why do you hate this so much /g/?
>pic
>>52639500
I don't, it's great
>>52639500import faggot
print op is a faggot
goto kill you're self
>>52639578
Python rules, it's a good scripting language. Gets the job done without much hassle
>>52639500
people hate 3.0, 2.7 is fine
>>52639578
>not "op is a faggot"
>>52639537
So nearly half of people familiar with type theory don't like neither?
>>52639642
Half the people familiar with type theory aren't programmers, they are mathematicians and logicians.
>>52639537
I have no idea what type theory is. And I only have a faint idea of when to use python, or any other of the languages I am a little familiar with.
>>52639642
Most people familiar with type theory don't give a fuck either way. Those that do care go for static typing. We all understand why static is superior, but the reason why is pretty much negligible given the things most python programs are written for. Nobody is writing an operating system in python or anything equally heavy duty and able to be abused so it doesn't matter.
>>52639619
C++ master race
>>52639710
Fuck off
t. Linus Torvalds
>>52639619
Scripting language? Its Interpreted...
>>52639537
I agree with this. It will blow your mind how how indispensable a low-level understanding of your task is. The incredible performance modern C/C++ can give you using the right libraries correctly is icing on the cake.
[ tab ] [ tab ] [ tab ]
import program that does what i want .py
reblog repackage +1'd subscribed follow me on github
./ad-nauseum
>>52639736
CPython, the most widely used implementation, is compiled.
>>52639500
>slow
>bloated
>meme syntax with mandatory tabbing
>GIL
>the users
>>52639578
>>52639632
>not ("op" + "is" + "a" + "faggot")
>>52639843
Also the whole Python 2 vs Python 3 controversy
Literally the only thing good about Python is the ability to slap together a quick script
>>52639843
Python as a language does not require GIL.
>>52639806
I'm familiar with Jython, that uses a JAVA interpreter, and if I recall, CPython is a source code interpreter... How ever python CAN be compiled to machine code, it is often not.
>>52639500
no hate
its amazing
>>52639868
>>52639632
>>52639578
I have never seen anybody so assblasted in my entire damn life.
/g/ is filled with skiddes who adhere to one language
python is easy to use and has a lot of support/libraries
packages like numpy, scipi, and matplotlib are useful to actual professionals
but lemme see your elegant haskell one liner :^)
>>52639642
They like dependent types.
>>52639891
Great. Know any complete implementations that don't have a GIL?
Also I'm not convinced that the entire Python standard library is fully reentrant.
>>52640110
Jython
>>52640128
>Jython 2.7 implements the same language as CPython 2.7, and nearly all of the Core Python standard library modules.
>nearly all
>2.7
Know any *complete* implementations that don't have a GIL?
eh, I'd rather just use lua desu
Horses for courses. Python is good for code where the programmer's time is more important than raw speed, when you know the logic will change a lot over time, and when the code won't grow into a million line monstrosity. In other words, clarity and productivity over theoretical purity. It's great as long as you don't pretend it's something it's not.
var op = "Faggot"
Now you scared me. I wanted to start to learn Python...
>>52640314
Don't listen to butthurt C++ faggots.
Python is a great language to learn, and is the best scripting language out there.
For large projects, I still prefer C; for everything else, Python.
>>52640283char OP[] = "faggot"
>>52639623
and that's part of the fucking problem holy shit these fucking faggots
you can't use 3.* on 2.7 code
HOW UNGODLY RETARDED DO YOU HAVE TO BE TO RELEASE SUCH A SMOKING HOT PILE OF SHIT?
>>52640360
Thank you. I even set up Emacs for Python work and did a little research before decision as what new language to start to learn. I am myself C but never understood all the hate towards languages.
>>52640376
>first character of variable is a capital letter
BUT THAT'S WRONG
WROOONG
>>52639897
>How ever python CAN be compiled to machine code
[citation needed]
how tho?
>>52640459
Wait what.
http://www.c4learn.com/c-programming/c-variable-nameing-rules/
>Valid Names
>Num1
>>52640492
same way as lisps are compiled. if there's anything that might need interpretation at run time, the interpreter is compiled in with the rest of things.
Name a good alternative.
I love it though.
>>52639500
The people who suggest C/C++/Java as alternatives to python are morons who dont understand what people use python for in the fist place.
Most people just use it to throw together quick scripts or small programs that do useful things without having to do too much work because there are some awesome libraries out there for python. No one is going to write a goddamn OS or virtualization software in python.
>>52639500
It makes me feel dumb when I don't know how to do anything in it even though I have no problem working in the C family of languages (C-like syntax)
perl is a better scripting language
>>52640492
>>52640740
Second google result: http://stackoverflow.com/questions/138521/is-it-feasible-to-compile-python-to-machine-code
>>52640440
emacs as a python IDE is so fucking good. Make sure to use the python REPL feature emacs for godly RDD
Is java better than python?
>>52641704
python is better at some things
java is better at some things
right tool for the right job
>>52641728
Then what's the best thing for starting off? Not sure what I want to do right now.
>>52639500
python is good for beginners. also not bad for something quick and easy.
>>52639500
I don't. I just started learning python... It's pretty cool. Also know Java and C++.
>>52641728
This
>>52641897
C or C++ for practicing basic memory manipulation.
