|
Notorious b.s.d. posted:all scripting languages are bad whats a scripting language anyways
|
# ? Jan 10, 2014 23:56 |
|
|
# ? Jun 5, 2024 09:38 |
|
( BOSS enters office) BOSS: I need you to organize this data and output it to a graph CODER: okay but this is a lot of data, this will take a while ( montage of CODER typing at keyboard for 6 minutes, jerking off, sleeping )
|
# ? Jan 10, 2014 23:57 |
|
can build a string without worrying about NUL terminator
|
# ? Jan 10, 2014 23:58 |
|
a scripting language that is java is a scripting language, a bad one
|
# ? Jan 10, 2014 23:59 |
|
MeruFM posted:i've started learning GPGPU with C++ because it seems to be the most supported language. Maybe it's my poo poo-stained glasses but C++ is pretty lovely between the large amount of C style functions, macros, and memory management of course made worse now that there's 2 discrete memory spaces to manage, one of which is basically a black box if you've got the budget and time, the next iteration of CUDA will have unified memory if not, look to stuff like the thrust library. if you're lucky that'll be enough to hide the worst bits
|
# ? Jan 11, 2014 00:03 |
|
coffeetable posted:if you've got the budget and time, the next iteration of CUDA will have unified memory Cool It's kinda sad and cynical of me, that after looking at the example slide, I can only think that everyone's just going to be: "Oh, memory problems? Add some more 'cudaSynchronize' and see if it works"
|
# ? Jan 11, 2014 00:54 |
|
coffeetable posted:if you've got the budget and time, the next iteration of CUDA will have unified memory how does this work w/ current pc architectures. is it just doing the copying for you? will you essentially get a free speedup once some of the console architecture stuff makes it to pc?
|
# ? Jan 11, 2014 01:15 |
|
Jonny 290 posted:whats a scripting language anyways you know it when you see it, like obscenity. here are some tips on how to identify one, with the languages sharing each trait in parens. i am not 100% on all of these so i may have missed some obvious ones
|
# ? Jan 11, 2014 02:04 |
|
Notorious b.s.d. posted:you know it when you see it, like obscenity. here are some tips on how to identify one, with the languages sharing each trait in parens. i am not 100% on all of these so i may have missed some obvious ones
|
# ? Jan 11, 2014 02:06 |
|
there's nothing "broken" about php's garbage collection. and php has no threading support so you can't really call it "broken" either. the other two things are good
|
# ? Jan 11, 2014 02:23 |
|
Cybernetic Vermin posted:it is telling that python might be the sanest of the p:s and they deprecated half the language for the sake of vision~~ and purity~~~~ python 2 is garbage and I can't wait for it to die it's almost liveable until you have to do anything with unicode, but in the end, you always end up having to deal with unicode then: lol
|
# ? Jan 11, 2014 03:12 |
|
today I found a line of code ending "#DEBUG TODO remove before putting into production". it disabled 90% of the functionality of the method it was in, and had been deployed to production sometime in late september. coding is hard. on the plus side, it made closing out the associated bug very easy. e: this is probably more of a cjs post but I'm putting it in the PL thread just to mix things up a little. we'll see how it goes.
|
# ? Jan 11, 2014 03:14 |
|
PleasingFungus posted:python 2 is garbage and I can't wait for it to die but, but, what about my non-portable serialization?
|
# ? Jan 11, 2014 03:50 |
|
PleasingFungus posted:python 2 is garbage and I can't wait for it to die They didn't remove asyncore from Python 3 and that makes it a bad language. Also, there was a new standard library thing called concurrent.futures that nobody likes and the only reason it got in was because people were too busy arguing over whether it should be called "futures", "deferreds" or "promises". No, seriously. Python 3's unicode support isn't great. It's impossible for me to build an HTTP server that serves bytes from arbitrary files without converting it to UCS-4 internally, which means that serving a 10MB file blows up to 40MB in memory. Also, you have to copy it like twenty times before it gets to bytes on the other side.
