00:03:37 <Sgeo> I don't think so
00:03:45 <Sgeo> Well, Cookies: is
00:03:54 <Sgeo> And Set-Cookies:
00:05:00 <Sgeo> wtf http://i.imgur.com/crossdomain.xml
00:05:14 <Sgeo> http://imgur.com/crossdomain.xml
00:07:21 -!- Phantom_Hoover has quit (Ping timeout: 258 seconds).
00:08:01 <kmc> don't get pickles in your cookies
00:08:07 <kmc> websec / cooking advice
00:09:18 <mnoqy> im sure someone enjoys pickle cookies
00:09:34 <Bike> someone you don't want doing your websec
00:16:40 <Sgeo> Tables do not work well in epub
00:16:47 <Sgeo> Or well, the ones in this book don't :(
00:22:30 <kmc> we used to make chicken cookies
00:24:12 -!- quintopia has quit (Ping timeout: 252 seconds).
00:25:46 -!- Phantom_Hoover has joined.
00:27:56 <kmc> we used to make chicken cookies in this country, now we just have our hand in the next guy's pocket
00:32:08 -!- zzo38 has quit (Remote host closed the connection).
00:34:26 -!- Phantom_Hoover has quit (Read error: Operation timed out).
00:34:59 -!- Phantom_Hoover has joined.
00:44:43 -!- Phantom_Hoover has quit (Read error: Operation timed out).
00:52:36 -!- Tritonio_ has joined.
00:54:08 -!- Tritonio_ has quit (Read error: Connection reset by peer).
00:54:45 -!- Tritonio_ has joined.
00:55:47 -!- Tritonio has quit (Ping timeout: 272 seconds).
00:58:57 -!- Phantom_Hoover has joined.
01:03:52 <kmc> python has some neat higher order functional stuff, even if they avoid the label
01:03:55 <kmc> like http://docs.python.org/2/library/contextlib.html#contextlib.contextmanager
01:04:10 <kmc> turns a coroutine that yields once into an exception safe resource scoping thing
01:05:17 <Sgeo> Hrm. What's that page that demonstrated a Javascriptless chat room using a gif?
01:07:07 <kmc> though i've realized I'm using it wrong, sigh
01:07:20 <kmc> shachaf: https://github.com/videlalvaro/gifsockets ?
01:09:28 <Bike> Sgeo: [shachaf's up-hand character]
01:09:57 -!- Phantom_Hoover has quit (Ping timeout: 248 seconds).
01:10:44 <Sgeo> As in, it would leave a connection to a gif open and the 'gif' would get a new frame when someone entered a line
01:11:12 <Bike> «to the anonymous person that submitted the fairly competent, seemingly unironic, furry digimon(?) inflation porn version of “Squirt Force”, as an obfuscated url, to our ask box, i have a couple questions» jerkcityhd is going well
01:13:52 <Bike> Sgeo: pretty sure that's kmc's link
01:15:16 <Sgeo> kmc said shachaf though
01:15:41 <Sgeo> Oh, that's what you meant by [shachaf's up-hand character]
01:15:52 <kmc> yeah I meant "Sgeo:" whooops
01:15:54 <Sgeo> I thought you were saying embedding images into chat
01:16:12 <kmc> the keys are practically right next to each other
01:16:42 <Bike> I don't know if that's the same.
01:16:46 <Bike> shachaf is mysterious.
01:16:53 <Sgeo> I could have sworn that there was an old implementation
01:17:11 <Sgeo> Which would imply not clojure
01:17:12 <Bike> There's a white one but not a black one, even though there are black ones for left and right...
01:17:28 * Bike adds "racist" to Unicode complaints, realizes that was probably already there from Han unification or something
01:19:09 -!- Koen_ has quit (Quit: Koen_).
01:28:23 -!- Phantom_Hoover has joined.
01:32:36 <Bike> i wonder if you could encode those scripts' characters with series of radicals or something (spoiler i don't know any chinese scripted languages)
01:35:00 <pikhq> Bike: Though not done, this would be entirely practical.
01:36:08 <Bike> hm, both the major non-unicode representations of chinese (big5 and guobiao) seem to just use characters the same way
01:36:52 <Bike> i'm having that "if this was a workable idea someone would have tried it" thought
01:38:10 <Sgeo> Is there any more reason to cater to NoScript people than no-cookie people?
01:38:11 <pikhq> It's definitely a workable idea. It's just that it's computationally intensive to compose characters that way.
01:38:35 <kmc> it might go against the Unicode idea of what a "single character" is
01:38:40 <kmc> which I'm still not super clear on
01:38:40 <pikhq> Compared with just using a giant lookup table of codepoint to font.