For example, making linked lists, B trees, sorting algorithms, etc.
Python, is also good for practicing algorithms, but not so much for memory manipulation.
It's also good for backend web development.
e.g. Django, Flask.
Java's good for Android Apps.
Javascript's also good for web development.
Look up meteorjs, Angularjs, jQuery.
It really depends on what you want to do.
Try to answer the basic question, why do you want to program?
Think about what specific things you like the most about programming and do a project that capitalizes on that.
>>52639632
>notfaggot="op"
print("{faggot} is a faggot".format(**locals()))
>>52641350
you can spend 3 hours or less to learn to use it.
http://stephensugden.com/crash_into_python/
>>52641897
>Not sure what I want to do right now.
Then make sure what do you want to do right now. Then choose appropriate language/framework/libs/etc.
>>52642130
underrated post.
finally some anon understood programming for real work.
>>52639500
hardlyreadablefunctionnames
Mainly because od that.
>>52642130
Thank you m8
>>52642130
>Why do you want to program.
to make money.
>>52642288
how_this_is_hard_readable?
>>52642343
Then study economics first.
>>52642374
ahead of you on that one. Took macro last semester.
I learned nothing tbqhonest family
>>52642288
Why doesn't any language just ignore underscores except for before and after all alphabetical characters and possibly ignore capitals except for types and constants, then give a warning for the programmer breaking their own style just in case?
>>52639500
/g/ hates it because they think they're edgy for rejecting a language that's so easy and high level, as if doing everything in C makes them more hardcore.
Granted, there are times when using Python doesn't make sense at all, but there are times when it is obviously among the best choices. If you're writing simple scripts, bots, crawlers, web scrapers, or anything that's meant to automate some trivial task, Python is the way to go.
>>52642465
sounds gross
you should design and implement a language around it
>>52639729
Filthy spergtastic hobbyist
Hate because python coders have jobs, which most of /g has never experienced. Which is pretty much the same reason /g hates on 3dpg.
>>52642527
I mean, it is disgusting, but is there a better way to not force programmers into a certain style or force them to mix styles?
You could always just implement it as an option for libraries.
>>52639775
So type theory is just understanding how your type is represented as actual ones and zeroes? I guess I can say I know type theory. I didn't know it had a name.
No multi-statement lambdas.
>>52640110
>Also I'm not convinced that the entire Python standard library is fully reentrant.
Your mother is fully reentrant.
>>52642627
>lambda
>statements
l2functional
>>52642624
No, it's abstracting with types.
>>52642647
>>52642343
>>52642459
As someone who was concerned about salary to the point that it was omg penis size when he was younger...
Do something that is hard and that you enjoy.
Something hard will make money, if you follow whatever is hot in tech right now you will be fucked because you will be behind the bell curve and struggling vs everyone who is hoping into it for cash sake. By the time you are good at whatever is "hot" everyone else will be too and you'll be back to slightly above average salary.
If you really want money over everything else look into the sales, managing and business ownership side.
If you want just money and have this burning desire to get to the upper rungs of society so that your progeny wont have to work... Ownership is only option.
Go look at http://wallstreetplayboys.com/ and learn how to stab people in the back and focus on getting in good with the tops of large business.
Oh yeah bonus points, people skills mean more than anything. You want the wallstreet brokers to feel comfortable snorting coke with you.
>>52639500
I wish it had even optionally declared types so I wouldn't have to remember library objects or large APIs in order to script them.
Also, learning python will get you to Google's secret page so it can't be all bad.
>>52642703
Expressions, statements, whatever. That stinginess really is /g/.
>>52641123
Angelscript
>>52642728
Like building arrays,queues, dequeues, vectors, trees, and heaps with pointers? I took an abstract data types class where we did that. Good fun but never came across anything called type theory.
>>52642348
issubclass(), bytearray(), raw_input() are python's built-in functions. The names of the first two functions are fucking disguising and unreadable for me. The third name is ok but the naming style differs and that inconsistency sucks. So why using a language with such anti-patterns?
>>52643015
any competent pythonista follows PEP8 which advocatessnake_case_you_fucktard
>>52640182
PyPy is the most complete, and it can interoperate with CPython with multiprocessing
>>52643015
raw_input was removed from python3
>>52639500
Because the reference implementation is crap and you can't consider the alternatives as "working".
>>52642130
*golfclap
Someone clearly over the age of 25 and employed. Glad to know there are some on /g/ whom are not total dragon dildo's
>>52643138
Ok. How is it crap? Fine to have that opinion, but back it up.
For example, I don't like c++ because of the constant abi changes in many implementations, and I think it lends itself to overly complex solutions.
So what is crap, in your opinion, about python?
>>52642795
gotcha, I'll learn haskell!!!
>>52639666
So actual smart people
>>52639897
Where can I go/what can I search to learn more about this? Definite blind spot for me
>>52640420
For anyone following along at home, and has ever wondered why it took Sun so long between Java releases, and now why it takes Oracle so long, this is why:
The ability to release over decades without breaking compatibility takes a shit ton of time and effort.
Python is remarkable because they just kind of went "eh, fuck this, let's just do what we want and push it", making it a clusterfuck for backwards compatibility.
>>52641897
Java.
