>  The original thought behind the present model for submission of patches
>  was based on the existing procedure for kernel patch submission - it
>  looked like a model which worked, and by implementing it OpenWrt would
>  "teach" people a skill which could be needed later on.
I don't disagree, but do not think OpenWRT has quite replicated that
structure.  We have good submissions and proper formatting, but
there's still no one (or at least seemingly so) responsible for
actually taking the submissions and implementing them.  I've seen "get
the package maintainer's attention and hope" on #openwrt-dev before,
but even that seems to seldom work.

>  * Keep things as they are.
>  * 'Ticket' based submission system.
>  * 'Ticket' based submission system with added notification to -devel.
>  * Launchpad (as new platform).
>  * (Did I forget something?).
IMO, yes - keep things the same but have a developer christened as
patch-wrangler.

>  I think it might be needed to make a distinction between submission of
>  patches targeted at packages, and patches targeted at the "base" system
>  itself?
That would be most certainly helpful, but it seems most of the devs
working on -base are nearly too busy to do their own changes, much
less review and apply 3rd-party patches.

>  It would be really good to extend the list of package maintainers, then
>  patches for packages could be send directly to the maintainer - or?
As long as those package maintainers are reasonably responsive and
have commit access.  It would also need to be very clear who each
package's maintainer is; right now, one has to read through the SVN
log and hope there's enough evidence to select someone that cares
enough about the package not to ignore you.
_______________________________________________
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
http://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel

Reply via email to