WordPress 3.0 Beta Performance

First impression is that even with 283 opened tickets for 3.0 milestone on WordPress TRAC, current version is fairly stable. And also, very important to notice is that WP 3.0 is faster and uses less resources than it’s predecessor. So far so good.

If you follow the discussions on the current WP 3.0 progress, you surely know that major problems remaining right now are Menus and custom post types. And while the custom post types are working good as far as I can tell (with few problems still), the same can’t be said for Menus.

Custom post types and taxonomies

Even though custom post types were implemented in WP 2.9, it was very basic implementation, and you couldn’t do much with them. But now, custom post types are on the same level with the posts and pages. And while there are some bugs still remaining, I am confident that it will be sorted out before the official release. This website uses custom post types for the central documentation, and right now the only thing I couldn’t implement is having comments with posts in documentation.

Custom taxonomies are much improved with additional hierarchy support. You can try my GD Custom Posts And Taxonomies Tools plugin to see how all that works right now.

Menus

As of yesterday, I managed to create first menu and to actually see the editor working properly and display nesting of the menu items. But, menus editor in my opinion is not good (not to use stronger word). I have several problems with the way the menus are implemented. They are forced into a taxonomy, even thou, menu has nothing to do with taxonomy. Menus need much complex data model to be able to allow much wider range of items and taxonomies are not good for that. Editor will be a pain to use if you need to create big menus that have hierarchies of pages or categories with no settings on how to handle page item that already has sub pages or categories with sub categories. In my xScape theme I have built menu designer that has much more options and more control over the menu items and how they are displayed including different way to render items. Also, all menus in xScape are saved into the WP options table as serialized objects, and that I find more logical than approach in with WP native menus. Also, it’s easy to expand such structure. But, deed is done, and I don’t believe that anything with menus implementation will be changed in the next 2-3 weeks. We will see who the menus will perform and how much problems will cause in the long run. Good idea, average implementation.

WPMU to Multi Sites

As for the merge with WPMU, I still have to test how the conversion of WPMU website will go to WP 3.0, but from what I have seen over the Internet, that should go without problems. As for standalone multi site installation I tested, there are many inconsistencies with WPMU and I think that there will be problems with migration of plugins and themes specifically written for WPMU. I will write more on the migration next week when I do that with xScape.info website.

Speed

Speed. WP 3.0 is much faster in all operations. This is not just subjective view, but I actually measured that speed improved by 15% and memory usage is now lower by 10%. Many changes made to meta boxes on post edit page are great and improve speed and usability, but they will break many plugins that relied on JavaScript to modify default meta boxes. But, I guess that with so many changes one can expect that problems with plugins can’t be avoided.

For the past month I have been improving all my plugins to work with WP 3.0, and so far everything works great. Same goes for xScape themes, and everything works in both normal and multi site installations. More on WP 3.0 coming soon, I will make more tests with menus after Tuesday when Beta 2 is announced. Migration from WPMU will follow next weekend.