You can do anything in Java, from desktop to embedded to web / server back end; and the tooling support is unmatched by any other language.
Python is definitely easier to learn, but you can also learn some bad habits.
Java gives you a far better starting point for everyone except the complete layman, in which case Python may be quicker to grasp and pick up from a zero knowledge start point.
>>52644996
java is good but it's not really OK to use java because it's so ghey
any thoughts on that, javafriend?
>>52639500
I'm going to a 6 month coding bootcamp to learn how to do data science and they're teaching it in python. They've guaranteed me 100k+ starting, this is a good idea right?
>>52642459
You took macro before micro?
Stupid.
You learned nothing in a whole semester?
Stupid.
Give up already, you're obviously never going to make significant money. Learn a trade, maybe?
If I just want to start being some fag that builds random things like on Instructables, is learning linux and python ok?
>>52645317
Yes.
>>52639537
>People who have extreme short-term memory loss
Stupid picture is stupid.
>Use dynamically typed language
>Never reassign variables
Wow look, it's exactly the same as a statically typed language!
>>52642627
Multi-statement lambdas are for plebs.
Check this out, pleb: https://gist.github.com/darkfeline/d29dc4e431f71617420a
>>52645967#!/usr/bin/env python3
import sys
(lambda input:(lambda ilen:(lambda m,olen:(lambda star:(lambda f,g:None)(list(map(lambda i:(lambda b:list(map(lambda x,y: star[x].__setitem__(y,input[i]),[i,b,m,m,i,i,b,b],[m,m,i,b,i,b,i,b])))(olen-i-1),range(ilen))),print("\n".join("".join(char for char in row) for row in star))))([[" " for x in range(olen)] for x in range(olen)]))(ilen-1,2*ilen-1))(len(input)))(sys.argv[1])
>>52639500
because I want to write useful and efficient applications, not draw pretty pictures with a turtle
>>52643124
>multiprocessing
get the FUCK out with this retarded bullshit
- pythonfags
- package naming like: pickle
- function naming like: dir
- retarded scoping
- missing basic control statements like for(not for each) and switch
- fake object system - its not a object it a just a tuple
- self
- class definitions aren't definitions they are functions that create special packages that can be completely modified at any time
- function default parameter values are not constant
- no interfaces
- no "object" property access modifiers eg every thing is pubic cus wes adult now
- inconsistent syntaxif cond:
stmtstmt if cond
- can't use assignment in a conditional because: var = expr is a statement or some other hand holding baby tier bullshit
- memory hog
- slow as shit
- pythonfags making excuses for this shit
I'd type up more but I'm on my phone.
>>52639500
There are two reasons, one's good and one's bad
bad reason:
> its for noobs lol
good reason:
> it's extremely fucking slow
>>52641312
No, people should be only using python for those thing but these idiots are using it for a lot more.
>>52646467
sounds like the same sort of problem every language (even C) has
>>52639500
It cares the baggage of c++, but add a terrible syntax to it.
/g/ hates things that actually work but that are also flawed. Like C.
w-wu
W-WHi
WHITEPSP
WHITESPACE
[/autism]
>>52646543
>pythonfags making excuses for this shit
Yeah c really has a problem with its object system. :^)
Pythonfag in full damage control mode.
>>52646638
fuck python, I was referring to idiots using a language far outside where it's meant to be used and where features of the language work against you.
Python is really good for implementing things really quickly and easily, especially for personal purposes where you don't intend to distribute it or for scientific things as a matlab alternative.
It often seems to end in spaghetti if someone tries to write a large and complex program in it though.
>>52639500
It's not purely functional.
Python :: shit
>>52646049
>>multiprocessing
>retarded bullshit
Is it some kind of meme?
>>52646241
>missing basic control statements like for
range/xrange provide enough functionality for iterating over indexes. Everything else supports for in.
>and switch
Have to agree with it.
>fake object system - its not a object it a just a tuple
What do you mean by this? What is "real" object in your opinion?
>every thing is pubic
As something bad. Every time someone make private shit, someone else (want to) use reflection to get access to this shit. Underscore is used to mark anything you shouldn't access in normal circumstances and that's enough.
>pythonfags making excuses for this shit
At this point I am not sure if you actually willing to reason or just rage-posting.
>>52640492
nuitka can compile python!
>>52644855
The fact that they aren't familiar with type theory means they are dumb in the ways that matter when it comes to programming language design. Just because you're smart when it comes to math and logic doesn't mean shit when it comes to designing a FAST language. Python can be great for other reasons, but its slow because of its dynamic typing.
>>52639578
>kill you are self
Nothing wrong with Python, its one of the more useful scripting languages and capable for larger projects, too.
>>52642647
KEK
>>52640091
Every program I've ever written in Haskell has been one line fām
(before reading through the whole thread)
Disclaimer: I have been using it (also Django/Flask, along with JS) to pay rent in the last four years
Language itself is... okay. Can live with it.
The problems start once you try digging a bit deeper.
Two things there: the community and, as a predictable consequence, the decisions they make - both technical and social.
Community is - well, it's competing with Node/JS community for the title of "largest bunch of SJWs this side of Gender Studies department". A strong third entrant arose recently - Rust/Mozilla (good bye Brendan, you had a great run).
