Nothing makes you write the next one

Every discussion about a new blogging platform ends up being about the tech which is the least interesting part of it.

What stack it uses, whether it’s static, what the markup looks like when you view source. All of that is secondary to whether you actually use the thing to build up a body of writing that shows what you know and what you think over a period of years. That’s what a blog is. It isn’t the tags and it isn’t the stylesheet. It’s the content sitting inside them, and any platform discussion that doesn’t get to that has stopped short of the point.

Where the next generation framing is right is in going after the friction between having a thought and having it published. That’s the real target and it’s worth attacking. There are three things any platform has to make trivially easy before it has a chance and the industry has comprehensively solved one of them.

The first is setup and maintenance. If my laptop dies and I can’t clone a repository and run two commands to be exactly where I was, it doesn’t work. That constraint rules out a surprising amount of what gets built and it’s the one thing the static site generation world has genuinely nailed.

The second is publishing itself. Most blogs run on Markdown with either frontmatter or a specific file path doing the work which is fine and puts most of the cognitive load onto the writer. You have to remember the shape of the frontmatter, and where the file goes, and what the slug should be, and whether you care about the URL structure enough to fight it. I use a static generator that’s very good and if I worried less about URLs than I do it would be close to perfect. But I do worry about them so it isn’t.

The third is the one nobody has ever solved and the reason most blogs die. There’s nothing that pushes you to write.

That’s the kicker. Every platform is built around the moment you already want to publish something and the actual bottleneck is the six weeks before that when you don’t. A tool that suggested posts you ought to write, based on what you’ve written before and what you keep circling back to, would produce more than any improvement to the publishing step. So would something that noticed you’d written and said so.

That sounds trivial and it’s slightly embarrassing to admit. It’s also the mechanism every social platform is built on and the reason people who never write anything anywhere will post four times a day to a timeline. The feedback isn’t a nice extra on top of the writing. For most people it’s the thing that produces the writing at all and a blog is a format that deliberately removes it.

One or two platforms have come close to taking that seriously. None of them have gone far enough and the ones that try tend to drift towards the timeline shape and stop being blogs. That trade might be unavoidable, and if it is then the honest version of a next generation blog platform is one that admits the hard problem is motivation rather than markup, and does something about it anyway.

The workaround most people land on is external and slightly undignified. You tell somebody you’re going to write a thing, or you set up a schedule you’d be embarrassed to break, or you find one reader whose interest you don’t want to lose. None of that is a feature of any platform and all of it works better than any platform, which tells you roughly where the real problem sits.

The tooling was never the bottleneck. It was just the part that could be improved by writing code.