|
tinselt0wn posted:also in lisp similar languages the program itself is even more data!!!! same with machine code i guess the lesson is, write machine code
|
# ¿ May 5, 2012 16:24 |
|
|
# ¿ May 13, 2024 16:19 |
|
proofing some outsourced software for a side gig. loly moly.
trex eaterofcadrs fucked around with this message at 00:34 on May 6, 2012 |
# ¿ May 6, 2012 00:27 |
|
most of the front end guys i know of are now slaves to bureaucracy and bike shedding. lol when legal starts getting involved.
|
# ¿ May 6, 2012 01:29 |
|
Rufo posted:when you formulate a dataflow analysis problem as an instance of a general dataflow framework (by defining an appropriate lattice + dataflow eq'ns + finding the "least" fixpoint of them + evaluating that at the cfg vertex youre interested in) you typically define abs/rep functions for mapping back and forth between the direct program state and the abstracted state. you need to ensure these conjugate pairs of funcs behave correctly. if you use funcs which form a galois connection, you can later use the fusion theorem to prove they still act correctly when you start dicking about with fixpoints I want. plus i'm totally curious what method do they use to find the fixpoint
|
# ¿ May 6, 2012 01:46 |
|
Rufo posted:its just mfp/mop, iirc the dragon book has something about it + the result about mop being ideal when ???? whatever the condition was. when something is monotonic?? idk don't feel bad, i used to know all sorts of poo poo and now i barely remember my middle name, welcome to getting older.
|
# ¿ May 6, 2012 02:06 |
|
fidel sarcastro posted:this is me pick a couple things, kick the poo poo out of them, move on.
|
# ¿ May 6, 2012 02:27 |
|
fidel sarcastro posted:yep, just need to figure out which things those are going to be aim a bit higher. learn something loving hard... i don't have good examples because i'm lame but just learning a language is easy compared to the magnitude of some of the problems you will encounter if you're anything above a code monkey.
|
# ¿ May 6, 2012 02:44 |
|
Stringent posted:mmm yes. a thorough understanding of a language is usually a terrible place to start. i mean, how are you gonna motivate yourself to reinvent the wheel if you already know it exists? maybe it's just me but the complexity of learning the idiosyncrasies of a language is not only boring but it's simply not hard, it's just tedious.
|
# ¿ May 6, 2012 02:53 |
|
but i guess if that's all you can handle then aim for that i guess. congrats on your mediocrity
|
# ¿ May 6, 2012 02:54 |
|
fidel sarcastro posted:tbf i did say make some things, not sit down and read the camel book cover to cover your comment emphasized language over problem. use whatever language you like man (woman?) just solve the problem well.
|
# ¿ May 6, 2012 02:59 |
|
ppp posted:i wouldnt use perl for learning computer science topics http://perl.plover.com/lambda/tpj.html
|
# ¿ May 6, 2012 03:04 |
|
Stringent posted:the best way to do it is to learn the tools then go find a problem that needs solving. you can worry about the solutions once you've found the problem. i wish that app was around for your mom when she aborted you
|
# ¿ May 6, 2012 03:19 |
|
nukethewhales posted:same re: your mother, she might not have changed her mind in a moment of weakness a shameful combination of "rah rah" and "rubber/glue" posting perhaps for your next trick you get out
|
# ¿ May 6, 2012 17:33 |
|
PENETRATION TESTS posted:R is a complete monster and i use it constantly and it's growing like crazy and has a billion packages to do useful stats crap the answer is clojure + incanter use the best language (a lisp) on the best platform (the jvm) to do cool stats poo poo (via incanter, an R-esque package)
|
# ¿ May 7, 2012 05:54 |
|
ahhh spiders posted:are you for real all three are good, if you need to do those sorts of things
|
# ¿ May 7, 2012 06:00 |
|
R is really good but it's clear to me it's a stats package first and a language second.
