5 ms·

I attended TechMesh 2012 in London. The creators of Haskell were there, and there was a number of other Haskell talks. They were almost all about how to model s

by augustl 14y ago

I attended TechMesh 2012 in London. The creators of Haskell were there, and there was a number of other Haskell talks. They were almost all about how to model state, I/O, etc. One of the Haskell creators also created Quickcheck, a generative testing tool where tests are written in Haskell. He talked about how he found state related bugs in Riak. Rich Hickey, author of Clojure, was there, he keynoted about state. I'd argue that the whole point of Clojure is to manage state. Specifically, have a model for how to work with the passing of time. In other words: functional programming and Lisp does not translate to ignoring state.

dschiptsov 14y ago
How to model a state is a meaningless question without asking state of what. Same is true for I/O. You must start from a protocol. Most of people probably underestimate how many brilliant people contribute to what we could call a modern Lisp. With all respect, Clojure is nowhere near in terms of sanity, consistency and uniformity. I don't event want to argue about Haskell. Seems like words "monad" is the same as words "chakra" or "dharma" in a popular culture.))
- irahul 14y ago
  > With all respect, Closure is nowhere near in terms of sanity, consistency and uniformity. It's Clojure http://clojure.org/ http://clojure.org/ Closure is Google's js library and tools https://developers.google.com/closure/ https://developers.google.com/closure/
  - dschiptsov 14y ago
    Yeah, thank you. It was a typo..
- PuercoPop 14y ago
  Ir you are talking about syntax uniformity, lisp is not very regular also http://xahlee.info/UnixResource_dir/writ/lisp_problems.html http://xahlee.info/UnixResource_dir/writ/lisp_problems.html Although I agree that clojure has is warts. For example clojurescript uses the numerical stack of JavaScript. So you get different and unexpected results than when running on the JVM
  - rafcavallaro 14y ago
    lisp is much more regular in syntax than most popular languages, and Xah's page that you point to amounts to little more than saying "I don't know lisp well, so I find various aspects of the language - backquote, quote, unquote-splicing, etc. - confusing."
  - cgag 14y ago
    One of Clojure's goals is to interop well with its host platform, which it does in both cases. I'm not sure I'd count that trade off as a wart.
  - jhuni 14y ago
    If you have issues related to the syntax uniformity of certain Lisp dialects, please speak for yourself rather then linking to that troll Xah Lee. Xah Lee recommends that developers subjugate themselves to Wolfram Research which controls the proprietary program Mathematica. Even though some Lisp dialects have a couple of quoting operators like ' and ` at least they don't have infix operators like + and * that Mathematica has. Besides, a few quote operators is no basis to say that Lisp syntax is "not very regular."
- mbrock 14y ago
  > Seems like words "monad" is the same as words "chakra" or "dharma" in a popular culture. Are there any more interesting concepts you stubbornly dismiss?
andrewcooke 14y ago
[aside] that looks like it was a very interesting conference. i can't find much related info. will it be annual? are they spreading to other places? who was behind it? [edit:] it seems to be an indie thing http://techmeshconf.com/techmesh-london-2012/contact/ http://techmeshconf.com/techmesh-london-2012/contact/
- augustl 14y ago
  They're redoing it this year, and considering doing one in New York this year as well. I think the organizers is a loose collection of Erlang companies (Trifork, Erlang Solutions, probably more). IIRC they organized an Erlang conference and decided to relaunch it as a general FP conference. It was _really_ good, you should definitely check it out :)

by augustl 14y ago

dschiptsov 14y ago

How to model a state is a meaningless question without asking state of what. Same is true for I/O. You must start from a protocol. Most of people probably underestimate how many brilliant people contribute to what we could call a modern Lisp. With all respect, Clojure is nowhere near in terms of sanity, consistency and uniformity. I don't event want to argue about Haskell. Seems like words "monad" is the same as words "chakra" or "dharma" in a popular culture.))

irahul 14y ago
> With all respect, Closure is nowhere near in terms of sanity, consistency and uniformity. It's Clojure http://clojure.org/ http://clojure.org/ Closure is Google's js library and tools https://developers.google.com/closure/ https://developers.google.com/closure/
- dschiptsov 14y ago
  Yeah, thank you. It was a typo..
PuercoPop 14y ago
Ir you are talking about syntax uniformity, lisp is not very regular also http://xahlee.info/UnixResource_dir/writ/lisp_problems.html http://xahlee.info/UnixResource_dir/writ/lisp_problems.html Although I agree that clojure has is warts. For example clojurescript uses the numerical stack of JavaScript. So you get different and unexpected results than when running on the JVM
- rafcavallaro 14y ago
  lisp is much more regular in syntax than most popular languages, and Xah's page that you point to amounts to little more than saying "I don't know lisp well, so I find various aspects of the language - backquote, quote, unquote-splicing, etc. - confusing."
- cgag 14y ago
  One of Clojure's goals is to interop well with its host platform, which it does in both cases. I'm not sure I'd count that trade off as a wart.
- jhuni 14y ago
  If you have issues related to the syntax uniformity of certain Lisp dialects, please speak for yourself rather then linking to that troll Xah Lee. Xah Lee recommends that developers subjugate themselves to Wolfram Research which controls the proprietary program Mathematica. Even though some Lisp dialects have a couple of quoting operators like ' and ` at least they don't have infix operators like + and * that Mathematica has. Besides, a few quote operators is no basis to say that Lisp syntax is "not very regular."
mbrock 14y ago
> Seems like words "monad" is the same as words "chakra" or "dharma" in a popular culture. Are there any more interesting concepts you stubbornly dismiss?

andrewcooke 14y ago

[aside] that looks like it was a very interesting conference. i can't find much related info. will it be annual? are they spreading to other places? who was behind it? [edit:] it seems to be an indie thing http://techmeshconf.com/techmesh-london-2012/contact/ http://techmeshconf.com/techmesh-london-2012/contact/

augustl 14y ago
They're redoing it this year, and considering doing one in New York this year as well. I think the organizers is a loose collection of Erlang companies (Trifork, Erlang Solutions, probably more). IIRC they organized an Erlang conference and decided to relaunch it as a general FP conference. It was _really_ good, you should definitely check it out :)