SJWs behave just like you'd expect - hysteria, stupidity, drama, harassment, sexism, racism (anti-straight white cis male only, of course) etc etc etc.
Django cofounder non-stop screeching on Twitter about MUH OPPRESHUN? Check.
Him again, proudly proclaiming on the conference "great programmers do not exist, we're all mediocrities"? Check.
Him again, saying right after "I'm not gonna answer any questions from the audience, you and I all know you hate that" (to the applause)? Check.
What do they think about freedom of speech? You guess.
A bunch of other, less visible, but sometimes even more unhinged assholes.
The quality of technical solution that they provide is, well, you can guess. Works for a small-ish web thingies, turns into morass once you start growing beyond mid-level project.
Saw a lot of complains about other parts of Python ecosystem in general, too.
Work alright as a glue to put bigger parts together.
Seems to be working alright for academia - NumPy, SciPy, etc.
All in all - a decent programming language, as long as you're not a programmer.
>>52639795
breddy mudj
also tweet
also last line of readme/site footer:
"made in SF with <3"
AAARRRRGGGGH where's muh baseball bat
>>52640110
>Know any complete implementations that don't have a GIL?
There isn't any, to my knowledge.
>Also I'm not convinced that the entire Python standard library is fully reentrant.
Closest approach is PyPy, but AFAIU it doesn't work with parts done in C, which is a significant part of standard lib, yes.
>>52639883
>>52640214
Yep, tons of modules/libs is a good thing.
Good for web and/or quick prototyping, but once you want to BUILD something that'd stand under pressure, you'd better look elsewhere.
>>52640091
> packages like numpy, scipi, and matplotlib are useful to actual professionals
...which AFAIK are a Python wrapper over C code.
But, yeah, these things work, Python is widely used in academia.
Like I said
>>52647874
"a decent programming language, as long as you're not a programmer"
>>52640314
eh, good for beginners
>>52644965
>making it a clusterfuck for backwards compatibility
...with the only major improvement being some slight adjustments to how Unicode is handled.
GIL problems are still there, etc.
>>52647222
>its slow because of its dynamic typing.
it's slow because it's bad
you could write a C# program making heavy use of dynamic and get way better performance than python
my personal peeve with python is the lack of lexical scoping
>>52642920
kek
>>52647764
Python is not a scripting language.
Why is Python so much more popular than Ruby? I write a lot of Ruby scripts at work, but Python is pretty much our standard and I'm the only one writing any Ruby. I just happen to work on Puppet mostly which requires Ruby, so i mostly stick to Ruby scripts outside of Puppet as well.
>>52644996
Yes, with your plastic Mr Beer barrel kit, you can brew anything from IPA to Lager, Stouts, meads and wines! Doesn't mean it's the best, or even right, tool for the job. There are other & better ways to home brew, and Mr Beer out those other home brew methods really can't compare to the professional brewing at the artisanal or industrial levels.
Just because you can, doesn't mean you should.
>>52649736
Lol mr beer. My friend had one of these 20 years ago, the beer tasted like corn. It was terrible. We were 18 so we drank it anyway.
>>52649727
I don't know, but any Ruby tool I've ever used had dystopic startup times.
Vagrant 5-10 seconds
Logstash 30 seconds, I'm not even exaggerating
>>52639500
Tbqh no one hates python as much, because there are more mainstream hateable langs like JS, Ruby on fails, Java and Go :)
>>52649727
Python is more universally used, yeah - webdev, some sysadmin stuff, quite some scientific stuff.
Not sure why. For academy-related stuff - van Rossum worked in the University when he was just starting on Python, so I guess he had some connections and was spreading the word.
>>52649727
A combination of luck, and the fact ruby got a lot of negative attention with all the RoR tryhards shitting up every non-ruby forum about how great they are, and shitting up the ruby forums with awful code and advice. They really pushed the idea that Rails==Ruby. They were a fucking cancer, and I'm glad they moved to Node.
>>52649736
I started hgomebrewing on one of those to get a feel for it, your sarcasm actually reveals a good point
>>52644996
is right
>>52644996
>Java
>Ever
>>52649780
>Go and Java
Both are well-designed languages even though Java is a bit dated and has legacy cruft. Python is insanely practical for small projects and getting shit done.
None of those three languages really deserve to be hated.
>>52649761
>>>52649727
>Logstash 30 seconds, I'm not even exaggerating
Yeah, but logstash has the excuse that it runs in jruby. It had to spin up a whole JVM first, before even loading the ruby portion.
It was designed at a time when MIR had scaling issues (<1.9), and i think takes advantage of the fact that you can compile some ruby code to native java bytecode. Also, java is already in corporate environments, so it's easy for upper ups to pretend to understand, and approve it.
>>52642920
Not even close.
>>52649874
>lang for tards
>good
literally a quote from rob pike
>The key point here is our programmers are Googlers, they’re not researchers.
>They’re typically, fairly young, fresh out of school, probably learned Java, maybe learned C or C++, probably learned Python.
>They’re not capable of understanding a brilliant language but we want to use them to build good software.
>So, the language that we give them has to be easy for them to understand and easy to adopt.
>>52639500
Because they honestly don't have a clue.
I'm not even a big Python fan.
