datasette 1.0a37
Simon Willison's Weblog Simon Willison
Simon Willison pushed out Datasette 1.0a37, another alpha build on the long road to a stable 1.0 release. It's a small update, but each alpha chips away at what's held back one of the more useful open-source data tools for years.
Based on reporting by Simon Willison's Weblog, Simon Willison — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Simon Willison dropped a new alpha of Datasette on July 14th, tagged 1.0a37, continuing a release cadence that's been going for what feels like forever at this point. Datasette, if you haven't run into it, is Willison's open-source tool for turning SQLite databases into browsable, queryable websites — you point it at a database file and suddenly you have a web app with search, filtering, and a JSON API, no extra code required.
The project has been sitting in alpha purgatory for a long stretch now, climbing through version numbers like a1, a2, and now a37 without quite crossing the finish line to 1.0. That's not necessarily a bad sign. Willison has been famously deliberate about Datasette's plugin architecture and API stability, and a project used in production by journalists, researchers, and hobbyist data nerds benefits more from careful iteration than from a rushed version bump.
What's notably absent from this particular post is much detail on what actually changed in a37. Willison's blog entries for these alpha releases are often terse changelog pointers rather than full writeups, and this one leans that way too. Longtime followers of the project know the drill: check the GitHub releases page for the actual diff, because the blog post itself is mostly a marker that another build has shipped.
Still, the steady drumbeat of alphas says something on its own. Datasette hasn't stalled out or gone quiet — it's still getting attention, still moving toward that 1.0 milestone, even if the path there looks more like a slow crawl than a sprint. For a tool that's quietly become infrastructure for a lot of small-scale data journalism and open government data projects, that persistence matters more than a flashy changelog entry would.
My take — AI-written commentary, not fact-checked reporting
I like that Willison ships these tiny, unglamorous alphas instead of going quiet for six months and dropping a monolithic 1.0 with no warning — that's the boring, correct way to build infrastructure people actually rely on. My only gripe is that terse release posts like this one make it hard for casual followers to tell if a37 is a big deal or a typo fix, and in an ecosystem this noisy, a one-line changelog note would go a long way.
Read more about this at: Simon Willison's Weblog
