George Svarovsky

> browser-level is really slooooow > in the demo, I actually use memory-level, but stream the state to the Cache API. It's faster to snapshot the whole thing than to...

Theoretical: transactions including an agreed datatype declaration may get fused with transactions using it, causing new clones to fail to apply the uses correctly

To include - [ ] all JS compliance (transport security, agreements, shape contraints, text), suitably expanded - [ ] chaos tests for the above

List representation in RDF/JS

1

> Lists in m-ld, using the @list JSON-LD keyword, do not transform to RDF Collections. They are more like RDF Containers (but still, not quite 🙂). Plain RDF Collections don't...

If a clone broadcasts operations against an unknown shared data type, a clone could store the operations without processing them. This would allow for generic tools like the Gateway and...

**thought:** The fact that this config requires so much massaging makes me suspect that the shape of the config may not currently be ideal. _Originally posted by @Peeja in https://github.com/m-ld/m-ld-gateway/pull/12#discussion_r1269777965_

Already exists as a test fixture, just needs to be made robust. https://github.com/m-ld/m-ld-js/blob/edge/test/datatypeFixtures.ts

Currently only available for [TSeq CRDT](https://edge.js.m-ld.org/classes/tseqtext.html)

Property-based testing for core CRDTs

1

Using e.g. https://github.com/dubzzz/fast-check/