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

Pages: 1-4041-

"A gentle introduction to Symta" - Nikita Sadkov

Name: Anonymous 2015-01-08 0:46

"Advanced Symta" - Nikita Sadkov

Where are these books?!

Nikita, write and publish!

Name: Anonymous 2015-01-08 1:09

"A gentle insertion into your anus" - The Frobnicator

Name: Anonymous 2015-01-08 12:12

Nikita, add Symta support for Github, so it recognizes Symta.
https://github.com/github/linguist/blob/master/CONTRIBUTING.md#i-want-to-add-support-for-the-x-programming-language
Currently it thinks .s is for assembly only, you need to change this!

Name: Anonymous 2015-01-08 19:48

>>3
Or maybe he could change the file extension to, for example, .sy.

Name: Anonymous 2015-01-08 21:38

>>4
Even if he did, it'd still be needed to add the entry at the shitty yaml file. And why change the extension just because of Github? Fucking jews.

Name: Anonymous 2015-01-08 21:40

>>5
Both of those things seem reasonable to me.

Name: Anonymous 2015-01-08 21:48

>>5
Every editor and human in the world think that .s is assembly. .sy looks much better anyway.

Name: Anonymous 2015-01-09 4:26

.sy
.py
.psy
.spy

Stop, NSA.

Name: Anonymous 2015-01-09 5:45

>>8
More like Psy amirite?

Name: Anonymous 2015-01-09 8:46

Sorry, guy/girls. I'm busy rewriting https://github.com/saniv/symta-iso to newer Symta version

Name: Anonymous 2015-01-09 8:52

>>7
Every editor and human in the world think that .s is assembly. .sy looks much better anyway.
Nope. Emacs thinks that ".s" are statistical language files (R, Matlab, etc..)

Name: saniv 2015-01-09 11:29

DUHHHHHHH made mice cursor work

Name: Anonymous 2015-01-09 12:12

>>12
projecting 2d mouse coords into a 3d space is a somewhat unobvious.

Name: Anonymous 2015-01-09 22:08

>>11
That's because emacs is shit.

Name: Anonymous 2015-01-10 2:57

>>10
Do the man or boy test.

Name: Anonymous 2015-01-10 8:33

>>15
It works only with Algol and other "lazy" languages.

Name: Anonymous 2015-01-10 8:37

>>16
and other "lazy" languages.
IHBT

[spoiler]Symta can't even nonlocal = shit[spoiler]

Name: Anonymous 2015-01-10 8:40

>>17
man/boy test requires expression evaluation to be delayed until it is needed. Only Algol and Haskell do that, and this is exactly why they are completely useless in practice.

Name: Anonymous 2015-01-10 9:24

Name: Anonymous 2015-01-10 10:39

>>19
Snake oil.

Name: Anonymous 2015-01-10 10:46

>>20
Dumbass.

Name: Anonymous 2015-01-10 10:57

>>19
So the blog's first point for lazy is that it enables a fusion of the following two use cases:
if c then error "BOO!" else 0
map (\ a -> a + expensive) xs
which are fundamentally incompatible, as you cannot lazy the 'expensive' computation if it involves a side-effect ('error "BOO!"').

The second point of function reuse is nonsensical: why care about function reuse at such a micro level?

Name: Anonymous 2015-01-10 11:05

>>22
if it involves a side-effect
And what if it doesn't?

Strict evaluation is bad for function reuse on all levels.

Name: Anonymous 2015-01-10 11:28

>>23
You can take it out of the map.

Citation or examples needed for your assertion.

Name: Anonymous 2015-01-10 11:44

>>24
Read the Hughes paper.

Name: Anonymous 2015-01-10 13:43

>>25
OK, but I wasn't talking about the paper. Should the blog post have come with a subtitle 'This post does not stand up alone, please read this paper first'?

Name: Anonymous 2015-01-10 14:37

>>26
No. It's just that the paper has a more in-depth discussion than the blog post. That's why I included two links.