I just know what's up.
>>52647943
it fucking isn't you dunning kruger fucking tard
100% of programmers need to know about types and python hides them as much as possible. that's just one of the many things that are literally absolutely ridiculously fucking disgusting as shit disgusting about python
>>52650036
Good point, senpai. I understood the error of my ways and will commit sudoku immediately.
>>52639500
I despise that the logo looks like the flag of Sweden.
>>52650184
very appropriate actually, given their SJW tendencies
>>52639775
Type theory has nothing to do with low-level programming, shithead.
I DON'T HATE IT BECAUSE I'M A DATA SCIENTIST AND PYTHON IS THE BEST LANGUAGE FOR THAT
>>52650325
>DATA SCIENTIST
That's a funny way of spelling "statistician"
It's called programmer, not "code ninja"
It's called analyst/statistician not "data scientist"
It's called salesperson, not "key account manager"
>>52650245
>their
Who's? Guido's?
>>52650308
So the amount of memory used by the computer to represent a particular value has nothing to do with the low level representation of that value? I need to get away from here for a while...
>>52650036
You've never actually used Python haven't you? You can use native c types you dungies
>>52650346
No need to add the unsubscribe button to your newsletter dude, I'm onboard, regardless.
>>52650367
>So the amount of memory used by the computer to represent a particular value has nothing to do with the low level representation of that value?
The amount of memory used to hold a value has nothing to do with type inference or what typing system is used, you fucking shit.
If you thinkvar a = 3;
somehow uses more memory and CPU cycles thanint a = 3;
you are a fucking idiot and should jump off a tall building.
>Hint: It's handled by the fucking compiler, assembly instructions have no concept of "types"; they only move stuff to and from memory into registers
>>52650346
I DON'T CARE IF SOME GUY ON /G/ THINKS MY JOB TITLE IS MADE UP, NOR DO ANY OTHER DATA SCIENTISTS
>>52650364
Yup. The whole Python/Django community (that latter even more actually) - van Rossum, Kaplan-Moss, Gaynor, etc etc.
>>52650432
>Django Team = Python Team
You do know that Python is a lot more than just a web framework, right?
>>52650409
>python
>compiler
>>52650409
And how will the compiler know how much memory to allocate, on the instantiation of "var a = 3"... What happens when the variable mutates? "a = []"?
>>52639500
There are two real problems I have with Python:
- Global Interpreter Lock
- No clean ways to enforce type safety
Dynamic types are fine by me, but why is it so hard for them to implement an interpreter that understands something like:def func(int x, y):
pass
>>52650464
We're talking about type theory, get it? People like >>52639537 don't like Python because it's dynamically typed.
This stupid fucker >>52639775 implied that dynamic typing is what makes Python slow. By that logic, any statically typed language should be fast, no matter if it's interpreted or compiled or ran in a VM.
But we both know that's wrong... As I said, type theory is food for thought for computer scientists and the abstract concepts of a high-level programming language.
With regards to runtime, all that matters is how it is implemented and how close to the metal it is. Dynamically or statically typed, interpreted languages ARE slower than compiled ones. Interpreted Python is slower than C. Interpreted BASIC is slower than C.
>>52650454
Yes I do.
Read through the list of candidates for the last elections for PSF BoD:
https://wiki.python.org/moin/PythonSoftwareFoundation/BoardCandidates2015
Look up DongleGate (>>52649640), PronounGate, check out Alex Gaynor's blog/twitter, etc etc etc
It's not about tech anymore, it's all feminism and diversity.
I've been somewhat closely following Python community for the last few years, it's shit through and through. Unfortunately.
>>52650409
except python has abritrary integer range so int a is different of var a for big values and in fact cost way more cycles as you have to read carry values over more memory, faggot.
>>52650478
>And how will the compiler know how much memory to allocate, on the instantiation of "var a = 3"...
It infers that 3 is a integer literal. This is literally what type inference is.
>What happens when the variable mutates? "a = []"?
It depends on the language and the implementation. However, an empty array is probably the easiest to represent as you need no information about each elements size. For languages that have only objects (and thus passes object references around), [] will be an size indicator only, that contains 0.
[ <some obj>, <some other obj> ] will be the size indicator (2) + two pointer-sized elements.
If you mix "native" types and object types, then you implement it differently. However, in these languages just "a = []" are rarely allowed because you have no type information. This is why you can't do "auto p* = {};" in C++ for example.
>>52650536
Becaus I know it's you...
>>52650409
Also, depending on the language, yes, 'int' would be allocated less memory than 'var' (where you've assumed 'var' to equal 3). So languages will very likely cast the var as a float.
Memory allocation is effective the foundation of low level programming, and being close to the 'metal', or whatever the fuck you're confused about.
>>52650536
dynamic typing is one of the things that make python slow
>>52650560
>DongleGate
A single cunt attending PyCon hardly qualifies as the "python community" anon. Not to mention that she was fired from her job too because of it...
>>52650567
You're still missing it, because in the case of Python "a" will be the size of a pointer. However, what "a" refers to will be a larger size.
See >>52650591
>>52650567
this
>>52647142
>>pythonfags making excuses for this shit
>At this point I am not sure if you actually willing to reason or just rage-posting.