|
# ? Jan 11, 2014 05:20 |
|
Mr Dog posted:(i mean boehm gc somehow deals with this problem so i guess it isn't an intractably broken design but this entire class of garbage collectors are lovely hacks to begin with) actually, in my book they're pretty neat hacks. they are the only way to get garbage collection in C. unless, you know, you are managing your own references at which point you're essentially running an interpreter. forgetting about the fact that registers can hold heap pointers is nothing more than a stupid newbie mistake. Notorious b.s.d. posted:design determined by implementation details of naive interpreter, instead of a language spec (perl, php, python, ruby, tcl, vb) i don't know, i think you're either being unjustly negative ("implementation details" (~> overparticularity), "naive" (~> thoughtlessness)) or unjustly inclusive (it's okay and inherent for language design to determine some details of implementation). for instance, i think python and tcl are sensible designs.
|
# ? Jan 11, 2014 14:46 |
|
quote:tcl lol no weak typing is terrible
|
# ? Jan 11, 2014 14:52 |
|
Mr Dog posted:lol no weak typing is terrible incorrect
|
# ? Jan 11, 2014 14:56 |
|
case in point
|
# ? Jan 11, 2014 15:03 |
|
Suspicious Dish posted:It's impossible for me to build an HTTP server that serves bytes from arbitrary files without converting it to UCS-4 internally, which means that serving a 10MB file blows up to 40MB in memory. i'll believe you but i have no idea how you would manage to gently caress it up that badly. like the worst implementation i can think of would only blow it up to UCS-2
|
# ? Jan 11, 2014 15:49 |
|
suffix posted:i'll believe you but i have no idea how you would manage to gently caress it up that badly. like the worst implementation i can think of would only blow it up to UCS-2 wchar is 32-bits on linux
|
# ? Jan 11, 2014 15:59 |
|
suffix posted:i'll believe you but i have no idea how you would manage to gently caress it up that badly. like the worst implementation i can think of would only blow it up to UCS-2 utf8 and ucs4 are basically all anyone should bother with now
|
# ? Jan 11, 2014 16:20 |
|
py3 strings vary in size based on the largest present code point now, mostly to save space on ascii-only strings
|
# ? Jan 11, 2014 16:37 |
|
Aha. That wasn't the case when I looked at it during the Python 3.2 era. That's a clever hack! but I still don't like the way they implemented Unicode support. There's a lot of cases when you want bytes as "text" strings without converting to a sequence of integers.
|
# ? Jan 11, 2014 17:09 |
|
Suspicious Dish posted:Python 3's unicode support isn't great. It's impossible for me to build an HTTP server that serves bytes from arbitrary files without converting it to UCS-4 internally *record scratch sound* say whaaat
|
# ? Jan 11, 2014 18:11 |
|
double riveting posted:i don't know, i think you're either being unjustly negative ("implementation details" (~> overparticularity), "naive" (~> thoughtlessness)) matz implemented ruby in the most straightforward and obvious way, the "naive" choice. mri code is mostly very easy to read (particularly before 1.8). this is good for hacking on the interpreter. this is bad for the language design. double riveting posted:(it's okay and inherent for language design to determine some details of implementation). for instance, i think python and tcl are sensible designs. you either can't read or can't type. many of python's problems come from the implementation defining the design. the arrow should flow the opposite direction obvious things that come to mind for python: * dumbfuck refcount gc (determined by interpreter's C api) * totally unfixable global interpreter lock when google can't fix your locking problem you know you are in trouble Notorious b.s.d. fucked around with this message at 20:11 on Jan 11, 2014 |
# ? Jan 11, 2014 20:05 |
|
double riveting posted:forgetting about the fact that registers can hold heap pointers is nothing more than a stupid newbie mistake. it's a consequence of going with a naive implementation rather than bytecode or anything. ruby objects are C structs and C pointers. this makes garbage collection and compiler optimizations a dangerous combination turns out if you compile mri with -O2, sometimes the compiler makes the right decision for C and the wrong decision for ruby p.s. it is a bad sign that your post is so full of stupid i have to quote it twice Notorious b.s.d. fucked around with this message at 20:10 on Jan 11, 2014 |
# ? Jan 11, 2014 20:07 |
|
