[11:46:05] Is there a way to implement Z28676? I tried a bunch of ways, but I don't really understand how errors work [11:57:17] It may depend on T406354. I’d be looking at getting the thrown error with Z853 and testing it with Z852, as outlined in Z8521… but I haven’t got round to trying this. [14:23:13] I'm starting to get some fun structures: : https://tools-static.wmflabs.org/bridgebot/27db1c10/file_74216.jpg [14:24:09] https://tools-static.wmflabs.org/bridgebot/b014abd9/file_74217.jpg [14:25:36] Thank you so much! (re @u99of9: ) [14:25:39] https://tools-static.wmflabs.org/bridgebot/91da24c7/file_74219.jpg [14:26:32] Just a detail: is the colour key available somewhere? (and are the colours accessible) [14:28:26] Colours are a hash of the return type at the moment. I haven't yet made a key, but you can find out what a colour is by going to one of the ZIDs with that colour and looking at the return type of the function. I agree they're not very accessible yet, especially when very dark with black numbers on top! (re @NicolasVIGNERON: Just a detail: is the colour key available [14:28:26] somewhere? (a [14:28:26] nd are the colours accessible)) [14:31:38] These tree versions only have one direction out of each function. They look at the composition(s) and find the highest ZID function called by the composition, and point to that. Eventually the paths converge on a root which is either a built-in or a function that only has code implementations. If we can write a composition for that root function, in the next dump it [14:31:38] will be join [14:31:38] ed to a bigger cluster. [14:34:20] Already a fantastic start, I can't wait to see the next version (re @u99of9: Colours are a hash of the return type at the moment. I haven't yet made a key, but you can find out what a colour is by going to...) [14:35:44] It's a bit late here now. 🎃 I'll have another go tomorrow if I get a chance.