Skip to main content
11 events
when toggle format what by license comment
Dec 11, 2016 at 0:05 comment added Justin Hilburn If you are just trying to connect programming languages and mathematics I would learn the typed lambda calculus and look at mathoverflow.net/questions/903/…
Dec 10, 2016 at 22:57 comment added Justin Hilburn The different kind of semantics are explained here en.wikipedia.org/wiki/Semantics_(computer_science)
Dec 10, 2016 at 22:51 comment added Justin Hilburn I would hold off on learning domain theory for a while. Church doesn't make any effort to give a denotation to lambda terms in his book. He gives an operational semantics instead (the algorithm for reducing a lambda term to a normal form).
Dec 10, 2016 at 22:29 comment added Polymer You've convinced me this is the correct answer, even though I'm not sure I understand it. Generally when I'm trying to understand properties of a function definition, the "language" isn't directly important (homomorphisms are conceptual, not syntactical). I evidently tried and failed to separate the syntax of lambda calculus from the concept of a certain function I thought was introduced. Thank you for your time. Any recommendations for learning Domain theory would be appreciated.
Dec 10, 2016 at 22:07 vote accept Polymer
Dec 10, 2016 at 21:58 comment added Justin Hilburn It is just like how there are many different models of the theory of groups, i.e., particular groups. It doesn't make sense to say that an element of a group is "really" a matrix even though that might be true in a particular model.
Dec 10, 2016 at 21:47 comment added Justin Hilburn As you said, one way of doing this makes lambda terms into elements of a domain.
Dec 10, 2016 at 21:46 comment added Justin Hilburn I think you need to be more careful about the distinction between a language and a semantics for that language. Lambda terms aren't anything but strings of symbols obeying a certain grammar and rewriting rules. One may attempt to associate mathematical objects to lambda terms (give a denotational semantics) but there are many ways to this.
Dec 10, 2016 at 19:46 comment added Polymer So if I understand this answer correctly. Untyped lambda calculus is deceptively complicated. The lambda terms are "really" continuous functions of some odd space. Is that correct?
Dec 10, 2016 at 18:55 history edited Justin Hilburn CC BY-SA 3.0
added 56 characters in body
Dec 10, 2016 at 18:47 history answered Justin Hilburn CC BY-SA 3.0