3 ms·
I was hoping it would reference roc-lang.org, which is actually doing in-place mutation for efficiency, but I didn’t see any mentions :-(
by zellyn 3y ago
I was hoping it would reference roc-lang.org, which is actually doing in-place mutation for efficiency, but I didn’t see any mentions :-(
- agumonkey 3y agoI stopped monitoring roc after feldman's early youtube chats, but it seems evolving in term of team size.
- zogrodea 3y agoI don't really mind Roc-lang not being mentioned since Koka is referenced with its novel idea of functional-but-in-place and Perceus Reference Counting which are the theoretical basis of Roc's implementation[0]. So it doesn't really bring anything theoretically unique to the table in comparison, although I'm interested in Roc as a practical language because of this too. [0] Ctrl-f "Perceus" on the following page and you will see a paper about Koka mentioned. https://www.roc-lang.org/ https://www.roc-lang.org/
- ghayes 3y agoSurprised there hasn’t been a well-received HN post on Roc-lang. It looks really interesting.
- sparkie 3y agoIn-place mutation in a functional language dates back as far as 1994, when it was introduced for Clean in the paper 'Guaranteeing safe destructive updates through a type system with uniqueness information for graphs'[1]. The FP2 paper references a subsequent 1995 paper from the same authors, 'Uniqueness Type Inference'[2]. More publications about Clean here: https://clean.cs.ru.nl/Publications https://clean.cs.ru.nl/Publications [1] http://www.mbsd.cs.ru.nl/publications/papers/1994/smes94-guaranteeing.pdf http://www.mbsd.cs.ru.nl/publications/papers/1994/smes94-gua... [2] http://www.mbsd.cs.ru.nl/publications/papers/1995/bare95-unitypeinference.ps.gz http://www.mbsd.cs.ru.nl/publications/papers/1995/bare95-uni...