Version 2

The second coming of my blog. Technically the third, but that's another discussion about semantics.

I've been wanting to re-work my blog for some time now, for a variety of reasons, but never found the time to do so. A couple sparse weekends gave me the opportunity to brainstorm, and a few recent cross-country flights provided the time.

Things I Wanted

static content - no PHP, no SSI, no database, easily updated
content to be stored on GitHub, both for backup purposes and to allow for user-submitted edits, pull requests even
ability to migrate the site to another host, with ease; tar -zcf and go, ideally
ability to keep the old blog content intact, that is to say, in the same place; no doubt the search crawlers would sort out any changes eventually, but dead links suck
to write in text, Markdown, or HTML, either remotely or on-server
mobile-friendly, which either means dead simple HTML or mobile-aware CSS
renewed focus on essays, technical HOWTO, and photography; ditch the rest for now

Things I Don't Care About Right Now

attractive site design: to be handled by drop-in CSS later
comments: because they suck
search: can add Google site search later
tags: perhaps auto-generated later, but not now
RSS: perhaps auto-generated later (bash, Perl modules, etc), in the meantime there are plenty of site-to-RSS services out there, RIP Google Reader, etc. For now just track categories and published/updated dates.

Decision

After fiddling with a lot of different engines, attempting to import the WordPress entries from XML, I just gave up. Pelican came close, except that the template system was monstrous.

I decided I was being stupid.

Why re-import all the old stuff? Why not leave mirror the WordPress site using wget and serve up the static HTML?

So that's what I did. Could you tell?

The HTML is a little larger on-disk than it needs to be given all the tag and category pages, but load times are faster and load on the CPU has dropped considerable. Both WordPress and PHP have a history of security vulnerabilities, so I'm glad to be off both of them.

With the old site mirrored, I rolled my own blog engine. About 300 lines worth of Bourne shell and awk scripts, which shouldn't be a big surprise to anyone who knows me.

The upsides: it does what I want, nothing more, with a KISS kind of elegance.

The downsides: to be determined, I'm sure.

Inspiration

Tim Bray's copyright page, to preserve copyright but also to license under Creative Commons, something that I've been wanting to migrate towards for a while.
Dave Gamache's work on Skeleton, which (I hope) is a drop-in CSS styling thingy that will JustWork™ for me in the near future.

Last Modified: 2020-08-09

Home | All Entries | Search | Errata