Re: [julia-users] Re: [Blog Post] A little fun with meta-programming

2014-06-21 Thread Tamas Papp
I am transitioning to Julia from Common Lisp, so I am more accustomed to macros. The use case in the blog post is indeed nice, but these things are routinely done in languages with macros (see, for example, Chapter 7 of Let Over Lambda). The problems you are talking about indeed exist, but they

Re: [julia-users] Re: [Blog Post] A little fun with meta-programming

2014-06-21 Thread Kevin Squire
On Sat, Jun 21, 2014 at 5:55 AM, Tamas Papp tkp...@gmail.com wrote: I guess that the Lisp family influenced the design choices in Julia fundamentally, even if this is not overemphasized in the manual. One of the creators of Julia, Jeff Bezanson, wrote femtolisp, which is used for Julia's

[julia-users] Re: [Blog Post] A little fun with meta-programming

2014-06-20 Thread Job van der Zwan
For someone who has hardly ever used meta-programming outside of the creation of containers, that was really enlightening! Maybe this is old hat to people who work with macros all the time, or compiler writers and other people close to the metal, but this is the first time I've seen the concept

[julia-users] Re: [Blog Post] A little fun with meta-programming

2014-06-20 Thread Tony Fong
Metaprogramming is immensely powerful. I think this use case is brilliant. From the short time that I have been experimenting with it for a bit, I'm a bit torn. I think we should be careful about tradeoff that this technique forces upon us. Starting from a moderately complexity, macros can get

Re: [julia-users] Re: [Blog Post] A little fun with meta-programming

2014-06-20 Thread Stefan Karpinski
This is one of the reasons we opted for very visibly distinct syntax for macros. Otherwise it's hard for any entity looking at the code to know what might be happening – be that entity a person or a program like a parser or a linter. On Jun 20, 2014, at 8:07 PM, Tony Fong