01:39:03 <Bike> hm, that's true, composition would be nasty
01:39:15 <pikhq> kmc: A Unicode "single character" is more-or-less an abstract notion of the glyph.
01:39:22 <pikhq> Which can be composed with one or more codepoints.
01:39:23 <Bike> composition aka "something you have to do to actually display the characters"
01:39:41 -!- Phantom_Hoover has quit (Ping timeout: 268 seconds).
01:39:47 <Bike> "something you probably want to be simpleish for a usable system"
01:39:56 <pikhq> However, in terms of it being *able* to express Chinese characters, it's certainly workable.
01:40:07 <lifthrasiir> Bike: that's why Hangul has both precomposed characters and composable characters
01:40:16 <lifthrasiir> (the former called syllables, the latter called jamos)
01:40:24 <Bike> lifthrasiir: isn't hangul like, way more reasonable than han in a lot of ways
01:40:24 -!- Phantom_Hoover has joined.
01:40:44 <Bike> correspondance between glyphs and sounds for instance
01:41:09 <lifthrasiir> Bike: well, Han ideographs also have Ideographic Description Sequence (IDS) for the composable characters, except that they are not normalizable at all
01:41:13 <pikhq> The relevant bit here is that the composition rules for hangul are simpler and more regular.
01:41:40 <Bike> gah, thinking about languages makes me think about phylogeny
01:41:45 <pikhq> Though I think there's actually *more* jamo than radicals?
01:41:47 <Bike> and phylogeny is just a stupid clusterfuck how appropriate
01:42:27 <lifthrasiir> when I'm saying "practically" it refers how Hangul was implemented in Unicode
01:42:29 <pikhq> But there's only a few fixed ways that jamo combine into a single syllable block, making it a lot easier to do on a computer.
01:43:09 <kmc> wait, are the precompose hangul blocks actually in unicode?
01:43:09 <lifthrasiir> right, but when the character set was defined for the first time (around 1980) it was not practical at all
01:43:21 <pikhq> kmc: Both must be in Unicode for round-trip compatibility.
01:43:36 <kmc> round-trip compatibility is unfortunate
01:43:36 <lifthrasiir> (that's a fine chunk of BMP space actually)
01:43:50 <Bike> Wait, why do you need both exactly?
01:43:56 <Bike> Round trip through what?
01:44:00 <kmc> other encodings
01:44:04 <pikhq> Bike: Legacy Korean charsets had both.
01:44:23 <kmc> a rather onerous design goal of Unicode is that you can convert any encoding to it and back without losing information
01:44:26 <pikhq> Unicode is designed so that legacy charset -> Unicode -> same legacy charset doesn't change anything.
01:44:33 <lifthrasiir> pikhq: one catch here is that legacy Korean charsets do *not* have composable characters at all
01:44:39 <Bike> You couldn't roundtrip other encoding's precomposed char -> sequence of jamo -> recomposed char?
01:44:42 <pikhq> lifthrasiir: Oh. Sigh.
01:45:21 -!- augur has joined.
01:45:29 <lifthrasiir> they are designed for the two-byte-sequence-fill-two-column terminals
01:46:06 <lifthrasiir> they even do not cover every modern Hangul syllable either
01:46:06 <pikhq> Right, yes, that particular oddity of legacy CJK encodings.
01:46:45 <lifthrasiir> and MS decided to add remaining 8822 of them in the way incompatible to EUC for just that reason
01:46:58 <lifthrasiir> (more standard-ish way involves 3-byte sequences)
01:47:32 <lifthrasiir> so it is more like a historic and political issue
01:47:42 -!- Phantom_Hoover has quit (Ping timeout: 268 seconds).
01:47:50 <pikhq> Character encoding is freaking complicated.
01:48:00 <kmc> praise God who sent His prophet Ken Thompson to bring us the good word of UTF-8
01:48:21 -!- Phantom_Hoover has joined.
01:48:37 -!- Tritonio_ has quit (Read error: Connection reset by peer).
01:48:44 <lifthrasiir> once upon a time there was Plan 9 (from whatever)
01:48:49 <Bike> clearly i should organize a committee of ROC and PRC reps to work out how chinese will work for once and for all. the resulting violent collapse of civilization will leave all character encoding problems solved by default
01:48:52 -!- kmc has set topic: Happy UTF-8 Appreciation Day everyone! | http://underhanded.xcott.com/?page_id=5 | http://codu.org/logs/_esoteric/.
01:49:00 -!- Tritonio_ has joined.
01:49:18 -!- augur has quit (Remote host closed the connection).
01:49:21 <Bike> > "Happy UTF-8 Appreciation Day everyone!"
01:49:22 <lambdabot> "Happy UTF-8 Appreciation Day everyone!"
01:49:29 <Bike> i was expecting homograph attacks.
01:49:30 -!- TeruFSX has joined.
