❤️❤️ the responses
Quote Tweet
PL twitter: quick, what's the scariest screenshot you can find of a paper with an overwhelming amount of inference rules?
Show this thread
1
1
20
Conversation
Serious question: do people (outside of the paper authors) actually understand what these rules mean?
7
4
I consume PL papers as part of my job, and I do actually get a lot of value out of these rules! Sometimes as a reader they can be hard to parse at first. Like anything it takes a bit of taste to use them effectively as a form of communication.
1
10
This said, I *do* think it would be cool to take a step back and think about how we design PL notation in a way that makes communication easier. I definitely think it's an unfortunate barrier to entry, and can be hard to deal with even if you are experienced with it.
3
6
I wonder how we could improve it. part of the issue is the tension between concision and completeness, I think. as indecipherable as judgment rules &c can be at first, I don't think people would prefer, like, 10 pages of English text thoroughly explaining the mechanism.
3
7
maybe part of it is just consistency. seems like many authors like to use slightly different notation, so you have to be careful to read their specific flavor notes prior to engaging with the material. maybe we need a PL standards office 😂
2
1
My idea for improving it would involve playing around with documenting language specifications in a clear, concise (potentially interactive) way. Jumping into standards seems a bit premature? I dunno 😅
oh for online docs, interactive would be so cool! but for as long as papers remain stuck in PDF/LaTeX, I fear we are constrained to some kind of math-y Greek notation lest our semantics become longer than the paper content. but, I dunno, there's gotta be something better...!
1
Yeah that's why I think language docs would be a fun place to start at least. I want to be able to build an interactive, hyperlinked specification for Pikelet and Fathom for example - that doesn't need to be a paper, afaik
1
2
Show replies