[wp-hackers] Easy-Hacks for WP

[Stas Sușcov]

În data de Tue, 31 Jan 2012 21:23:22 +0200, Jesper Jarlskov  
<jesper at jarlskov.dk> a scris:

> I have on several occasions talked to the people behind the Linux
> distribution Exherbo[0]. They have had great success getting new
> contributors onto the project by keeping a bunch of low hanging fruits
> around. This is stuff that is not critical, ie not security issues and so
> on, but small easy to grasp usually nice-to-have features, that could
> easily be fixed by the core developers. They make sure to always keep  
> some
> of these issues around to give aspiring contributors an easy way to get
> started on contributing.
>
> I really like this idea and I know, as mentioned, that other projects has
> had great success with it. I think one of the biggest problems to getting
> new contributors on a project is that first commit.
>
> [0] http://exherbo.org/
>
> Jesper Jarlskov
>
>
> On Tue, Jan 31, 2012 at 8:11 PM, Chris Williams <chris at clwill.com> wrote:
>
>> IMHO, the last thing a relatively mature codebase like WP needs is  
>> people
>> randomly tossing in a bunch of "easy hacks".  Every piece of code needs  
>> to
>> understand the context it is in, needs to respect the coding, data, and
>> structural paradigms, and needs to play well with the past, present, and
>> future of the codebase.
>>
>> In my experience, some of the most insidious bugs in the projects I have
>> worked on were introduced by well-meaning people who simply didn't
>> understand the context well enough.  I would rather have some of these
>> small, even trivial, bugs lie fallow rather than have them "fixed" in a
>> way that introduces more bugs, more complexity, or even hamstrings the
>> codebase against some larger, more elegant, long-term solution that may  
>> be
>> in the works.
>>
>> But that's just my $0.02,
>> Chris
>>
>> On 1/31/12 8:20 AM, "john malc" <dimitrijenko at gmail.com> wrote:
>>
>> Can i get more ideas by people on it. Agree/ Not agree etc
>>
>

Hi,

All this easy-hacks thingy sounds much like a primitive code review  
workflow.
Nowadays this problems are solved by tools like gerrit/patchwork or  
continuous integration systems.

I understand that everyone wants his patch accepted or reviewed, but the  
problem you described can't be solved like that, simply because more  
patches doesn't mean more code gets accepted. And it's a bit unfair to ask  
a developer to do a job that can be automated (like codestyle check,  
validity of the diff file, quality of the code).

The example with a linux community is also a bit unfair, since a linux  
project is usually split into submodules and sub-projects that have own  
maintainers.

To sum up my thoughts, it's not the workflow that has to be changed, it's  
that WordPress needs more tasks to be automated before a patch review  
request is fired to devs.
In the end, as scribu said, it should mainly simplify the commit process,  
not replace it.

As a parallel note, there's a thread "Unit Testing" that was started,  
which questions solved, could definitely simplify the commit process.