01:49:41 <kmc> well one thing to appreciate is that it's a superset of ASCII :)
01:49:47 <lifthrasiir> Bike: there are Chinese characters for every discovered atomic elements
01:49:57 <Sgeo> "This has resulted in a substantial number of web application vulnerabilities, but to this day, Internet Explorer developers seem to have no regrets and have not changed the default behavior of their code."
01:50:18 <Sgeo> (About content-sniffing HTML when being told it's a text/plain)
01:50:56 <kmc> Hong Kong and Macau also use a different character encoding from mainland PRC
01:50:59 <pikhq> Aside from a handful really old elements, IIRC the element characters have some rather nice regular properties.
01:51:11 <Bike> do you mean trad? or something worse
01:51:15 <shachaf> Speaking of Ken Thompson, "A few minutes ago I spoke with Ken Thompson. He confirms that he and his colleagues never wanted argc to be 0; and he agrees with me that the authors of C89 should not have allowed that case to be legal. He is pretty sure that Plan 9, in particular, always ensured that argc>=1, by supplying a default environment if necessary."
01:51:24 -!- augur has joined.
01:51:28 <kmc> well I guess it is related to the trad / simplified split
01:51:38 <kmc> but i meant Big5
01:51:47 <lifthrasiir> pikhq: radicals represent the type of those elements at STP?
01:51:48 <pikhq> i.e. all the metals have 金 (metal), the gasses have 気 (air/gas/spirit/this character has a lot of semantics), the liquids 水 (water/liquid).
01:52:06 <Bike> oh, chinese does the breath-life thing too, eh
01:52:56 <lifthrasiir> fortunately for Chineses their pronounciations follow those of non-radical parts
01:53:19 <pikhq> I'd be tempted to still call those "radicals", but yeah.
01:53:19 <lifthrasiir> (it is actually common for many Chinese characters)
01:53:48 <pikhq> Though that's for lack of a better term for the more generic "Chinese character component".
01:54:04 <lifthrasiir> but they still have to memorize the meanings and pronounciations of at least thousands of basic characters
01:54:09 <pikhq> As opposed to "Chinese character component that is used for dictionary lookup, that generally has some amount of semantic meaning."
01:57:31 -!- Bike has quit (Ping timeout: 245 seconds).
01:58:03 <shachaf> mnoqy: got any monqy wisdom and/or mnoqy wisdom for me today
01:59:27 -!- Bike_ has joined.
02:00:05 -!- Bike_ has changed nick to Bike.
02:00:14 <Sgeo> It's always fun when a book on web security begins a sentence with "To their horror and disbelief"
02:00:36 <Bike> the implication that they do this often enough for you to notice is good
02:01:26 <Sgeo> No, sorry. Although things that imply sheer pain do occur often enough
02:01:30 <kmc> yeah I like the writing style of that book
02:01:39 <kmc> it's engaging and not dry
02:01:49 <kmc> the author finds it all amusing in a "we're all fucked" kind of way
02:02:00 <kmc> I think there are some intrinsic properties shared by good exploits and good jokes
02:02:18 <Bike> that would explain roughly everything about hacker culture
02:02:26 -!- Phantom_Hoover has quit (Read error: Operation timed out).
02:04:33 <shachaf> Probably I should read that.
02:04:47 <shachaf> Did you see http://lcamtuf.coredump.cx/postxss/ ?
02:04:47 <kmc> i should finish reading it
02:05:02 <Bike> http://en.wikipedia.org/wiki/Reverse_migration_(birds) how could anyone not like biology, i ask you
02:05:03 <Sgeo> o.O coredump.cx is real?
02:05:41 <shachaf> Is the book like that page?
02:06:33 <kmc> oh, well, it's the same author
02:06:35 <Bike> .cx is the best tld
02:06:41 <kmc> i think it's pretty similar
02:06:45 <shachaf> That page is full of cleverness. A lot of these sorts of things are.
02:06:51 <Bike> what the heck, does anyone use <textarea> now
02:07:08 <kmc> Bike: i know someone who got into a LONG wikipedia edit war about whether goatse.cx should be listed on the .cx page as one of the notable domains
02:07:34 <Bike> kmc: awesome (it totally is)
02:07:41 <pikhq> That's pretty neat, but c'mon, .su is for a non-existent state!
02:08:12 <kmc> http://en.wikipedia.org/wiki/Country_codes_of_Serbiais amusingly complicated
02:08:12 <pikhq> North Korea exists, it just has no right to. :P
02:08:38 <kmc> pikhq: I want to see alternate history fiction where the USSR survives into the Internet Age
02:08:53 <Bike> i'm working on some in my head >_>
02:08:56 <pikhq> kmc: Arguably it did. :P
02:09:00 <kmc> well, but just barely
02:09:11 <Bike> also a guy i know elsewhere sometimes bans .kp and the one i forget for east germany
02:09:12 <pikhq> .su was created a mere 14 months before the collapse.