python is fine and nobody in the real world actually has problems with the GIL, which is also solved in other implementations to satisfy those spergnerds. python is still the best for exploratory data manipulation/analysis, gluing parts together and hooking things up to the web. if you're writing system software in python then lol. who uses just one language anyway
|
# ? Jan 11, 2014 20:13 |
|
i have successfully pushed most of my non-python work onto newer coworkers than me its rly nice i dont have to do poo poo
|
# ? Jan 11, 2014 20:15 |
|
kraftwerk singles posted:python is fine and nobody in the real world actually has problems with the GIL, which is also solved in other implementations to satisfy those spergnerds. python is still the best for exploratory data manipulation/analysis, gluing parts together and hooking things up to the web. if you're writing system software in python then lol. who uses just one language anyway python's badness may or may not affect its applicability to your problem bad languages can still be useful edit: also, lots of system software is written in python. did i ever tell you about the time that python's broken garbage collection put my ~*~enterprise linux~*~ systems into a situation where they could not patch themselves? Notorious b.s.d. fucked around with this message at 20:22 on Jan 11, 2014 |
# ? Jan 11, 2014 20:20 |
|
also the reason that the GIL doesn't give anyone problems in "the real world" is that every python developer is taught from day one to stay the gently caress away from python threads once burnt, a thousand times shy it's still a really bad and stupid problem to have in a language
|
# ? Jan 11, 2014 20:21 |
|
everything is a poo poo fractal. except maybe C, but its also bad because it takes way too long to use to solve small problems
|
# ? Jan 11, 2014 20:22 |
|
Socracheese posted:everything is a poo poo fractal. except maybe C, but its also bad because it takes way too long to use to solve small problems i disagree. java and .net basically own bones this is because they were written by people who had already learned some hard lessons about language and platform design. on rare occasion, the "second system effect" is a positive good.
|
# ? Jan 11, 2014 20:23 |
|
Notorious b.s.d. posted:also the reason that the GIL doesn't give anyone problems in "the real world" is that every python developer is taught from day one to stay the gently caress away from python threads p much this you just use an eventing lib if you need concurrency
|
# ? Jan 11, 2014 20:23 |
|
is there a way to even write .net not using visual studio
|
# ? Jan 11, 2014 20:26 |
|
Socracheese posted:is there a way to even write .net not using visual studio comedy option: http://monodevelop.com/
|
# ? Jan 11, 2014 20:30 |
|
my liberal obama brainwash college got me doing all my coding on unix and i havent coded on windows since using bloodshed C++ compiler on windows NT or something coding on windows seems p awful imo
|
# ? Jan 11, 2014 20:30 |
|
Socracheese posted:is there a way to even write .net not using visual studio emacs c# support is really crude/bad, dont bother trying it unless you are just in love with cedet. monodevelop wasn't terrible the last time i used it. people tell me the xamarin ide is nice but it aint cheap.
|
# ? Jan 11, 2014 20:31 |
|
Socracheese posted:my liberal obama brainwash college got me doing all my coding on unix and i havent coded on windows since using bloodshed C++ compiler on windows NT or something coding on windows is pretty nice it is everything else that is terrible: putting up with microsoft's sudden and capricious deprecations, negotiating microsoft volume licenses, managing microsoft app servers, the lack of open source, working with coworkers dumb enough to be working in a windows shop etc
|
# ? Jan 11, 2014 20:46 |
|
nah windows programming is still terrible c# loving owns but the standard library is a very mixed bag. in particular asp.net is horrible garbage: it tries to simulate a stateful control hierarchy within a single html <form> by serialising and HMACing the state of that control hierarchy and round-tripping it through the client, which will use some prehistoric magic javascript poo poo to attach an event to the state blob on every POST back which will fire once the server deserialises the control state. web development is not client ui development ffs. otoh this was the state of the art in like 2010 so maybe they've just walled off that part of the standard library (which must nonetheless continue to exist in perpetuity) and created a new framework from scratch since then.
|
# ? Jan 11, 2014 20:55 |
|
|
# ? Jun 5, 2024 09:38 |
|
it's called asp.net mvc and it owns bones
|
# ? Jan 11, 2014 21:02 |