Return Styles: Pseud0ch, Terminal, Valhalla, NES, Geocities, Blue Moon. Entire thread

Scheme and exceptions

Name: Anonymous 2011-05-30 10:29

R7RS will have EXCEPTIONS. Does it mean Scheme is ready for enterprise now?

Name: Anonymous 2011-05-31 7:23

What now, the ``in Lisp'' guy is the finitist antijew troll?

Fuck you /prog/

Name: Anonymous 2011-05-31 7:40

>>41
Wait, you really didn't realize it?

Name: Anonymous 2011-05-31 8:04

>>41
He's also the Sussman.

Name: VIPPER 2011-05-31 8:35

He's also VIPPER.

Name: VIPPER 2011-05-31 8:36

>>44
Everyone is VIPPER, except anonymous.

Name: Anonymous 2011-05-31 8:52

I still think that continuations seems to be an over-glorified version of a goto :(.
But I haven't read SICP so I guess I can't complain, though I'd really appreciate an explanation.

Name: Anonymous 2011-05-31 8:53

>>46
Goto is only a local jump.

Name: Anonymous 2011-05-31 8:58

>>46
If you use them for everything (e.g., for loops), they are not much different than goto. The point is, you never explicitly use continuations, you abstract them away with macros and higher-order procedures to implement your control structures.
Also, they will never be as abused as goto, since they are structured and harder to use.

Any control structure can be expressed in terms of call/cc.

Name: Anonymous 2011-05-31 9:01

>>47
Not BASIC's goto. The two main differences are that call/cc is structured, you can only jump to a continuation you already entered once, goto is text based, you jump to a position in text, and that call/cc stores the whole program's context, goto is just a jump.

Name: Anonymous 2011-05-31 14:54

>>46
I'd really appreciate an explanation.
Look at continuations as a fast setjmp. So fast and consistent, that it could be use as a control structure.

Name: Anonymous 2011-05-31 15:04

>>50
as a fast setjmp.
Ruby's callcc would like a word with you. (implemented as stack copying)

Also, continuations are more like {get,set,swap,make}context, not setjmp/longjmp.

Name: Anonymous 2011-05-31 15:10

>>51
That's because ruby is a fucking slug.

Name: Anonymous 2011-05-31 15:22

>>51
Ruby's callcc would like a word with you. (implemented as stack copying)
Ruby is so slow by itself, that won't notice the slowness of it's continuations.

Name: Anonymous 2011-05-31 15:26

>>51
continuations are more like {get,set,swap,make}context, not setjmp/longjmp.
Dunno. I've seen setjmp implementation of syntatic closure in C/C++: http://users.telenet.be/stes/block98.html

Name: Anonymous 2011-05-31 15:27

>>54
implementation of lexical closures
self fix

Name: Anonymous 2011-05-31 15:51

>>54-55
Aren't setjmps one-shot continuations? The *context procedures do exactly what continuations do: save and replace the program's context.
Also lexical closures != continuations, unless you convert all the code to CPS.

Name: Anonymous 2011-05-31 15:56

>>56
{get,set,swap,make}context are linux-specific.

Name: Anonymous 2011-05-31 16:10

>>57
They are POSIX, but yes. That doesn't make continuations any more similar to setjmp/longjmp, unless you're the ``in Lisp'' guy that again negates reality.

Name: Anonymous 2011-05-31 16:13

>>58
That doesn't make continuations any more similar to setjmp/longjmp
It's as best as you can get with C/C++. Java won't allow even that.

>you're the ``in Lisp'' guy that again negates reality.
You cant blame me! The current state of reality is pure horror, with it's Unixes, Javas and C/C++.

Name: Anonymous 2011-05-31 16:19

>>59
We were not talking about C/C++/Java. We were talking about continuations.

Name: Anonymous 2011-05-31 16:22

>>60
Most people, like >>46, don't know about them because of C/C++/Java best practices.

Name: Anonymous 2011-05-31 16:46

>>61
We were not talking about languages that don't have continuations. We were talking about continuations.

Name: Anonymous 2011-05-31 16:50

>>62
Consider an average college student. Explaining continuations to him would require usage of language he familiar with, which is most likely JAVA.

Name: Anonymous 2011-05-31 16:53

the ``in Lisp'' guy is the finitist antijew troll?

This place has nothing anonymous, you get to know every single of the few users.

>>TILFAG (the ``in Lisp'' finitist antijew guy)

Ok. Your point was interesting, never thought infinite could be a poisonous concept or whatever. Now that we heard it, a spammer is just a fucking spammer. You ought to keep higher standards.

Name: Anonymous 2011-05-31 17:02

>>64
This place has nothing anonymous, you get to know every single of the few users.
We know him because he constantly spams every not-already-spam thread in /prog/.

The other one I know is the Common Lisper which posts code in [m][/m]tags and bolds parens.

Name: Anonymous 2011-05-31 17:09

We know him because he constantly spams every not-already-spam thread in /prog/.
Yes! Yes! Deanonymize and kill him! In the name of Jahweh!

Name: Anonymous 2011-05-31 17:19

The jews are after me.

Name: Anonymous 2011-05-31 21:20

>>66
What's the point of daemonizing him if you're just going to kill him?

Name: Anonymous 2011-05-31 21:26

>>68
So he can disassociate the jew from the terminal.

Name: Anonymous 2011-06-03 21:19

(call-with-current-continuation
 (lambda(bump)
  (fuck '(this thread))))

Name: Anonymous 2011-06-04 1:43

>>70
The fuck? You don't even do anything with the bump argument. Shit code.

Name: Anonymous 2011-06-04 7:58

>>71
It's used implicitly in the fuck procedure :)

Name: Anonymous 2011-06-04 8:00

>>72
that is not how scheme works

Name: Anonymous 2011-06-04 8:16

>>72
But lambda doesn't dynamically bind its arguments in Scheme.
IHBT

Name: Anonymous 2011-06-04 9:01

>>72
* :-)

Name: Anonymous 2011-06-04 9:02

>>74
Huge mistake. Look at Emacs Lisp, it's dynamically typed and is much more successful than Scheme because of this.

Name: Anonymous 2011-06-04 9:06

>>76
Emacs Lisp has lexbind too, now.
(setq lexical-binding t)
(let ((x nil))
  (let ((f (let ((x t))
         (lambda () x))))
    (funcall f)))
; => t
IHBMT

Name: Anonymous 2011-06-04 13:26

MORE LIKE
A MANLY PROGRAM IGNORES ERRORS AND KEEPS ON GOING WITHOUT TELLING YOU ANYTHING IS WRONG
DECEPTION not EXCEPTION
AMIRITE LOLLLLLLLLLLLLLLLLLLLLLLLZzz!!11oNE!!1ONE1!

Name: Anonymous 2011-06-16 5:30

NO EXCEPTIONS

Name: Anonymous 2011-06-28 0:36

call/cc can implement exceptions trivially. I'm 12 and what is this
Newer Posts
Don't change these.
Name: Email:
Entire Thread Thread List
All trademarks and copyrights on this site are owned by their respective parties. All comments and posts are the responsibility of their own posters.
- Tinychan + 2ch Mode -