I decided to port my blog over to WordPress from my homegrown blogging software, WheatBlog App. In order not to muck up links and such, I am, for the moment at least, leaving everything in place over at wheatdesign.com. Henceforth, I’ll be using wheatblog.com (RSS) as the home of the blog, and probably most of the content that is currently at the old site. But some things will probably remain at wheatdesign.com. I’m still trying to decide how to best divide up content.
Since WordPress can import from RSS, I repurposed some code and created a mammoth RSS file of all the posts over at wheatdesign.com and imported these into the new blog. That did pretty well, but the old DB was pretty crufty, and a good 150 posts or so came through with bad dates (“December 31, 1969”). I hand edited these. I’ve noticed a few other date inconsistencies (i.e. posts being off by one day) and found at least one post missing entirely.
There were quite a few duplicate entries in the wheatdesign.com database (I’ll explain why if anyone cares, but it had to do with some particulars of my own site, not with the blogware itself). WordPress, to its credit, refused to import these. But that also means there’s no longer a 1:1 relationship between the post IDs in the old system and the post IDs in the new system. So, using mod_rewrite to forward the old posts to the new ones isn’t going to work, and I’m not sure what will (other than a massive .htaccess file or a lot of JavaScript redirects).
But that’s all phase three and four type stuff. Phase one is complete. Phase two is to get the comments ported over, which will be tricker and might even have to be done by hand.
If you’ve read this far, the next question is probably, why bother? And I’m still working on the answer to that one, too. But I basically felt that WheatBlog App had served its purpose. Working on that app taught me a good deal about how to code and how to manage a project. But it lacked a lot of what have become standard features of blogware today. And, more importantly, it lacked an API or any sort of plugin architecture.
There was a time when a blog could be an island unto itself, but the web has become a more interconnected place, and blogware has become more extensible and powerful. I don’t have time, these days, to hack in new features. And there’s no point since there’s already a more mature product out there, one I can learn to extend and modify to my own nefarious purposes. This is no slack on the fine work that Josh and Pete did on WheatBlog App. They made it a decent tool. And it’s still a decent tool, but It’s no longer my tool of choice for blogging.
I’ve been using an older version of WordPress for jamesnotjim.com since its inception. But, when I heard about Zeldman’s involvement in the 2.5 version, that pretty much sealed it for me. So, I’ll be working on the theme to give it a look I like. But the engine under the hood is entirely new. And, all this work porting the posts over gets me a excited about blogging again. So look for more frequent updates.
