Re: Patch policy

[Jonathan Larmour <jifl@eCosCentric.com>]

Andrew Lunn wrote:
>>This is policy that we would have to develop (decide on).
>>One way would be to only accept patches via BugZilla.  Then the
>>onus would be on the submitter.  To make this policy work, maybe
>>we'd want to have the default owner of the "bug" be the patches
>>list, or at least send a copy there.
> 
> 
> I think part of the problem is that there is no clear owner of a
> patch. So i would not set the default owner to "bug", but rather the
> owner of the component, so we have a real name we can point a finger
> at. The assignment can be changed manually if its not appropriate.

One important thing we're missing though is a sort-of "checklist" for 
patches. We (I) have been lax in the past sometimes, and then the problem 
multiplies when someone uses a laxly written port as a basis for _their_ 
port. Part of that would be coding standards - not so much indent level 
and such like as just making sure comments in files refer to the current 
filename not an old one, author is correct, doesn't mention the 
architecture/platform it was derived from etc. But we have to solve our 
own inconsistencies there too!

A (script-driven) overhaul of our standard preamble would help a lot too - 
authors vs. contributors is somewhat ambiguous. I would prefer 
"maintainer" and contributors, where the former is the person responsible 
for the file (not necessarily one of us, i.e. not necessarily a maintainer 
maintainer :-)) and the latter is _anyone_ who's touched the file (if they 
want). But this is just one of the low priority things to do that there's 
never enough impetus to do.

Jifl
-- 
eCosCentric    http://www.eCosCentric.com/    The eCos and RedBoot experts
--[ "You can complain because roses have thorns, or you ]--
--[  can rejoice because thorns have roses." -Lincoln   ]-- Opinions==mine