News for hackers. Stuff that matters.
Re:Acing the technical interview by nice_byte on Thursday March 16, 2017 @03:15PM (#13888148)
Wow so all the stereotypes about lisp programmers are true after all
Re:Acing the technical interview by nolemurs on Thursday March 16, 2017 @03:32PM (#13888351)
Every few months I think to myself "maybe I should learn Lisp," but then I see things like this, and my suspicion that Lisp has been fetishized beyond all reason starts seeming more likely, and I decide to put that exercise off a while more.

At this point, I suspect I'm never going to choose to learn Lisp.

Re:Acing the technical interview by nice_byte on Thursday March 16, 2017 @04:18PM (#13888789)
I don't think you're missing out on much. Lisp is probably the least interesting of the functional programming languages. I think Haskell has more interesting ideas if your interest is purely from PL research standpoint. For practical purposes, I don't even consider those languages. Their main value for me is the slow cross-pollination and trickling down of ideas (such as lambdas, type inference, algebraic data types, etc) to more practical, mainstream languages.
Re:Acing the technical interview by justinhj on Thursday March 16, 2017 @04:46PM (#13889014)
Lisp languages are not even functional programming languages, they're mostly general purpose. Lisps are interesting for other reasons such as the s-expression syntax being similar to the AST, lists and other data structures as first class entities and so on. Also they are very practical languages. Clojure and Common Lisp are used in real world production systems for example. Missing out on lisp is missing out on a lot of interesting ideas.
Re:Acing the technical interview by draw_down on Thursday March 16, 2017 @03:33PM (#13888364)
I think it's worth learning. But the idea that it's some glorious font of enlightenment? To paraphrase Frank Zappa, pheeeeeeuw.
Re:Acing the technical interview by nolemurs on Thursday March 16, 2017 @03:37PM (#13888407)
One of the things that slows me down is that whenever I see Lisp code, my mind goes to the equivalent Haskell code, and the comparison is basically never favorable.

I'm a firm believer in the value of learning a functional language, and I totally understand that from a historical perspective Lisp is important, but does it really have much to recommend it over modern functional langauges?

Re:Acing the technical interview by marcosdumay on Thursday March 16, 2017 @04:36PM (#13888933)
Lisp goes for its simplicity and metaprogramming power. I wouldn't recommend anybody to learn functional programming in it.
Re:Acing the technical interview by curuinor on Thursday March 16, 2017 @03:44PM (#13888472)
This is clojure so it seems damned modernish
Re:Acing the technical interview by AnimalMuppet on Thursday March 16, 2017 @04:03PM (#13888650)
You can write functional code in Lisp. But Lisp isn't really a functional language. That is, you have to kind of fight the language to write non-functional code in Haskell. In Lisp, you don't.

So if you learn Lisp, you learn a language that transcends programming style. It's much more universal than Haskell.

Re:Acing the technical interview by platz on Thursday March 16, 2017 @04:24PM (#13888833)
if you learn Lisp, you very quickly learn to lean heavily on macros, which is very much a programming style.
Re:Acing the technical interview by platz on Thursday March 16, 2017 @03:43PM (#13888463)
> my mind goes to the equivalent Haskell code

yup, I was doing this too.

notice also he did give a nod/mention in the post

Re:Acing the technical interview by rev_null on Thursday March 16, 2017 @06:34PM (#13889793)
Yeah, Church-encoding is a lot nicer to use when you can wrap the function in a new type.
Re:Acing the technical interview by ep103 on Thursday March 16, 2017 @04:01PM (#13888630)
Re:Acing the technical interview by Verdex_2 on Thursday March 16, 2017 @04:54PM (#13889075)
Just read On Lisp and find a primer on Scheme macros. Then if you're still interested read Let Over Lambda. Few languages do macros like lisp, so it's worth looking at the macros to understand a new concept. Besides that, though, I don't think lisp has anything world shaking.

Also if you happen to like what you see in macros, try also checking out parsing words from forth or factor.

Re:Acing the technical interview by Apocryphon on Thursday March 16, 2017 @05:25PM (#13889353)
The original SICP is in Scheme. Surely Scheme is worth leaning for SICP?
Re:Acing the technical interview by flukus on Thursday March 16, 2017 @09:57PM (#13890756)
The LISP guys spent 40 (50 now?) years telling the world how good lisp is, the ruby and python ones went and actually build cool shit with the language. The later is a better strategy.
Re:Acing the technical interview by rurban on Friday March 17, 2017 @03:06AM (#13892128)
On the other hand the one lisp guy produced cool working programs you are not able to understand, while the ruby and python guys solved trivial problems with 100 fragile dependencies, need 10x more time and people to support it. The first is the better strategy.
Re:Acing the technical interview by nolemurs on Saturday March 18, 2017 @01:59PM (#13902582)
> On the other hand the one lisp guy produced cool working programs you are not able to understand,

It's this precise attitude that turns me off Lisp. First, it's not actually that hard to understand. Sure, it's harder than most other languages, but if you've done programming in half a dozen languages (some functional, etc.) Lisp holds no real mysteries.

More importantly, being hard to understand is a bad thing. A large part of what makes a good programmer is the ability to write clear, robust, easy to understand code. Yet Lisp advocates seem to love to talk up how hard the language is to learn and how subtle and mysterious it is. I'm sure that appeals to teenage boys, but it is objectively a bad thing! Like, literally - hard to understand code has all sorts of obvious disadvantages, and no compensating advantages over easy to understand code. It's a bad thing!

Re:Acing the technical interview by kazinator on Saturday March 18, 2017 @05:20PM (#13903574)
I've never heard any Lisp advocate talk about it being hard to learn or use (let alone in some bragging terms). The most memorable advocacy quote for me is "cuts through hard problems like a hot knife through butter".

When the language is easy to work, you delve into harder problems.

Re:Acing the technical interview by flukus on Friday March 17, 2017 @03:22AM (#13892195)
What cool working programs (emacs aside) would these be? For such an old and constantly promoted language the list of working programs is incredibly sparse.
Re:Acing the technical interview by rurban on Friday March 17, 2017 @08:11AM (#13893335)
lisp programs are used in the industry mostly. everything which is bit more complicated, and is happy with a one-man dev team. extremely cool apps. open source I only know gmane and sawfish. emacs and autolisp certainly not, these are atrocious dialects.

Famous are schedulers or AI apps, such Orbitz and DART. http://www.paulgraham.com/carl.html

https://stackoverflow.com/questions/543579/what-is-the-most-... and http://franz.com/success/ have lists.

Re:Acing the technical interview by LeonidasXIV on Friday March 17, 2017 @04:48AM (#13892498)
That's already a lot more than the number of cool Ruby and Python programs I use.

Going by that metric, we should all switch to C and C++ as they are obviously the way to get software done.

Re:Acing the technical interview by rqebmm on Thursday March 16, 2017 @04:49PM (#13889035)
My advice would be: Do learn functional programming, don't learn Lisp.