Over the last few days there has been a lot of discussion on the WordPress Hackers email list about having wp-plugins.org become an “official” repository for WordPress plugins. The thread really picks up steam around this post from Ryan Boren.
After skimming through most of the thread the points seem to be focusing on get your plugin into wp-plugins.org or we may leave you out in the rain. Features like plugin updating and maintenance are being thrown around, with the idea being that if your plugin is part of wp-plugins.org you’ll get these features and services and if you aren’t then you are on your own. This of course has made some plugin authors not too happy. Like so many other things in life, both sides have valid points and are discussion is quickly becoming a mess as each side tries to talk over the other in an effort to get them to see the light.
In some cases people are torn between two positions, like Jason:
As a developer, I hate this idea.
As an end-user, I love it.What a conundrum :(.
From what I can divine from the conversation WordPress is hoping to make things better for the end user (with updating and maintenance of plugins as mentioned above), but is doing so in such a way that makes it harder (for at least some) plugin authors. No surprisingly this has lead to comparisons with other projects the Linux kernel and Debian. I don’t want to get into those comparisons much other than to say that most (if not all) of these comparisons are completely off.
I’m not sure that it is a good plan to shift more of the burden upon plugin developers. That said, having plugins break all without getting fixed isn’t much help either, especially when new versions of WordPress come out. I will say that at least some of the problem is with communication. When WordPress 2.0 came out there was little warning and when the warning did come with was with in a few days of Christmas. With less than five days (during the holidays) doesn’t give plugin authors much time to test and verify that their code works with the new release.
One thing that could be done is to give authors at least two weeks and have an email list for just plugin authors that could be used to alert them of upcoming releases. Although there would be a temptation to send out other info to such a list, I thing that would makes things worse and not better. Keeping it limited, with a specific focus, will allow people to spend more time working on code and less time trying to keep up with email.
In the end Matt will probably be the driving force behind what ever decision is made. Hopefully it will be one that will balance the end user experience with any additional burden on plugin developers.
One reply on “WordPress Plugins”
There were 12 days between when we said the release was coming out, and when it did. Also a months-long cycle where every new feature (object cache, usermeta, etc) was discussed on wp-hackers. There were thousands of people running nightlies, apparently none of them plugin authors. 😉
However despite all that, there were still lots of plugin compatibility probs with 2.0. I think some of this might just be user error, they don’t know when something is new. An update notification in the core would be easy to add and not cause any security problems.