Spent most of Saturday doing something I've been putting off for probably three years now: actually going through the old drafts folder for this blog. Not the published posts, the drafts folder, the graveyard of half-finished things I started and abandoned. There are 214 of them. Two hundred and fourteen. Some are just a title and nothing else ("why nobody talks about..." — talks about what, past me? you didn't say). Some are 1,800 words that just stop mid-sentence like I got up to make coffee in 2014 and never came back.
I didn't set out to write a nostalgia post today, but here we are.
This blog started in November 2011, which means it's closing in on fifteen years, and going through that folder is like an archaeological dig through my own opinions. There's a draft from early 2012 arguing that Retina displays were a gimmick that would never make it to laptops because the pixel density math didn't make sense for battery life. I was extremely confident about this. I was also completely wrong within about four months. There's another one, undated but clearly from around the Siri launch on the 4S, where I'm annoyed that voice assistants "will never understand context" and predicting the whole category dies out by 2015. I want to reach back through time and just shake myself a little.
The thing that actually got me, though, wasn't the bad predictions. Everybody who writes about tech in public racks up a pile of bad predictions, that's basically the cost of admission and I've made my peace with it. What got me was how many of the abandoned drafts weren't wrong, they were just boring: half-finished reviews of routers and USB hubs and a truly baffling number of posts about RSS reader migrations. I counted four separate unfinished posts about switching RSS readers. Four. Apparently every eighteen months or so, past-me would get fed up with whatever reader I was using, start typing up a whole essay about it, and then just... not finish, presumably because I got distracted setting up the new reader instead of writing about the old one.
I kept about a dozen drafts and deleted the rest, which felt more violent than it should have for a folder of text files nobody was ever going to read. There's something oddly final about deleting a draft versus just letting it sit there unfinished forever. An unfinished draft is technically still possible. A deleted one is a decision.
One thing I didn't expect: reading old posts from 2011 and 2012 made we want to write more like I did back then, not less. Early-blog me was way less careful. I'd just say a thing was bad, full stop, no three-paragraph hedge before and after it. I miss that a little, honestly — somewhere around 2016 or so I think I started writing for an imagined audience of people who'd disagree with me on Twitter, and it made everything softer and more qualified. Nobody asked for that. I'm not going to pretend I'm fixing it in one blog post, but it's on my mind.
Also, purely as an aside because I can't let it go: I found the original hosting invoice from when this thing started, and I was paying more per month in 2011 for a shared hosting plan that could barely survive a hundred concurrent visitors than I pay now for something that could handle actual traffic without blinking. Inflation is supposed to make everything more expensive and yet here's a case where the opposite happened so hard it's almost funny. Compute got cheap while everything else got expensive. Weird decade.
If you keep any kind of personal archive (a blog, a notes app, a folder of old projects), I'd genuinely recommend doing this at some point, going back through the unfinished stuff instead of just the polished output. It's uncomfortable in a useful way. You get to see the shape of what you were interested in before you decided which parts were worth finishing, which is a different and honestly more honest picture than the "best of" version everyone else sees.
Anyway. Draft folder's down to about thirty now. Give it a year and it'll be back up to two hundred, I have no illusions about that. But it felt good to look.