Bots minus Twitter

I turned off the Twitter versions of my fleet of bots last week, and today I went through their various home pages and took out the Twitter timelines - which seem to have stopped working anyway - and replaced them with embedded timelines from their Mastodon incarnations, using Emfed, which was really …

more ...

Bots on Mastodon

One of the nice things about the influx of Twitter users to Mastodon has been peoples' reaction when they find out that most of my bots are on the Fediverse, too. Even though they are still posting on Twitter, I've updated bots.mikelynch.org so that the front-page links go …

more ...

Two kinds of constraint

A fun quality of Mastodon - a social communication platform I hang out on a bit too much - is that individual instantiations can fork its basic form and add odd twists which suit just that community. Oulipo.social notoriously bans that fifth Latin glyph which is most common in many linguistic …

more ...

Glossatory retrained

Glossatory is a recurrent neural net trained on all of the definitions in the lexical database WordNet. I recently retrained it on a GPU, increasing the number of hidden units to 512 and the layers from 2 to 3. The results are now a lot more grammatical, and there's a …

more ...





Neuralgae Updated

I've updated the algorithm behind neuralgae. It's still the same basic principle: render a deepdraw image based on some random categories, feed the results back through the classifier, and then repeat with the new categories. But the draw part of the cycle now uses the weights from the classification stage …

more ...

A Mighty Host

As an exercise in Haskell, I wrote a monad which threads random-number generation state through a context-free grammar, allowing the creation of arbitrary yet valid sentences. This was kind of clever.

Then I used it to generate a series of descriptions of bands of warriors along the lines of Homer's …

more ...