At this point you're trying to rationalize away deficiencies of a imperfect language to support your emotional state. Nothing is perfect, and this includes python which is far from perfect. You don't need to create workarounds and excuses for something that's not deficient.
You're the reason I included that line. Python apologists are arguably the worst thing about python.
>>52647931
>Good for web and/or quick prototyping, but once you want to BUILD something that'd stand under pressure, you'd better look elsewhere.
which languages should we look to then?
>>52650603
>Also, depending on the language, yes, 'int' would be allocated less memory than 'var' (where you've assumed 'var' to equal 3). So languages will very likely cast the var as a float.
They wont// Here is some C++ code
auto a = 3;
What type do you think a is? Do you think it is a float?
>>52650614
Hardly. What makes Python slow is that it is interpreted. The impact of dynamic typing in that case is like a drop of water in the pacific ocean.
>>52650653
>What makes Python slow is that it is interpreted.
compiled python is slow too you tard
>>52650603
Python is not C++. That's where your fuzzy argument breaks down.
Here. Read this.
http://deeplearning.net/software/theano/tutorial/python-memory-management.html
>>52650685
>compiled python is slow too you tard
Do you mean the intermediate bytecode format (aka pyc)? That's hardly "compiled" anon, as it is just a pre-parsed format for the interpreter.
Or do you mean the various Python compilers? First of all, these are highly experimental and there are a lot of valid Python that doesn't work in these. Second of all, these are also mostly unoptimised, at least compared to hyperoptimised C compilers such as GCC, Clang and ICC.
>>52650625
>Pointer
so its 8bytes vs 4bytes, so more
>>52650653
Quoted from http://deeplearning.net/software/theano/tutorial/python-memory-management.html:
"Let us focus on the 64-bit version (mainly because that’s what we need the most often in our case). None takes 16 bytes. int takes 24 bytes, three times as much memory as a C int64_t, despite being some kind of “machine-friendly” integer. Long integers (unbounded precision), used to represent integers larger than 263-1, have a minimum size of 36 bytes. Then it grows linearly in the logarithm of the integer represented.
Python’s floats are implementation-specific but seem to be C doubles. However, they do not eat up only 8 bytes:"
>>52650728
First of all, you need to differentiate between reference and value, anon.
Second of all, see >>52650699
Third of all, a pointer is 4 bytes on a 32-bit architecture. An integer is at least 16-bit but can be 32-bit or 64-bit too. Even 24-bit.
>>52639578
Watch out everyone...
Master of python, minor of English coming through
>>52647874
Completely agree. The major of the python community are extremely annoying. They seem to think that their excrement is made of gumdrops.
>>52649908
What's wrong with a language easy to adopt that allows you to build good software?
>>52650750
>Python’s floats are implementation-specific but seem to be C doubles. However, they do not eat up only 8 bytes:"
I already know this, the point was that this has everything to do with how Python represent objects and little to do with dynamic vs static typing and discussions surrounding type inference.
Object representation != type theory
>>52650625
>A single cunt attending PyCon hardly qualifies as the "python community" anon.
Okay, one last time since I suspect I'm actually talking to a SJW, which is a waste of time.
1) It's not a 'single cunt'. She was supported by the community, I do recall Gaynor at least actively supporting her - and he started PronounGate; plus, he's quite an asshole overall.
2) Still, she's welcome enough in the community to run for the FUCKING BOARD OF DIRECTORS!
If some organizations lets an Aryan Brother run for its board, it tells us something, innit?
3) Did you actually read the link? She's not the only one like that, the list is chock full of this shit, it's feminist this and diversity that, and muh oppreshun and muh soggy knee. Fuck this shit.
4) Then, I did follow others in the community; did not save the links, but SJWism is very widespread.
There was another bitch who _boasted_ not understanding how pointers and dereferencing worked. Got hired by (Red Hat?) after talking on a conference about diversity - hired as a programmer. Left shortly afterwards citing 'lack of support' - which means that they actually got fed up of hand-holding her and (naively) expected some results - any results!
It's corruption, plain and simple. They're driven by the made-up goal of having 50% female programmers, and as the result are pumping shit people in the industry, without any regard for their skill or integrity.
>>52650781
>What's wrong with a language easy to adopt that allows you to build good software?
It makes programming accessible to normies.
I have a single skill, Python takes this away from me.
>>52650810
>She was supported by the community
She most definitively wasn't, she was shunned and mocked and ridiculed and even fired.
How the fuck is that "supported by the community"
>I do recall Gaynor at least actively supporting her - and he started PronounGate; plus, he's quite an asshole overall.
Okay, a single person defended her... So what?
The rest, I'm not even going to bother. Twitterers and single female technology enthusiasts hardly make up the "community" and they most definitively don't represent the team behind the language.
If your broken logic was true, then the Linux community would be full of fucking maoists.
>>52641897
I think Java would be a good choice because it's syntax (sort of like grammar or sentence structure in English) is similar to other popular languages like C# and C++. Python is pretty unique from what I've seen. Java has pretty good library documentation.
C# never gets mentioned, but working with Visual Studio and .Net libraries is pretty nice, if you don't mind being tied to Windows. Microsoft has been open sourcing stuff lately, so the Windows dependency might change.
