So it's been a while since I've been on the Fediverse. I stopped working on Pterotype - some combination of burnout, not managing expectations and getting really tired of writing PHP. And I was embarrassed about that. I had spent so much time talking up the projec that it felt like it had just become my Fediverse identity.

So that's that. But I'm going to give it another shot. I'm not working on Pterotype anymore! I've moved on to something new: I'm building a database! Intro here:

And to be clear I am still super excited about ActivityPub and the Fediverse! There's definitely a future where I do more ActivityPub projects. Just not right now - first I gotta see where this database thing takes me.

Show thread

@jdormit While you were gone, someone asked about exactly such a database where one would be able to see how it looked in the past.

This is basically what Wikipedia does. Didn't they already build such database or do they do all of that at the application level?

Wikipedia is built on the MediaWiki project, which uses MySQL I believe:

So you're correct, all the change tracking for Wikipedia is done at the application level. UnifyDB tracks changes to records by default at the database layer, making it much easier to write history-aware software like MediaWiki.

Datomic has a couple of gaps that I'm trying to fill - a production Datomic deployment is very complex, and Datomic only has Clojure and Java clients. I'm designing unifyDB to be able to run in various deployment scenarios (single-node or distributed, plugging in to multiple backends for data storage/message queuing/caching) and I'm planning on releasing clients in all the mainstream languages. I love Datomic but I think there's a reason it never became more widely used in the industry.

Sign in to participate in the conversation
Mastodon for Tech Folks

This Mastodon instance is for people interested in technology. Discussions aren't limited to technology, because tech folks shouldn't be limited to technology either!