> We should also acknowledge the pain in the other direction - there is a
>> category of "small fixes" where our response (although understandable give
>> our limited resources) is viewed as out of line with respect to other open
>> source projects.
>>
>
> Can you provide some examples? I don't work very well with general
> statements.
>

Should of been more clear that I agree with everything you say, and was
trying to add the idea of how we look to potential contributors into the
mix. Each case I can think of (jdbc timezone fix that I rejected, wps pull
request that went from simple to insurmountable) has a danger of
distracting from the point - that developers expect to be able to
contribute a fix and walk away. Both of those examples had very clear
reasons why they were rejected - but the impression remains that our
codebase is hard to work with.

A review guide as you describe it, can help both those new to the geoserver
community conduct a review and feel confident in the result, and those
contributing to the project to know what to expect and that we are treating
them fairly/consistently.

It seems by your comment one way to handle this would be to increase the
> resources, so that the reviewer of the day is not so hard pressed to do
> things quickly and get back to work/family.
>

Yeah, the real strain here is lack of resources on the project.


> If it's relaxing the amount of stuff to be changed in order
> to require a contribution agreement, I don't see a problem, the current
> practice has been set basically following Locationtech advice, I don't
> think anyone around here wanted to make it so strict.
>

That was more apache advice when we were setting up OSGeo, but I do
understand your point.

If it's not requiring tests, then you have my unconditional -1, too many
> people making too many chances unaware of each other, if there are no tests
> you might as well not contribute, the change is going to be broken soon
> anyways. Again, you're talking too much in generic terms for me, could you
> propose something more specific?
>

A test case and documentation are often considered unreasonable. We do a
good job of communicating why they are required. And I agree with our
limited resources submitting a fix without a test case is wasted effort.

So for something more specific - at the end of a "review checklist" we
could include a paragraph inviting the reader to join geoserver-devel and
ask questions. Or at least acknowledge that this is hard work and the
checklist helps us share the work more evenly.
------------------------------------------------------------------------------
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel

Reply via email to