S410@kbin.social
on 06 Dec 2023 21:36
nextcollapse
"Our goal is knowledge, so we're going to obfuscate everything to fuck and make things unreadable"
Cannacheques@slrpnk.net
on 07 Dec 2023 11:23
collapse
No. Details and scope limitations please
monotremata@kbin.social
on 06 Dec 2023 22:16
nextcollapse
It looks to me like they did it this way so that it could have natural-language names in many languages. So, the function Z10096 is called "is palindrome" in English, but if you're coding in Japanese you can call it "回文の判定". I don't think the idea is for people to refer primarily to the alphanumeric soup version; I think that's just the unique identifier for the database.
It does look like it's leading to some issues, though. E.g., someone added a test for the "is palindrome" function which uses a somewhat common example: "Straw? No, too stupid. I put soot on warts." Now, a human would probably say that this is a palindrome, because it's got the same letters forwards and backwards, but most of the implementations disagree, because they consider the spaces, capitalization, and punctuation to be part of the string; that is, they test whether the input string and its reverse are equal. So someone (possibly the same person) has added a second python implementation which ignores spaces, capitalization, and punctuation, and mentions that in its name on the page.
Fundamentally this function is solving a different problem than the others (as demonstrated by the differing results on the relevant test), so should it get its own number and page? should there be a "palindrome disambiguation" page? This seems like something the site will have to figure out how to handle.
WetBeardHairs@lemmy.ml
on 07 Dec 2023 00:42
collapse
I think you’re absolutely correct about them choosing those awful identifiers so the functions are not language specific. It just hurts to read and thoroughly makes it harder to understand because my brain doesn’t tokenize “Z10096” and “Z10096K1”.
Tbf it wasn’t so much the names that tripped me as the weird Python operator. I would’ve never guessed that’s a string reversal if you hadn’t told me (I don’t know much Python beyond recognizing the syntax). If I had to guess I would’ve said it’s an array pop.
WetBeardHairs@lemmy.ml
on 07 Dec 2023 01:30
collapse
When I originally typed it, I made a function for string reversal and called that. But I didn’t include it since I didn’t want to define that too.
Honestly… this wiki has a seriously difficult path ahead of it. I mean - it’d be fantastic if it did simplify things like that to let you write simple, elegant, and easy to read functions while linking to other functions.
But it’d also have to lint those and make sure that contributors don’t implement recursive dependencies.
I hope it’s going to be used instead of machine learning. Seems much more correct, secure and efficient to me.
dukk@programming.dev
on 08 Dec 2023 20:20
collapse
This is fucking cool. I can imagine the many times this could’ve helped me quite a bit, and honestly even if I didn’t find the function I needed I could still probably hack out a decent implementation in whatever language and actually contribute towards this. In 5-10 years, this could be really useful.
threaded - newest
Is commenting for engagement boost a thing?
This is incredible and I never could have imagined it. This will democratize the power of programming
Comments impact some of the lemmy sorting algorithms such as active.
Yeah active is currently the default sort in programming.dev until scaled sort comes out so commenting helps the most out of everything here
Default sort is a profile setting too though so many may have overridden that.
Neat. I don’t like that the implementations have to name the function by some cryptic identifier, though. Real words matter in source code.
Who can tell me what this function is?
def Z10096(Z10096K1): return Z10096K1 == Z10096K1[::-1]How about this?
def isPalimdrone(myString): return myString == myString[::-1]"Our goal is knowledge, so we're going to obfuscate everything to fuck and make things unreadable"
No. Details and scope limitations please
It looks to me like they did it this way so that it could have natural-language names in many languages. So, the function Z10096 is called "is palindrome" in English, but if you're coding in Japanese you can call it "回文の判定". I don't think the idea is for people to refer primarily to the alphanumeric soup version; I think that's just the unique identifier for the database.
It does look like it's leading to some issues, though. E.g., someone added a test for the "is palindrome" function which uses a somewhat common example: "Straw? No, too stupid. I put soot on warts." Now, a human would probably say that this is a palindrome, because it's got the same letters forwards and backwards, but most of the implementations disagree, because they consider the spaces, capitalization, and punctuation to be part of the string; that is, they test whether the input string and its reverse are equal. So someone (possibly the same person) has added a second python implementation which ignores spaces, capitalization, and punctuation, and mentions that in its name on the page.
Fundamentally this function is solving a different problem than the others (as demonstrated by the differing results on the relevant test), so should it get its own number and page? should there be a "palindrome disambiguation" page? This seems like something the site will have to figure out how to handle.
I think you’re absolutely correct about them choosing those awful identifiers so the functions are not language specific. It just hurts to read and thoroughly makes it harder to understand because my brain doesn’t tokenize “Z10096” and “Z10096K1”.
Tbf it wasn’t so much the names that tripped me as the weird Python operator. I would’ve never guessed that’s a string reversal if you hadn’t told me (I don’t know much Python beyond recognizing the syntax). If I had to guess I would’ve said it’s an array pop.
When I originally typed it, I made a function for string reversal and called that. But I didn’t include it since I didn’t want to define that too.
Honestly… this wiki has a seriously difficult path ahead of it. I mean - it’d be fantastic if it did simplify things like that to let you write simple, elegant, and easy to read functions while linking to other functions.
But it’d also have to lint those and make sure that contributors don’t implement recursive dependencies.
I thought they were struggling for money?
They are not struggling for money. In fact they have so much money they don’t know what to do with it.
www.makeuseof.com/…/wikipedia-millions-bank-beg/
Either way they did get some additional funding for this project.
It’s to teach AI more properly?
I hope it’s going to be used instead of machine learning. Seems much more correct, secure and efficient to me.
This is fucking cool. I can imagine the many times this could’ve helped me quite a bit, and honestly even if I didn’t find the function I needed I could still probably hack out a decent implementation in whatever language and actually contribute towards this. In 5-10 years, this could be really useful.