[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Experimental or unstable.



Hi DDs,

> [[ Marc Haber ]]
> Experience with adduser shows that no-one besides the maintainers
> themselves and their closest environment uses experimental packages.

> [[ Simon Huggins ]]
> See this worries me a bit.
> 
> I'd love for Debian users to test some more cutting edge versions of
> packages (partly so upstream gets more testers, partly so I can see the
> bits I hate about the new versions and try to get them fixed) but I
> don't want them in testing now until they are properly released as a
> stable series.  In fact, I don't really want them in unstable if it
> means that in order to fix bugs which appear in testing I have to revert
> to the stable set of packages in order to get them in.
> 
> I guess the compromise is just to realise that they just won't be
> thoroughly tested when they hit unstable if they've only been in
> experimental before.
> 
> But then it does make you wonder if it's worth putting anything in
> experimental in the first place.
> 
> Is there any better way I can get snapshots/betas tested by the majority
> of users?  Do people think that this is the sort of thing that should
> just be uploaded to unstable and allowed to flow into testing?
> 
> Simon.

I wouldn't say that package maintainers are the only ones who use experimental
packages. My main system is running experimental/unstable, and I usually report
problems with the new packages immediately. I might not see all bugs, because
I don't do strictly daily upgrades (and I don't have udev installed, so I can't
comment on it), but I try to test every experimental package that hits the
archive.

But you surely won't reach the "majority of users" via experimental. Although
I haven't experienced any major problems, I think I wouldn't dare running
experimental if I didn't have some knowledge of Debian's mechanisms and tools,
assured that I could fix upcoming problems myself in an emergency. Retaining a
running system comes before reporting bugs and submitting patches...

Yet I simply can't believe that the number of people who are gamy enough to run
experimental is so small...


Regards,

Jan

PS: My opinion definitively is to keep experimental.

Attachment: signature.asc
Description: Digital signature


Reply to: