I’ve given my V7 redesign project the unofficial tagline “The Procrastination Destination” since the significant progress it’s seen in the past few months has come mostly in stolen moments, some of which turned into extremely productive (and perhaps troublingly obsessive) deep dives. This recent movement has been pretty non-linear, and the tasks in play are all interdependent enough that none of them are really done until all of them are, but I seem to be approaching an inflection point where a number of important questions will be answered. Here’s the current order of operations, all of which I’ll discuss in more detail soon:
- Metadata structure: I’ve honed a front-matter format that will accommodate a wide variety of entry types and will structure their metadata in a way that will give me a lot of options for how to present the entries. It remains to be seen if Eleventy can interpret the data the way I want, but I’m just about ready to get a test instance up and running to validate assumptions and solidify the format. I’m cautiously optimistic!
- Twitter archive reformatting: If you’ve been following my journey on this redesign, you know this item is huge. After years of wondering how the hell I was going to wrangle my thousands of tweets for republishing on my own site in my own specific way, back in December I started poking at my downloaded Twitter archive with Node.js. I came back to it last week with a vengeance and I’m extremely excited to say that as of last night, it’s done! Over 6,000 tweets neatly formatted in Markdown files using the front-matter structure mentioned above. If that front-matter structure changes, my tweet reformatting process will need some tweaks, but the heavy lifting is done, and it feels amazing.
- Article layout framework: I wasn’t sure how far I wanted to go with layout options for my posts in V7, but a few months ago, I started hammering out a modified rewrite of the expansive article layout framework I originally made for ProPublica. I still have a bunch of QA to do, but looking at the code now for the first time in weeks, I’m stunned by how compact the rewrite is.
- Local “CMS”: One thing that’s less than ideal about static site generators is that the relative laboriousness of assembling entries in Markdown files—especially adding the date and time by hand—can discourage smaller posts. It just doesn’t seem worth the effort; there’s a vast chasm between a Twitter-style tap-type-send UI and an SSG-style create-and-format-and-populate-a-text-file workflow. So I’m making a cozy little local UI that will look something like Craft or WordPress, live in my browser, and generate Markdown files for me.
- Graphic design: Lest I get too bogged down in code and information architecture, I’ve been thinking about the site’s look and feel, too. I recently had the epiphany that the design of the small portfolio site I put together a little while back (which I’ve mostly just been sharing privately with prospective clients) could actually be a great starting point for V7 to evolve from.