C++ would give you the most control and help you gain a deeper understanding of things like memory, but it's harder.
>>52650787
I give up.
https://en.wikipedia.org/wiki/Type_safety
>>52650649
You need to talk to people more experienced than myself, but Linux is built with C; DB engines, like Postgres or SQLite are built with C AFAIK; OSes are built with C or C++ mostly.
Erlang is used in telecoms - quick, reliable, fault-resistant.
(Facebook is built with PHP, but don't tell anyone I taught you that.)
>>52650860
Again, type theory != how object representation is implemented.
Dynamic typing means dynamic typing. It doesn't mean "hurr durr use 24 bytes to represent an integer". That's an implementation detail.
>I give up.
You do that, and come back when you've read your SICP.
>>52639537
Reassigning variables is considered harmful, as it completely fucks up the compiler and/or JIT's representation of the AST, making most optimizations impossible.
So in other words that image is bullshit.
>>52650867
>Erlang is used in telecoms - quick, reliable, fault-resistant.
Erlang is very little used in reality.
Telecom firmware is still C and C++. OS support for telecom protocols are mostly written in C and C++ too.
>(Facebook is built with PHP, but don't tell anyone I taught you that.)
It's a truth with modifications. For a long time, Facebook compiled their PHP to C++. More recently they implemented their own super-fast virtual machine in order to execute PHP code very fast, but not all PHP is actually supported (so it's more a Facebook dialect of PHP).
>>52650917
>So in other words that image is bullshit.
No shit
See >>52639666
>>52650846
>She most definitively wasn't, she was shunned and mocked and ridiculed and even fired.
By _programmer community_ at large - yes, mocked and shunned. Python establishment? WELCOME!
>The rest, I'm not even going to bother.
...because you don't have anything to say. Searching for "divers" has 48 hits on that PSF elections link - it's not about tech anymore. PyLadies. DjangoGirls.
Ah, fuck it, it's a waste of time like I said.
> If your broken logic was true, then the Linux community would be full of fucking maoists.
Don't you fucking dare talking to me about logic, you SJW moron.
And, long live Chairman Mao.
>>52650966
>Python establishment?
What "establishment"? Are you deliberately making things up?
>...because you don't have anything to say.
Yes, because you're absolutely deluded.
>Searching for "divers" has 48 hits on that PSF elections link - it's not about tech anymore. PyLadies. DjangoGirls.
Female tech interest groups hardly pose any threats. I'm sure you'll find BlackProgrammers too.
>>52650753
>32 bit pointers in 2016
>32 bit arch in 2016
unless you're using some random RISC architecture or you have a PC from 2003 you're on 64 bit.
Also an int can be as low as 16 bits if it wants, tho its more compiler standard to have 4bytes in size so yeah its 8bytes vs 4bytes.
>References
>Not wasting memory
>>52650939
Thanks for additions. So, yeah, C/C++ mostly.
I heard Haskell and some other FP languages are used in finances, so there's that.
>>52650781
Its not about being easy to adopt is about the GoLang team making decisions to have a weak lang just to catter to retards.
See generic functions for example.
>>52651009
>unless you're using some random RISC architecture or you have a PC from 2003 you're on 64 bit.
>random RISC
You are aware that there are such things as mobile devices?
You realize there are these things called microcontrollers, right?
You are aware that most hardware addresses are still in the 32-bit range (so if you program CUDA with device pointers, it's going to be 32-bit)?
>Also an int can be as low as 16 bits if it wants, tho its more compiler standard to have 4bytes in size so yeah its 8bytes vs 4bytes.
>compiler standard
You realize that this means nothing, right?
Saying that an int is 4 bytes and that a pointer is 8 just shows how ignorant you are. According to the standard integers are at least 16-bit, that's the only guarantee you have.
>>52651065
yeah because you're going to be programming in fucking python your cellphone or microcontrollers, fucktard.
How does it means nothing? gcc, clang and icc compile integers to 32bits so a pointer is bigger than an int. While a pointer is TIED to the arch bit size, faggot.
>>52651016
>I heard Haskell and some other FP languages are used in finances, so there's that.
Yes, but again so is Java. But yeah, generally things close to hardware (OS-level and downwards) are usually written in C or C++, things above that can be written in mostly anything that doesn't have any super-special requirements to latency (in which case they are usually written in C or C++).
Java, for example, is frequently used in high-frequency trading, though often not with HotSpot as the JVM: Zing, for example (4chan thought the link was spam, search for Zing JVM).
>>52650994
>What "establishment"?
(sigh)
Python board of directors. Gaynor. Kaplan-Moss. Van Rossum. Noller. Coghlan. Root. Klabnik, while not with Python, is a close associate of these.
> Female tech interest groups hardly pose any threats.
PronounGate. DongleGate. Graham interview with Tiku, and another flareup more recently. If we go beyond Python - ShirtGate; harassing Tim Hunt; harassing and outing Eich. There was plenty of other scandals (something something elevator coffee atheists Dawkins something something), but I stopped following at some point, discarding feminists/SJWs/Python&Rust&Node.js communities altogether.
Similarly, I'm going to discard you. This is my last answer to you in this thread.
>>52650994
SaltyJewishWoman, shalom mein nigga. The triggered, it cums.
