Why developing your own blog software isn’t always a good idea

David Goodwin made the following comment on my post about migrating to WordPress:

Despite every ‘web2.0′ language/framework having a ‘create a blog application in 5 minutes’ example, it’s amusing to see even a developer default to Wordpress, which is still written for php4

I can see his point, and I agree that is amusing. Of course I’m not going to miss a chance to get on my soapbox though.

photo credit: -SW-

Coding a blog-style application is easy. In MVC terms you only really need two controllers: one for posts and another for comments. Everything else is just gravy.

Developing a blog-style appliction is hard. You have to support feeds (preferably RSS and Atom). You should support Trackbacks. Comment moderation and spam protection are other must-haves for a site with any amount of traffic. These might be gravy, but without it the meal is incomplete.

Maintaining a blog-style application is even more work. XSS, SQL injection, security, spam. The list goes on and on.

Sure, coding a blog is fun and easy, and something every developer should do at least once. Developing a blog application is much more work, and something that will never be truly complete. Maintaining that application will be far more work than you ever planned.

So the question has got to be: is it worth it?

For me - and obviously lots of other developers - the answer is a resounding no. Our time is better spent creating something new than re-inventing the wheel.

As for David’s comment about PHP4, by all accounts the code for WordPress is not pretty - I’ve been advised not to look under the hood - but that’s not my problem. I do wonder why no-one has managed to re-create WordPress in one of the Web 2.0 frameworks though. Is it because they can’t, or just because they want to try to be different?

This post was written using PyRoom - writing without distractions.