maintainence change proposal

[tridge at linuxcare.com]

> It is worth noting that I have already received several (probably about
> 10) private requests for the "inherit mode" patch, which indicates:

sure, but I hope you can appreciate our caution with this one. Putting
it in is easy, maintaining it for years isn't. These sort of special
behaviour patches need to be looked at very carefully before being
incorporated. 

> Can you clarify:  isn't this basically what already happens?  (With the
> possible exception of "not [a team member] ... work up a patch".)

there are some big differences

- in the new system all submitted patches are public
- right now the traffic you see on the mailing lists is only the tip
  of the iceberg. We get thousands of messages to samba-bugs, but you
  don't see those.
- most patches received in a pretty bad state, requiring us to at
  the very least add docs, and often rewrite them completely. I hope 
  that with the new system we will get a higher quality of patches.

> Against what version?  For ourselves:

in many cases the version doesn't matter, particularly if it is a good
patch that impacts a small part of the code. The team member applying
the patch has to see if it still applies cleanly and if not then ask
the submitter to update it.

> 2. A patch might itself require changing from release to release. 
> Specifically, both my "inherit mode" and "veritas quotas" patches
> fail to carry over from 2.0.4b to 2.0.6 because the source files
> themselves changed.

yep, and until the patch is accepted it is the submitters
responsibility to keep it up to date or find someone who can do it for
them. I realise that this is a change that puts more work on patch
submitters and that is quite deliberate. The team just isn't keeping
up, so we have to distribute work. If a patch submitter can't keep
their patch up to date then in all likelyhood the patch won't get in. 

> How about a Web-based "test patches" area?  If front-ended with a brief
> and simple (and it must be brief and simple!) form, then at least a count
> could be kept of requests for a particular patch.  (Example:  I would
> upload my "inherit mode" patch there, and direct anyone who wanted it to
> that place.)

that might be good, but I'd like to try the initial steps I've
proposed first and then see how that works before adding more
complications. 

Cheers, Tridge