Post BAdzQhdA1xCtIiYCuG by icedquinn@blob.cat
 (DIR) More posts by icedquinn@blob.cat
 (DIR) Post #BAdzQhdA1xCtIiYCuG by icedquinn@blob.cat
       0 likes, 0 repeats
       
       @shibao :blobcatwaitwhat: you needed a robot to add a closure table?
       
 (DIR) Post #BAdzwI20k8UhAnEhTk by icedquinn@blob.cat
       0 likes, 0 repeats
       
       @shibao mostly use it to discover that closure tables exist in the first place (although i learned that from stack, so, neural search more than LLM :blobcatgoogly:)we need more term-rewrite systems in existence. this is exactly the kind of thing iMatix GSL was meant to do and term rewriting + datalog is exactly the kind of thing that is very good at "machine model of data -> generate boilerplate."haven't quite learned maude to try the idea sadly. just have ideas rolling around for something where you define the data in some basic spec and then decide how to shape it later. it's what i did for the matrix server. defined everything as a relational graph with XML and used templates to make the actual tables