Fork me on GitHub
#clojure-boston
<
2022-05-16
>
Chris O’Donnell00:05:43

@mjdiloreto I am! Wilker explains it better than I could in the beginning of this talk if you're interested in some more tech talk. https://youtu.be/YaHiff2vZ_o

Matthew DiLoreto12:05:29

@U0DUNNKT2 Ah! I was trying to search for your name but I guess slack couldn't index based on your username alone. Thanks for the link, that's exactly what I'm looking for!

Matthew DiLoreto21:05:54

Ah, so at 16:55 in the LinkedIn video he talks about how graphql does not solve the "data navigation problem" in general because the "read types" (the entities returned in a graphql query become "fat" (one type contains all possible information), while the "write types" (the types the client sends in mutations) "explode" (you create a new mutation type for every kind of mutation you want to do. Pathom, on the other hand, considers each attribute as the basic element of the protocol, so you don't have to arbitrarily define containers (the entities you were talking about in your talk) for sets of attributes, you can query or mutate only what you want without referring to entities. Let me know if I missed anything, but this is fantastic! Thanks a bunch for the pointer @U0DUNNKT2

Chris O’Donnell22:05:12

Sounds about right 👍

Chris O’Donnell00:05:36

And I'm glad you enjoyed the talk. 🙂

Matthew DiLoreto12:05:58

It was fantastic! I really appreciated when you pulled up the fulcro rad demo and said "I'm not going to walk you through this, do that on your own time". Got straight to the point and that was developing with fulcro rad, so thanks for that 👍

👍 1