|
|
# ? Mar 6, 2014 20:19 |
|
|
# ? Apr 19, 2024 02:18 |
|
|
# ? Mar 6, 2014 23:39 |
|
i fed the floridaman twitter account into a markov chain generatorquote:Florida Man Poses as Santa and Bath Salts
|
# ? Mar 9, 2014 05:47 |
|
Florida Man Arrested For Not Looking Enough Like Britney Spears
|
# ? Mar 9, 2014 08:32 |
|
Luigi Thirty posted:i fed the floridaman twitter account into a markov chain generator lol
|
# ? Mar 9, 2014 12:46 |
|
Luigi Thirty posted:i fed the floridaman twitter account into a markov chain generator i lol'ed
|
# ? Mar 9, 2014 13:30 |
|
802.11weed posted:Florida Man Arrested For Not Looking Enough Like Britney Spears
|
# ? Mar 9, 2014 17:23 |
|
quote:Florida Man Poses as Busty Female Friend, Actually Picks Up Dominos Delivery Man Shoves Drugs Gave Him at IHOP; Demands NSA Phone Taped to Check Status of Chicken Statue Hitched To Be a Warrant, So He Could Steal Nikes, Says Privacy is Not Before Being Invaded Because Mean Neighbors With Pole Made Out at Neighbors Home, Removes Clothes, Starts Shouting 5,6,7,8 Florida Man Has a Busy Day
|
# ? Mar 10, 2014 06:03 |
|
c idiot spare time project s: did you know that arduinos are a really excellent little teaching tool to demonstrate memory register overflows? i learned that today.
|
# ? Mar 10, 2014 06:06 |
|
non-computer idiot spare time project: homebrew beer batch #2 is now conditioning. unlike batch #1, it actually carbonated.
|
# ? Mar 10, 2014 07:27 |
|
i'm writing my own operating system inspired purely by theoretical physics its pretty cool, hilariously math and metamath, programming and meta-programming heavy but i think its gonna own bones e: and actually be legit, too, pm me if you want in on the git repo Mr SuperAwesome fucked around with this message at 08:01 on Mar 10, 2014 |
# ? Mar 10, 2014 07:51 |
|
I've been doing the mstasano embedded security ctf. it's a lot of fun but a huge time sink b/c i spend hours stepping thru asm in a web based debugger on a fictional embedded device microcorruption.com
|
# ? Mar 10, 2014 09:02 |
|
Mr SuperAwesome posted:i'm writing my own operating system inspired purely by theoretical physics
|
# ? Mar 10, 2014 09:08 |
|
baiscally the idea is i am a lazy programmer and a lazy physicist and i don't like working but i do like thinking. apply this to programming and u lean towards best practices instead of doing a bunch of hacking - SOLID, GRASP, the unix philosophy apply this to sysadmining and u lean towards full automation, ansible for all ur linux janitoring needs apply this to webs and u just use pre-built frameworks apply this to web apps and you get MVC apply this to getting poo poo done and you get: buy a loving mac and iphone and ur life is easy apply this to physics, and you get fundamental lazy but hard math like lagrangian mechanics (study of evolving systems - powerful because it only considers external constraints, state, and a set of rules), and also quantum mechanics (very fundamental, very powerful, also loving weird and hard to explain) - remember feynmann said if u can't explain something to a freshman undergraduate, we don't understand it yet, this is true so ok you've got all this poo poo and you wanna build a physics thing in the cloud. so u janitor up ur servers and poo poo and sperg the gently caress out with git and because you've done it all right it's really well documented, really nicely organised in separate chunks (different sets of things, buckets, repos, u get the idea), along with states (git hashes, hashes), or crypto functions (essentially using a one time pad of a certain state - whether its fartboners ur lazy password from when u signed up for some porn site or w/e or a full spergy SSH key thing, its just a complicated hash of a one time pad and given enough time you could rainbow tables it (hence salts and why we laff @ adobe and r scared by the nsa). since uve gone to all this effort to get the stack and poo poo easily replaceable and scaffoldable, bootstrappble (scaffoldable: list of routines, boostrappable, quickly adding an extra config or setup or template thing) it seems a shame to waste it all on the physics so start off with data "measurements" "observables" as input and other poo poo as output and steal a few things from functional programming (its math after all) - if you want something to be realiable as a data input function its gotta be able to handle malformed input, so a pro way of doing this is with a monad then you've gotta try figure out how the underlying stuff all fits together. thats all just math functions basically. just lots and lots of equations. nothing but equations. loving hell that's boring its a good thing our friend stephen wolfram of wolfram mathematica fame (which saved my first year of my degree because i could just push butan get answer instead of doing a physics [it also explains it which helps]).... has written a book about "a new kind of science" using just number patterns. seriously check out that book it owns bones and has some scary as gently caress pictures in - just generated from weird patterns of iterative sequences... and they do some interesting things. sometimes these iterative patterns converge and do a thing, sometimes they go loving crazy, but most of the time they're pretty boring.... kinda like physics dont u think F = ma, linear boring, rocks falling about, who gives a gently caress thats old news. quantum mechanics and nanoscale poo poo