I’m not sure whether you did notice, but as the amount of whimsy posts grew, the page load times became increasingly intolerable. This is because this blog was based on a Python script dynamically generating its responses from plain text files, an approach I’ve chosen because I grew frustrated with the static blog generator solutions I’ve evaluated at that time.
Hyde is different. It doesn’t try to be particularly smart about cache invalidation techniques or to impose yet another irregular templating language on you. Adding support for translating my existing posts was as simple as writing a few lines of Scheme. That’s why I went the full route and rebuilt my blog in nearly all of its entirety with it. Full sources are available on Github.
Given the low amount of hits for category-specific views, feeds and pagination links, I’ve decided to kill off these features. Unfortunately all old links are broken, so please update to the new Atom feed URL in your news feed reader.
edit: Nginx is pretty amazing:
location / { rewrite ^/feed/?(elpa|emacs|other)?$ /feed.atom permanent; rewrite ^/post/(.*)$ /posts/$1.html permanent; }
In other words, nearly all links will keep working. I could perhaps add some manual exceptions for the cases where the different slugification algorithm did make a difference in URL…