For one it is plain illegal. 
> Plus, considering what I just mentioned about the temporareliness of metadata 
> of such packages, it is likely to remain "noherd" for foreseable future.

Yeah, the example was about the comment really, you are reading things
into it. I don't like no-herd much either and wouldn't use it myself,
I'd try to find a suitable herd or create one (it's also about indexing
what herds are needed).

> > In my opinion it is most important to get full maintainer/herd coverage
> > of packages asap.
> Just adding whatever in order to get that damn metadata in is a bad thing IMO. 
> We will need anoter major push down the road, but this time to get trough 
> "noherd" packages (I hope this is the only kind of junk we collect in 
> metadata atm). Why don't we do it all in one go instead?

Hm, what do you mean ? 'no-herd' is a valid entry, it is no problem to
have no-herd packages as long as it has active maintainers. So there's
really nothing to go trough.

> So, the only change I can envision doing is, instead of "unofficially" taking 
> care of these packages as time allows, to outright drop them and throw the 
> related bugs back to bug-wranglers.

You shift the problem to developers outside yourself, i don't know what
packages you talk about exactly. If you were at one time maintaining a
package, you are probably still one of the best people on the team to
take care for it as long as it has no herd assigned.

I know there are a lot of devs who have in the past happily added new
packages to portage, but when it came to maintaining them they really
let it go a bit. I don't know if that is the case here and i do not say
you did this, but the whole herds idea is about responsibility and right
now the responsibility for you is to get these packages certain
maintainership.

The move to a full covered maintainership of packages in the tree trough
herds is a big job (which has already taken too long imo) and i think
every dev has a responsibility to do his part of the job to make it
happen. Pushing it all on the bug-wranglers as you suggest is no option
in my opinion.

These are my views, i don't know how people involved in bug-wrangling or
the herd initiators think about this.

- foser


--
[EMAIL PROTECTED] mailing list

Reply via email to