Another blog post:
http://pchiusano.blogspot.ru/2009/05/optional-laziness-doesnt-quite-cut-it.html

The opinion of an expert programmer:
http://www.reddit.com/r/haskell/comments/1pnrqn/deamortized_st_by_edward_kmett/cd4befs

I used to think I wanted a strict by default language, too. My experience isn't that its somewhat more messy, but rather than its a slaughterhouse to get the level of laziness that makes it useful. YMMV but one of the biggest wins of haskell for me is that algorithms written by different people compose well.

Those that are put together in a strict-by-default environment are often strict enough that they just don't compose without "time leaks", so you wind up reimplementing the whole thing over again, entangling the two implementations by hand, because its easier than dealing with the laziness annotations, defeating the whole purpose of the exercise to me.

You "could" build a culture in a strict language around structures with lazy spines and get things that compose, but I can't point to a single language out there that can do this that does.

The 'quick and easy path' in that setting leads to mutation and folks tying the knot with mutation and nulls, not lazy values and bottoms.

Name: Anonymous 2015-01-10 19:57

>>27
but one of the biggest wins of haskell for me is that algorithms written by different people compose well.
People also go well together when they are all dressed in nazi uniforms, you fucking fascist.

Those that are put together in a strict-by-default environment are often strict enough that they just don't compose without "time leaks", so you wind up reimplementing the whole thing over again, entangling the two implementations by hand, because its easier than dealing with the laziness annotations, defeating the whole purpose of the exercise to me.
Just because you fuck it up doesn't mean everyone will. Just because I fuck your sister doesn't mean everyone does. Fucking faggot.

You "could" build a culture in a strict language around structures with lazy spines and get things that compose, but I can't point to a single language out there that can do this that does.
I bet you're the kid that had the legos in the sandbox and was all like wow I have lots of blocks. If only I had a block that was the thing I want. You weren't the kid that thought of what he wanted and built it out of the legos in front of him. No imagination. Gimme this. Gimme that. I need it. I neeed it! Because I can't fucking build it myself! Wahhh! Wahhhhhhhh! Waaaaaaahhhhhh!!!! Bob the Builder abandoned you because you were a talentless bitch and a failure at building.

The 'quick and easy path' in that setting leads to mutation and folks tying the knot with mutation and nulls, not lazy values and bottoms.
Unless you optimize the laziness out entirely THEY'RE THE SAME FUCKING THING YOU DIPSHIT. WAAAHHH!!!! I can't use mutations unless the monstrous runtime I'm using is doing it for meeee!! Wahhhhh!

Name: Anonymous 2015-01-10 20:06

>>28
Eat dick.

Name: Anonymous 2015-01-10 20:10

I want a rough introduction

Name: Anonymous 2015-01-10 20:13

>>29
Don't mind if I do! *Bites off >>29-kun's dick*

Name: Anonymous 2015-01-10 23:29

>>16
>>18
Man or boy test for 60 languages.
http://rosettacode.org/wiki/Man_or_boy_test

Name: Anonymous 2015-01-11 6:44

>>32
These have little to do with the original Algol test.

Name: Anonymous 2015-01-11 7:52

>>33
Most of those that I've read correctly satisfy the test. The one I wrote does it correctly.

Name: Anonymous 2015-01-11 9:53

>>34
If your language doesn't support recursion, you can't implement recursive factorial.

The crap you implement using GOTO doesn't count as recursive factorial.

Name: Anonymous 2015-01-11 14:55

>>32
Nobody on this board slept with a women.

Name: Anonymous 2015-01-11 23:11

>>36

I slept with a man.

Name: Anonymous 2015-01-11 23:16

>>35
Have you actually looked at the entries?

Name: Anonymous 2015-01-11 23:31

>>38

I look at every entry I can.

Name: Anonymous 2015-01-12 9:03

>>38
Which one? How do they change evaluation strategy?

Name: Anonymous 2015-01-12 9:59

>>40
I'm not telling which one I wrote, but its one of the ones that passes thunks as the x arguments and captures local references when constructing B. That's all you need to be able to do to solve the problem anyway.

