If you make use of any of my plugins then it’s likely you’ve noticed several fundamental changes, along with some tougher system requirements. I think now is a good time to actually write about these changes and discuss the reasons behind them.
I’ll start by making an admission: the release of WordPress 2.7 was a big headache for me. There, I’ve said it. I’ll continue on by stating that I love all the new features, and appreciate the updated interface and the amount of work that has gone into it (and continues to go into it). However, as a developer with around 20 plugins for WordPress, the changes in 2.7 were substantial enough that that it took considerable effort to get all my plugins working ‘properly’ again.
Since updating everything I’ve taken the decision to baseline all plugins to WordPress 2.7. Most of my plugins have supported versions from 2.0 onwards, and this was gradually clogging the code and creating a testing nightmare. WordPress 2.7 provides features that ease the work for plugin developers, so this seemed a sensible point to fix development from. In truth the plugins should still work in 2.5 and 2.6, but there may be some inconsistencies.
No, I’ve not taken it this far… yet. While I don’t necessarily agree with Matt’s reasons why WordPress won’t be using PHP5 anytime soon, I can understand the desire not to alienate a large percentage of users. However, as a developer it is particularly frustrating not to be able to make use of ‘new’ features (and by new I mean several years old) of PHP5 that would simply life. It would be interesting to see figures for usage levels of PHP4/PHP5 for WordPress users to weigh up if it’s worth taking the plunge or not.