|
# ¿ May 7, 2012 06:04 |
|
java would be much more fun to read code:
|
# ¿ May 8, 2012 17:14 |
|
Shaggar posted:check out all the cool things written in python like: mercurial
|
# ¿ May 8, 2012 18:19 |
|
Shaggar posted:trash lol shaggar, your main man java is housed in a hg repo
|
# ¿ May 8, 2012 18:29 |
|
please tell me you use svn
|
# ¿ May 8, 2012 18:31 |
|
Shaggar posted:of course i use source control (svn) hahahahhahahhahahahah
|
# ¿ May 8, 2012 18:34 |
|
Shaggar posted:also every time someone has a problem with svn its because ... multiple people are editing the same stuff. holy poo poo holy poo poo holy poo poo
|
# ¿ May 9, 2012 01:49 |
|
rotor posted:also git is hella slower than svn at my work but idk if that's normal if you're running windows, it's cause of windows, use mercurial instead
|
# ¿ May 9, 2012 04:27 |
|
MononcQc posted:Git and Mercurial and distributed source control is usually worth it if only because it lets you work and commit stuff without being connected to whatever server. Oh, how hilarious it is when some server crashes or network splits and half a day of work is lost to some lovely source control software that works over the network without ever thinking the network could go out for some reason. i don't use git a lot cause i became an hg fag first, but one thing i immediately noticed about hg is how much faster it was at calculating changes than svn. even if i just used hg like a svn replacement and didn't use anything else, it was a huge gain just from that
|
# ¿ May 9, 2012 14:29 |
|
rotor posted:the cool thing is when someone loses a bunch of work because they accidentally delete their local repo consider it like a canary. if your idiot coworkers delete their local repo, perhaps you should gtfo
|
# ¿ May 9, 2012 15:07 |
|
man this thread is about to hit peak trolling
|
# ¿ May 9, 2012 17:53 |
|
ahhh spiders posted:that's just lazy undersexed nerd crap though. look at some of the giant diatribes about how women shouldn't be in "male" professions one guy went out of his way to point out what will happen to "family values" or some poo poo if women don't stay in the kitchen
|
# ¿ May 9, 2012 22:30 |
|
Sulk posted:does anyone in here use cakephp or know anything about it drupal is pretty bad too
|
# ¿ May 10, 2012 23:26 |
|
Nomnom Cookie posted:im kinda pissed at java guys, they act like gc is super cool and everyone should use it and then when your heap gets over a couple gigs they're like "oh, you were going to *use* that heap? lol sucks to be you". at this point it looks like my best option is to break the one monster jvm into a bunch of little ones that can full collect independently and do something dumb to serialize requests to each baby jvm. all because I'm too dumb to manage my own memory and don't get to have memory pools how big is your heap? have you looked into azul
|
# ¿ May 13, 2012 01:11 |
|
that's actually not a horrible strategy; spin up worker vm's with huge heaps every time you need to run a job and reclaim them immediately when the job is complete. there's other mechanisms i heard about from azul like using some huge bytebuffer as some kind of scratchpad and never letting go of it but i've never had to go above ~ 8GB of heap and definitely never needed to gc that much so good luck i guess :/
|
# ¿ May 13, 2012 02:14 |
|
BonzoESC posted:you also have to remember that some ruby people think wearing a helmet is key to extreme programming hand on mouse? not programming.
|
# ¿ May 13, 2012 15:52 |
|
fidel sarcastro posted:i'm pretty sure it still needs programmers two drums and a cymbal fall off a cliff
|
# ¿ May 14, 2012 02:05 |
|
Internaut! posted:that reminds me, so that flash guy I mentioned who interviewed at google didn't get the job but he's determined to learn the foundations of CS and whatnot so I said to get cormen and k&r but is there an accessible book that describes computer languages/compilers/runtimes, like the process from writing a line of HLL code through to it being executed lil bit beyond that but: http://www.amazon.com/The-Elements-Computing-Systems-ebook/dp/B004HHORGA/ref=sr_1_2?ie=UTF8&qid=1336768358&sr=8-2
|
# ¿ May 15, 2012 21:29 |
|
BonzoESC posted:yeah pretty much that was tim sweeney of EPIC MEGAGAMES and he was talking about how cool haskell is Tim S posted:Factoid: C# exposes more than 10 integer-like data types, none of which are those defined by (Pythagoras, 500BC). beware, comic sans approaches http://www.cs.princeton.edu/%5C~dpw/popl/06/Tim-POPL.ppt
|
# ¿ May 16, 2012 14:54 |
|
Internaut! posted:
it is awesome
|
# ¿ May 17, 2012 03:09 |
|
groovy is kinda sorta sponsored by vmware, by virtue of them buying most of the dudes who made it and grails
|
# ¿ May 17, 2012 12:58 |
|
Internaut! posted:did not know this it's ok, i used to be a real groovy fag but now i just use it for xml processing and grails (i did write an edi builder in it that was pretty bad rear end, it was good for that too) it has some syntactic and semantic idiosyncrasies from when the language was young that have carried into the 1.8 release that are supposed to get fixed by 2.0. little edge case-y poo poo that will burn your rear end. its got some pretty leaky abstractions and once i saw clojure i was like "welp sorry groovy"
|
# ¿ May 17, 2012 15:00 |
|
Shaggar posted:eclipse owns. sorry ur dumb or w/e both eclipse and rcp are loving terrible holy poo poo never again
|
# ¿ May 17, 2012 15:41 |
|
Mr Dog posted:So I just learned that Python iterators throw an exception to indicate end of sequence in my head i thought "no way" sure enough: http://docs.python.org/library/stdtypes.html#iterator-types posted:iterator.next() guido....
|
# ¿ May 17, 2012 15:43 |
|
|
# ¿ May 13, 2024 16:19 |
|
BonzoESC posted:can't you just pass a block to the iterator instead of doing scrub poo poo like writing your own loops? idk poo poo about python cause i went ruby instead but wouldn't that use the same "feature" just under the covers?
|
# ¿ May 17, 2012 16:08 |