Long Term Project
It’s been five months give or take since I decided to move my blog to a new domain and to migrate all my entries to Jekyll.
(Incidentally, Jekyll was used as the front-end for healthcare.gov. But it wasn’t the front-end where all the major problems lay….)
I’d initially decided to try and use Ghost (you’ll notice that my current Jekyll theme is essentially a clone of the default Ghost theme) and I even went as far as to subscribe to DreamHost’s VPS service just so I could run node.js apps, but I was daunted given my complete lack of familiarity with node.js and I didn’t really find an easy way to migrate all of my old blog entries.
So I decided to go with Jekyll because it’s written in Ruby and because it tickles my nostalgia, reminding me exactly of those days when I published blog posts by running XML files through XSLT and using Makefiles to handle selective rebuilding.
It also reminds me somewhat of Blosxom, which was the first dynamic blog engine I had ever run, and which also uses flatfiles with embedded metadata instead of using a RDBMS as has been the more recent fashion.
So I’ve got a sense of going full circle.
I’ve been slowly manually converting my old, old blog posts that I had originally written in XML into Markdown + YAML. I’ve also been editing the entries that I migrated from Mephisto and Typo (now called Publify) via Jekyll’s built-in importers (none of which handles metadata particularly well—I’m not entirely sure how to extract Mephisto’s tags from the database and the way Ruby handles timestamps is kind of messy.) I also wrote a simple script that added YAML frontmatter to all my Blosxom blog entries.
Granted, there is a part of me that thinks that I should just leave my old entries where they lie and make a clean break. But partly because I want to save a little money on domain names and hosting and partly because ヾA彡 has made me paranoid about how visible my existing blog is on the web, I think I’m gonna have to go for it.