howerer, thats cool as gently caress - gimme some fuckin 10 vars + consts shcroedinger equation to scare the poo poo out of me and generate cool results (neverwet, computers, etc) but the ultimate _building blocks_ - which is what we look for as a programmer, after all, programs is legos on steroids - in math are much more fundamental which is basically what stephen woflram is hinting at. equations modelling dynamic systems aren't ACTUALLY true to be honest. most of the time they are spherical cow approximations and in real life its gonna be weird because there something u haven't accounted for. but in the simplest case it'll still work as exepected. but given that they're all made up of legos, and any physics thing is just a pretty thing made out of math latex shaped legos, then math must be building blocks too so how about we write an object oriented math? well this takes you down into the annals of ZFC set theory, godel's incompleteness theorem, hilbert's fundamental questions, manifolds, topology, complex analysis, blargh. its all loving hard poo poo. but the basic takeaway point is its all just one big matrix. its all just nested arrays, and each of things in arrays are a bitmask for a specific constant or variable in the equation. manifolds defines what the vars are, topology defines how they fit together, and the order of the algo for that situation (be it simple like adding, or complicated like doing a schroedinger state evolution thing. but either way it can be formulated in terms of operators, observables, measurement outcomes, projection spaces and poo poo like that. but given it's all matrices and just a weird rear end arangement of them with some numbers, if u know the theory, u can generalise that to ur situation just like we've been going for the general lazy option before. so how do we represent math objects? this is set theory. a thing is just a set of sets, each with stuff in them (or not). think like a binary tree with pointers. it doesn't matter what you actually point to or how big it is or what it does, jsut that it is. also all our operators/equations/functionals/whatever are again just iterated steps. even x^n, ln n, e^n, sin(x), e^{ix}, different versions of similar poo poo (like E = 1/2 mv^2 and E = mc^2) but each is contextualised and without context of the state ur using it in (and what ur doing with it) it makes no loving sense other than just a pretty latex picture. there's continued fraction representations of most of these too, or other iterative approximations... see where we're going here? so regardless of how complicated your equation, it's just iterated steps on an input - lambda calculus, sw8 this is easy so we go back to the model idea, what input do we start with for our functions? well, numbers, sure butwhat types of numbers? well even numbers themselves are sets. the number {0} is the empty set which is the only thing required by axiomatic set theory other than the axiom of choice - think the choice like turing's instruction head on the halting tape machine ] e: gently caress this is getting long i better post before i accidnetlaly cmd-q ee: wc 52 1117 6350 lol Mr SuperAwesome fucked around with this message at 09:51 on Mar 10, 2014 |
# ? Mar 10, 2014 09:49 |
|
smoke weed write code do math
|
# ? Mar 10, 2014 09:54 |
|
Mr SuperAwesome posted:baiscally the idea is i am a lazy programmer and a lazy physicist and i don't like working but i do like thinking.
|
# ? Mar 10, 2014 10:07 |
|
Mr SuperAwesome posted:baiscally the idea is i am a lazy programmer and a lazy physicist and i don't like working but i do like thinking. losethos 8.1 lookin' good
|
# ? Mar 10, 2014 10:13 |
|
are you irl all right mr superawesome
theadder fucked around with this message at 10:16 on Mar 10, 2014 |
# ? Mar 10, 2014 10:14 |
|
not especially rigorous but this works 99% of the time, and that's half the point. math is trying to work out the core abstractions and they're gonna gently caress up a couple of times. this is why we developed calculus, and the calculus of variations feynmann's thing where he explains the principle of least action in lagrangian/quantum mechanics is really good at showing how powerful this is essentially the takeaway message is that as long as your dimensionality is right, and you know the vector spaces you're dealing with (strictly hilbert spaces i think) then all you need to do is vector callculus, grad/div/curl operators on these, and it doesn't matter how close you are to the right answer (nash equilibrium), it just matters that you're moving in that direction as opposed to not moving, its just the magnitude of ur momentum in a certain direction. whether this is a ball on a rubber sheet or a charge in an electric field. it also implies path independance, as long as there's no external energy input (i.e. no work is done by an outside force on the thing) - which is analogous to functions not changing state of objects in programming (which they won't unless you do an explicit thing with some sort of stdio or stdout
|
# ? Mar 10, 2014 10:18 |
|
