[wp-hackers] Community Views on Now and the Future

Ryan Boren ryan at boren.nu
Sun Mar 5 21:09:24 GMT 2006


Scott Merrill wrote:
> I think it's largely a cathedral, versus the more traditional open
> source bazaar.  A privileged few drive most of the development.

Whoever contributes code and ideas drives development.  Many people 
shape the development of WP.

>>What are its strongpoints? 
> 
> 
> The end result is largely consistent, because the privileged few driving
> development are mostly "on the same page".

I think it is good to share a philosophy.

>>What are its flaws? 
> 
> 
> There is no publicly defined vision.  There is no documented roadmap.
> The "same page" used by the developers is not shared with anyone else in
> a meaningful way.

Well, in large part the next release consists of whatever ideas people 
come up with along the way.  We try them on and see what sticks.

> Short of bug fixing, it's not clear how new people can meaningfully
> contribute.

Just do it.  Really.  Scratch an itch.  People who wait to be told what 
to do aren't contributing.

> Contributions from new (or unprivileged) participants are often rejected
> without any kind of positive reinforcement to encourage future
> participation.

Well, I'm not their mommy.  I tell people what is good and bad with 
their patch, and the people who really want to contribute will work with 
me and come back with another patch.  People who pitch a fit if I don't 
accept their patch wholesale are difficult to work with and not worth 
the time it takes to coddle them.  So, it goes both ways.

> This has long been a problem.  Way back in the 1.0 days, I submitted an
> ugly patch directly to Matt to provide per-user posting level support
> (like the view levels plugin now).  Rather than say "Thanks", or
> encourage me to rework the patch, Matt's word were, quote, "Does it have
> to be so complicated?".  That was my first attempt to really contribute
> to the core code, and it was rejected out of hand.  I should have
> stopped participating right there, in truth.

He says the same thing to me all the time.  ;-)  It's his primary job 
and a necessary one.

> My second real attempt to contribute found me working with Ryan.  He
> pointed out to me specifically where I needed to pay attention to stuff
> in the get_next_post() and get_previous_post() functions.  My patch
> evolved, with Ryan's encouragement, and was added to the core.

Thanks.  I do try.

> A lot of people are using Trac, and a lot of things are being marked
> "wontfix" with only terse explanations.  That's not helpful to the
> original reporter: they took the time to file a ticket on something that
> was important to them.  Curt rejections do not encourage participation.

We need to do better, but there needs to be some balance.  Some people 
expect huge discussion over why their pet idea was rejected.  That pet 
idea is usually one that has been suggested dozens of times before and 
everyone is tired of seeing it.

> I think there's not much leadership.  I think there's an autocracy,
> whereby one person (Matt) makes the bulk of the policy decisions.  Matt
> doesn't inspire participation.  He doesn't participate much on the
> mailing lists, when people are struggling with things that specifically
> need "official" response (see Robert Deaton's plea to squash the
> security rumors); but he has plenty of time to let us know he's going to
> hand out tee shirts at conferences.

A security announcement is forth coming.  A lot goes on behind the 
scenes that sometimes delays "official" response.

> I think most technical issues have reasonable discussion with Ryan,
> Andy, and a few blessed outsiders (Mark, Owen).  On the whole, the level
> of information sharing has increased in the recent months, with a marked
> decline in the number of things that are being added without the
> introduction of a corresponding trac ticket first.

They are not "blessed".  They are simply the people who do the work. 
How does one become a dev?  You fix bugs and write code.

Ryan




More information about the wp-hackers mailing list