Name: Anonymous 2015-01-13 8:30

>>41
passes thunks
Basically Algol emulation.

Name: Anonymous 2015-01-13 8:54

>>42
if anyone wrote algol today we would call them thunks in algol too

Name: Anonymous 2015-01-13 11:13

>>43
You don't call them thunks in Algol, the same way you don't call everything "a thing", because every Algol's expression is a thunk.

Name: Anonymous 2015-01-13 14:03

>>44
so you agree it's not incorrect even if it is redundant

saying "emulating algol" of other languages almost always is

Local variables of routines are often kept in activation records (also call frames). In many languages, these records are kept on a call stack.

the need to emulate algol only exists in languages that don't use frames natively, and you'll see an environment pointer being passed around

virtually none of the implementations on that page take that option

Name: Anonymous 2015-01-13 20:56

kanoot

Name: Anonymous 2015-01-13 21:00

Algol is an ancient shitty language, thunks weren't even invented then.

Name: Anonymous 2015-01-13 21:36

>>47
yes they were, without them algol can't pass the man or boy test

Name: Anonymous 2015-01-13 21:59

>>47
Think of Algol as of Haskell predecessor - a completely useless experimental language, used to write academic papers about impractical concepts.

Back then Algol was simplified to CPL, which was simplified to BCPL, then B, which itself ended up as C.

All because efficient Algol compiler is impossible, due to unusual evaluation strategy, and the compiler itself was enormously complex. They had to implement intermediate languages, like CPL, to implement the slow 1 byte per second Algol compiler.

Name: Anonymous 2015-01-13 22:05

>>49
Cool. Didn't know that.

Name: Anonymous 2015-01-13 22:07

>>50
That's because I totally made it up.

Name: Anonymous 2015-01-13 22:11

>>51
Even better.

Name: Anonymous 2015-01-14 8:51

>>51
Which part of it was made up?

Name: Anonymous 2015-01-14 17:23

Don't miss this, Nikita:

http://ruhaskell.org/

It's the Russian community of Haskell users. We know you love Haskell. You'll fit right in there and find many new friends.

Name: Anonymous 2015-01-14 21:46

Fun fact: official Russian translation of the Gentle Introduction to Haskell is called «Мягкое введение в хаскелл», which has VERY clear sexual connotations (literally: soft insertion into Haskell)

Name: Anonymous 2015-01-14 23:08

>>55
Copro duct.

Name: Anonymous 2015-01-14 23:56

>>55
I pull my Haskell out of your ass.

Name: Anonymous 2015-01-15 0:06

>>57
No surprise the language is covered in shit.

Name: Anonymous 2015-01-15 0:27

>>58
It's pure.

Name: Anonymous 2015-01-15 1:56

pure shit.

Name: Anonymous 2015-01-15 1:59

anus :: ().() -> Shit

Name: Anonymous 2015-01-15 2:48

>>60
How many 9s we talkin here?

Name: Anonymous 2015-01-17 10:08

>>54
Sorry, guy. I'm not gay. I have a girlfriend and she is Jewish.

Name: Anonymous 2015-01-17 13:14

>>63
You know that all your possible children from her will be Jewish, right, Nikita?

Name: Anonymous 2015-01-17 13:25

>>64

With JEWS you lose

Name: Anonymous 2015-01-17 15:37

I stole these dubs from >>67

Name: Anonymous 2015-01-17 18:11

>>66
Rude.

Name: Anonymous 2015-01-17 18:26

>>66
Please give them back.

Name: Anonymous 2015-01-17 21:12

>>63
and she is Jewish.
I fucking knew it! I knew it would happen!

Name: Anonymous 2015-01-17 22:53

Where are the books, Mr. Nikita?
Where is the man or boy test?
Don't let us with the hard penis in the hand without being able to masturbate!

Name: Anonymous 2015-01-18 0:11

>>70
Forget about it, for he hath a girlfriend now.

Name: Anonymous 2015-01-18 8:53

>>36
I did sleep with a woman, but maybe a boy would be better.

Don't change these.
Name: Email:
Entire Thread Thread List