So, I was looking at this thread http://dis.4chan.org/read/prog/1183997706/1-40 and all the various language bashing threads and was wondering... What, in the opinions of anonymous and 4chan in general, would make a perfect programming language? I am interested in both syntax and features and general flames, since they entertain me.
Name:
Anonymous2007-07-14 2:35 ID:VH4WdvW3
>>36
But a dangling pointer is not a violation of the type system, it is not even an error unless you dereference it. Statically typed languages without GC are still considered strongly typed, no?
I was really quite sure that there was a good reason for GC in functional languages above and beyond the usual convenience and safety, but maybe I was mistaken.
Feel free to read the entire thread. It's interesting stuff.
Statically typed languages without GC are still considered strongly typed, no?
I can't think of any non-trivial ones (read: has a heap).
Name:
Anonymous2007-07-14 4:22 ID:kRvNV/Ut
i'm not too sure what would be a perfect language... it'd have to be enterprise-level, with intrinsic attributes toward goal-specified objectives that are unbound by dictations of paradigm. It would be dynamic in a working environment, and allow for opportunities to unfold under a workplace basis.
Name:
Anonymous2007-07-14 4:27 ID:Gv9jVVIu
One thing that I read when working with Java was that it was still possible to create a memory leak if you had a circular reference since the GC would look at the one instance, see the other instance still had a reference to it and not deallocate it, and same when it looked at the other instance, even though no active objects had references to either object. Is this a general concern among garbage collection or just another example of Java being braindead?
>>32
Just spent the last while looking at Factor and must agree that Factor is indeed a very nice feature rich language. Definately intend on learning and using it.
Name:
Anonymous2007-07-15 9:22 ID:tZJSlFBr
Brevity.
Programs should be as short as possible.
Every keystroke is a possible bug.
Nevermind readability. Absolute brevity is more important.
Ruby is a stepping stone for many toward brevity. Lisp and Haskell and terse Perl are too much of a leap all at once.
Name:
Anonymous2007-07-15 9:25 ID:ZqwSpMPN
Yet most single-keystroke bugs are caught by a real compiler, duh. Bettar focus on your actual use of the language.
Oh, sorry, I forgot -- ruby DOESN'T CATCH UNDEFINED IDENTIFIERS. Explains your attitude, yeah. "slef" and all that.
Name:
Anonymous2007-07-15 9:29 ID:SE82Lakt
>>40, wikipedia
The Backus–Naur form (also known as BNF, the Backus–Naur formalism, Backus normal form, or Panini–Backus form) is a metasyntax used to express context-free grammars: that is, a formal way to describe formal languages.
Name:
Anonymous2007-07-15 11:09 ID:22OGH1NK
It should have clear and compact grammar so I could easily write code generators. Also, lisp-like metaprogramming.
>>53
APL greek letters + closures and high order functions.
Arousing.
I wish I knew how to write compilers.
Scheme non-optimized interpreter is supposed to be easy.
Yes. Yessss.
>>49
Brevity is a must, however, IMHO, so is readability, because if its hard to understand, its just as easy to make mistakes (in fact, I'd argue easier and harder to debug because they're unlikely to be syntax errors) as a language that requires minimal characters but is less readable.