Deroos: But its a hard-line comparative
Emch: That’s not really what I meant by saying “I don’t consider Python to be a productive language.” Mostly, writing prototypes in Python and refactoring them is a pain, almost as much of a pain as writing the same thing in Java.
Obar: Sure, you don’t have types, but that doesn’t make it force some arbitrary structure on you any less
Deroos: Huh, prototyping and the ease in refactoring is exactly why I value python
Deroos: I can get something to work and easily rearrange the working parts in to pretty code
Deroos: I can do this faster than in any language
Lauchaire: Deroos: have you tried any Smalltalk environment yet?
Seikel: Or some modern Lisp dialect, like Racket or Clojure?
Deroos: I think on the dynamic programming side, I am utterly served by Python
Deroos: And have been for more than a decade
Deroos: I’ve only recently been looking for something that will help me out statically
Deroos: And now I’m totally served there too
Deroos: But yes I have used smalltalk in an exploratory sense
Cataldi: Well, if you’re comfortable with your technology choices, that’s good for you. I certainly am not, and so far everything I’ve tried ****s
Deroos: Well, you know what I’ll say
Deroos: Also, for games, es6 is taking from my Python/Nim mindshare
Deroos: Es6 + lodash + phaser, is probably the best game development experience I’ve had so far in terms of the tech
Prause: Dolby: Any special features you’re looking for?
Redinbo: Madsy: computer-aided construction of programs. So, a mix of programming-by-demonstration, and interactive iteration REPL/Worksheet-like. A lot of these are being solved in tools like Lamdu and Unison, but there’s much that they don’t solve yet. In particular, I’d like the computer to help me solve problems in the most efficient way possible, understand
Kalfayan: Any solution I or someone else comes up with, and compare different solutions contextually to pick the best one. This requires refactoring the entire codebase to be a trivial operation, of course.
Shodunke: I’ve a question about Edge detection that I’ve written here https://www.reddit.com/r/javascript/comments/3jzsyw/im_not_sure_why_the_edge_detection_is_different/ , basically one version is using steps here https://github.com/geo7/reflection/blob/reflection-with-the-steps/main.js#L91 and the other is just checking if x or y exceeds width or height here https://github.com/geo7/reflection/blob/reflection-with-previous/main.js#L96 , I’m not
Mosby: Sure why they’re reacting so differently, and I’ve included videos of that here https://www.reddit.com/r/javascript/comments/3jzsyw/im_not_sure_why_the_edge_detection_is_different/
Dobek: Dolby: Have you looked at Coq and Idris?
Udicious: Madsy: Kay actually wrote a lot about similar topics.
Nigro: Coq is generally used as a proof ***istant, but it is just as powerful as any programming language
Hollering: Madsy: yes, but Coq is a theorem prover, and Idris is a wanna-be-practical dependently typed language. They don’t achieve any of the things I want.
Gearing: Madsy: what web frameworks are written in coq
Nuhn: Sure, more tooling could make them better, but there’s a lot of work to be invented there. DT isn’t practical yet, by a long shot.
Massingill: DT would solve one awful problem with programs though: the overt reliance on social conventions for sharing code. Constraints would be just types, and in that way you don’t need versioning anymore
Garretson: Dash: None I know of at least. Why did you ask?
Gearing: Madsy: why would anyone take a programming language without a web framework seriously?
Kampmann: Dash, people who uses C or C++ usually takes these languages seriusly, or so I hope
Gearing: Dolby: maybe on a small scale, but certainly not in general :
Kesek: Dash: Because every language starts small and without library implementations