posted 2013-01-10 1335 by masen
After a weekend hack-a-thon which extended to this morning (and perhaps throughout the rest of my life), I deployed my latest version of this site this morning.
Although it doesn't look much different, the way it's edited, generated, and rendered is mostly new. As I describe on my projects page, I've been working on going static since about July, but I've been working especially since last Friday.
The result is that now my site doesn't require python on the webserver. Although python is used to build out all the html pages, all of the site is assembled once when a change is made and then served without any dynamic code running on each request. This is a great benefit to scale as well as simplicity. However, the largest benefit comes from the archiving of content written here, from here on out. Instead of being locked away in a proprietary SQLite database, all content, templates, code, and files will be versioned on github in plaintext, human-readable format.
With my excitement to move to the new platform, there are some neglects:
- RSS feeds no longer update (i'm working on this)
- Old permalinks like "/blog/3" still point to the old CMS and old templates and old links. I have migrated the old url data to my new files, but haven't yet created a good redirect method. (Right now leaning toward mod_rewrite's [L] option)
- Comments are gone: the fb social comment system was never one that I liked. Since I didn't want to sign up for a fb API key, it was a pain to manage and no one really used it. A new comment system will be written that meshes better with the feel of the site (i.e. is static based) and allows anon comments.
- Documentation for the generation platform is out of date and sparse. I need to write it up before it leaves my head, or I may have to scrap the site again and start over. Ha.
Thanks loyal fans, Masen