yeah i know this sounds totally nuts so i have actually structured the thoughts and stuck it in git. the concepts are quite elgeant but they're quite a few levels deep (kinda like ramanuja style) - you know there's something weird going on underneath that describes an underlying connection, but since its very abstract, explaining this conceptually is hard. words are hardwikipedia posted:Srinivasa Ramanujan (22 December 1887 – 26 April 1920) was an Indian mathematician and autodidact who, with almost no formal training in pure mathematics, made extraordinary contributions to mathematical analysis, number theory, infinite series, and continued fractions. its all the same sort of math styles. in the same league as mathematicians such as Euler and Gauss. - both of these guys also had insanely deep mathematical insight, all the way down to number theory, manifolds, and topology (gauss's laws of EM led to maxwell's vector wave formulation of electromagnetics) as did von neumann, hilbert, godel, and nash its less about the rigour but more about the concepts basically Mr SuperAwesome fucked around with this message at 10:23 on Mar 10, 2014 |
# ? Mar 10, 2014 10:20 |
|
dont compare yourself to ramanujan e: holy hell, or euler or gauss, or von neumann, hilbert, godel or nash
|
# ? Mar 10, 2014 10:24 |
|
-
coffeetable fucked around with this message at 10:48 on Mar 10, 2014 |
# ? Mar 10, 2014 10:24 |
|
this came off me trying to solve poker with computers which is very doable and taught me all this number theory information theory poo poo theres a bunch of awesome poo poo on github where peopel have pre-computed pretty much all the "equivalence classes" for poker hands onto hashed numbers as a lookup table thing and it looks really really great so i got baked for a few days and hacked together a theory wrote a paper and took it to my phys tutor whos a quantum information theoriest and it was exactly like a game he'd been doing to get unbreakable crypto, and hed been figuring out that as long as you have a one time pad - which can be as simple as flipping a coin - you can have perfect information securty (without quantum encryption) and gave me a couple of ideas (the papers not public until theres no longer money in poker. lol) i did just come up with this all while yosposting stoned and reading wikipedia so it could all be complete trash, but i'm like 3 hours away from getting a mathematica from steroids together purely on metaprogrammng (it'll also be fast as gently caress because it's working with pure types - no floats needed, chars, literally everything is a pointer. this might be like c++ but i have no idea gently caress learning c++ it'll also self-compile i think and be parallel from the get go
|
# ? Mar 10, 2014 10:40 |
|
the turing halting problem seems to be the axiom of choice with regards to your input (are you feeding it an infinite series of digits? then it will never end), or output (are you doing an infinite loop with flow of control) if everything is pointers you can find this with recursive unit testing - and just test it against the domain of the function in a few places and see where it might lie
|
# ? Mar 10, 2014 10:44 |
|
jesus christ
|
# ? Mar 10, 2014 10:44 |
|
lmao did someone discover markov chain generators
|
# ? Mar 10, 2014 10:49 |
|
does yr "what would you do if you were hacking stoned and accidentally created skyne" thread have anything to do with this
coffeetable fucked around with this message at 11:00 on Mar 10, 2014 |
# ? Mar 10, 2014 10:51 |
|
holy poo poo
|
# ? Mar 10, 2014 11:16 |
|
5ing this thread before it gets sent tot he goldmine
|
# ? Mar 10, 2014 11:58 |
|
Nelson MandEULA posted:5ing this thread before it gets sent tot he goldmine
|
# ? Mar 10, 2014 12:26 |
|
Mr SuperAwesome posted:i think this is actually really simple. it's just a turing machine that takes 0, 1, ? as the symbols on the tape, the key is the ternary as opposed to binary view. the 0 and 1 represent truth, the ? represents inherent uncertainty, at a QM level this is H bar, otherwise its the uncertainty in your measurement. (systematic error + random) tobleronetriangular would be proud
|
# ? Mar 10, 2014 12:44 |
|
Mr SuperAwesome posted:...and hed been figuring out that as long as you have a one time pad - which can be as simple as flipping a coin - you can have perfect information securty (without quantum encryption)... Groundbreaking poo poo right here.
|
# ? Mar 10, 2014 12:59 |
|
otps are super secure
|
# ? Mar 10, 2014 14:02 |
|
Bloody posted:otps are super secure I filled one hard drive with random poo poo, and the other holds all my backups. I xored them and then wiped the random drive. Now my stuff is encrypted so deep nobody gonna peek into my poo poo.
|
# ? Mar 10, 2014 14:10 |
|
Bloody posted:otps are super secure they're inherently secure until they are known by another party, however
|
# ? Mar 10, 2014 14:42 |
|
truely bleeding edge research happenng itt
|
# ? Mar 10, 2014 14:55 |
|
|
# ? Mar 10, 2014 15:09 |
|
Nelson MandEULA posted:losethos 8.1 lookin' good
|
# ? Mar 10, 2014 15:27 |
|
Mr SuperAwesome posted:i did just come up with this all while yosposting stoned and reading wikipedia so it could all be complete trash, but i'm like 3 hours away from getting a mathematica from steroids together purely on metaprogrammng (it'll also be fast as gently caress because it's working with pure types - no floats needed, chars, literally everything is a pointer. this might be like c++ but i have no idea gently caress learning c++ is this the real life or is this just fantasy so you solved poker, are a math god, recreated mathematica, but c++ scares you??? how's that weed psychosis setting in?
|
# ? Mar 10, 2014 16:05 |
|
|
# ? Apr 19, 2024 02:18 |
|
Weed: Not even once.
|
# ? Mar 10, 2014 16:11 |