>>52651112
>yeah because you're going to be programming in fucking python your cellphone or microcontrollers,
http://navinbhaskar.blogspot.no/2015/03/python-on-intel-galileo-part-1.html
https://embedded.communities.intel.com/thread/7188
http://qpython.com/
https://github.com/kivy/python-for-android
>gcc, clang and icc compile integers to 32bits so a pointer is bigger than an int. While a pointer is TIED to the arch bit size, faggot.
A pointer is tied to the address space, yes. But integers are not guaranteed to be 32-bit. If you pass the right flags to GCC, it can decide that it wants to use 128-bit registers to store your local int variables instead.
>>52651122
Thanks! I mostly heard of C++ and FP stuff for HFT/algo, good to know.
>>52651129
>Gaynor. Kaplan-Moss. Van Rossum. Noller. Coghlan. Root. Klabnik
Except for Gaynor, you haven't pointed out how any of these are "SJWs"
> If we go beyond Python - ShirtGate; harassing Tim Hunt; harassing and outing Eich. There was plenty of other scandals (something something elevator coffee atheists Dawkins something something), but I stopped following at some point, discarding feminists/SJWs/Python&Rust&Node.js communities altogether.
>if we go beyond
Okay, since you're now just mentioning random other stuff, I'm gonna assume that you suffer from a mental illness and your main problem is really with general consensus regarding women all together, not really specific communities. Maybe concider joining ISIS?
>>52651129
If using Python makes you an SJW, then I guess using Openbox makes you a MTF trans person and GNU software a communist feminist.
>>52650961
>people who get shit done
with emphasis on shit
>>52651170
you're beating a dead horse a var is going to waste more memory and more CPU cycles if it passes the range of 64bits deal with it.
>Intel galileo
Just a grain of sand in the sea of microcontrollers who mostly are programmedi n C.
>Python for android
Cool cool, where are the apps tho?
>>52651252
>>52651201
Listen up, argument by elaboration, ya lost nigga.
>>52651310
>a var is going to waste more memory and more CPU cycles if it passes the range of 64bits deal with it.
>Hurr durr what are MMX registers
>>52644639
Let's put it simple.
I can live with that Python wasn't designed to be fast and that CPython reflects that.
What triggers me is that CPython has none of the advantages an interpreter could deliver, like:
* really fast startup times (CPython is way behind most other interpreters, because of the shit-fest that is the initialization of the standard stuff)
* good embedding API (it's shit tier)
* really good debugging experience (pdg is mediocre at best)
* additional security/sandboxing options
Additionally, Python fan/fuccbois claim that the design of the interpreter was to avoid premature optimization, but good architecture isn't premature optimization.
All in all, CPython is a shining example how not to build such things.
TL;DR: Get into building actual good interpreters if you want to know what's wrong with CPython.
>>52639537
plebeian type theorists prefer Java
patrician type theorists prefer Python, Javascript and Lisp
>>52651559
No. Learn to Google, on my first search I found a websockets framework for Apache and fastcgi (as in whatever the fuck you want to write it in)
>>52651465
show me what op lets you add two 128bit integers
protip there isnt one, you cuck
>>52651627
There is
Google SSE intrinsics
>>52650887
That means it's abstracting the low level representation of a scalar value, removing a certain amount of responsibility from the programmer. Seeing as you're just going to keep throwing jargon at me, lets simplify the argument...
When you assign numerical value to a variable in in a dynamic language, it doest know if you plan to store it with 8bits - (allowing a max val of 255), 16bits (max val 65,535), or how ever many fucking bits. Low level programming assumes an understanding of this trade off.
Tbf to you, a strongly typed language implentation doesn't necessarily mean you have access to this level of control, but you usually do. If you are seriously telling me there's no difference between working in this way, and working with a language that always assumes every number will be a double then you are a little stupid. No matter what book you want to throw at me.
What's the difference between ruby and python, and why would you use ruby over python?
>>52652175
show me the fucking op then
>>52652668
>What's the difference between ruby and python,
ruby is even shittier than python
>and why would you use ruby over python?
to use some ruby on rails gay web dev fag shit
>>52652668
>What's the difference between ruby and python, and why would you use ruby over python?
Basically the only difference is that Ruby has working alternative implementations for different needs.
>>52651252
>GNU software makes you a communist feminist.
This is accurate
>>52639578
Is 'your vs you're' a meme, a bait or everyone on the internet is fucking retarded?
>>52652852
it's a meme. People on the internet are retarded for using memes.
>>52652852
it's a meme you dip
>>52639500
Every programming language has its use.
But hey, what do I know.
>>52653027
>reddit is a gay site for fags.
>>52639500
because /b hates everything popuplar. If tomorrow a greater mass of ppl would rush to buy a ThinkCrap, then /g would throw their ChinkPads into a river.
/thread
>>52652852
Please stop posting, and lurk for at least a year. Thank you.
>>52652868
>>52653027
Just because you've been here for three months, you don't need to jump to spoonfeed even newer newfags.
>>52654130
it's a meme you dip
R or python?
>>52654429
Rhydon
Perl is better anyways
>>52647900
>AAARRRRGGGGH
it's "argc" and "argv" you pleb.
>>52651548
>patrician type theorists prefer javascript