02:09:33 <kmc> i mean like the age of people generally giving a shit about the internet
02:09:43 <kmc> man, remember the days when you could say "The Internet is just a fad" seriously?
02:09:55 <Bike> too young. sorry!
02:10:09 <pikhq> Bike: East Germany never got one assigned.
02:10:10 <shachaf> Bike: It's actually kmc who's too old.
02:10:38 <pikhq> That was never assigned.
02:10:53 <Bike> that doesn't matter to silly bans!
02:10:55 <kmc> (or the dark interpretation: the open Internet *is* just a fad, it will eventually be replaced with walled gardens run by Google and Facebook and Apple)
02:10:58 <pikhq> DD is a valid ISO 3166-1 code, but that TLD never existed. :)
02:11:16 <Bike> "For a time, praise.hm offered a URL shortening service directed at Christians." cctlds are the fucking best
02:11:55 <kmc> they stuck South Sudan with .ss? awkward
02:12:16 <Bike> i doubt anyone around south sudan has occasion to give a damn about the schutzschaffel, really
02:12:26 <Bike> staffel. fuck. fucking german
02:12:50 <kmc> huh SF is an obsolete code for Finland
02:12:54 <kmc> "Suomi Finland"
02:12:58 <pikhq> del.icio.us is one of the better uses of ccTLDs though...
02:13:05 <pikhq> In that the site was actually American.
02:13:21 <kmc> http://responsiveurl.co.uk
02:13:22 <pikhq> lifthrasiir: Please tell me they use .co.ck for commercial services.
02:13:23 <lifthrasiir> and it is totally possible to have .co.ck domains
02:13:50 <lifthrasiir> it is pity that they don't support two-level domains yet
02:13:53 <kmc> 'In the British sitcom Nathan Barley, the principal character registers his website, "trashbat", in the Cook Islands simply so that it has the amusing ".co.ck" suffix, which he always pronounces "dot cock."'
02:14:24 <pikhq> "Domains considered profane will not be considered on any level, and the application will be dissolved with the applicant being notified, and future requests for the same domain name will be ignored."
02:14:39 <pikhq> So, is it possible to actually register a dot cock?
02:15:02 <kmc> and the application will be dissolved with the applicant being notified, and the earth will be salted so that nothing can grow for 40 generations
02:15:05 <Sgeo> I misread that as meaning that the application wouldn't be notified, which would suck
02:15:28 <Bike> dragon.co.ck has been rejected
02:15:44 <Sgeo> "andn the applicant will be dissolved with the application being notified"
02:16:11 <kmc> also https://www.youtube.com/watch?v=Qu-YbU7IIns
02:16:15 <pikhq> But what if I have a company called Dragon in the Cook Islands that sells dildos?
02:16:27 <pikhq> "Get your Dragon cocks at Dragon dot cock!"
02:16:34 -!- Phantom_Hoover has joined.
02:16:50 <kmc> more famous for https://www.youtube.com/watch?v=ss8LDBNcsWc
02:16:53 <Sgeo> We can only assume that anything sexual is completely illegal in the Cook Islands
02:17:30 <kmc> http://loves.co.ck/ sfw
02:17:36 <Bike> kmc: errrr what
02:17:44 <Sgeo> kmc, don't feel like opening page, is that the one where he's trying to say top cop?
02:17:50 <Bike> @ chickenfucking
02:18:05 <kmc> there's one where he misreads a url ".com" as ".cock" and then briefly corrects to "dong" before settling on "com"
02:18:37 <kmc> Bike: watch the face of the woman sitting next to him
02:18:48 <Bike> a camcorder recording of a tivo'd program. i love the future
02:19:47 <pikhq> This is a lot like how the MPAA recommends that copies of DVDs for educational use be made using camcorder.
02:20:39 <mnoqy> reminds me of https://www.youtube.com/watch?v=8zB4ItYcX4I
02:20:46 <kmc> http://en.wikipedia.org/wiki/List_of_Internet_top-level_domains#Internationalized_country_code_top-level_domains a fun list too
02:21:03 <Bike> pikhq: do they advocate getting actors to reenact plays rather than recording them
02:21:13 <pikhq> Bike: I don't think so.
02:21:17 <pikhq> The point here is DRM.
02:21:47 <Bike> bad at jokes again, woe
02:22:09 <kmc> Tamil script is pretty damn cool looking
02:22:14 <shachaf> reminds me of https://www.youtube.com/watch?v=nOtrEalLyaViDEo
02:22:27 <mnoqy> shachaf: that's not really a video, is it
02:23:02 <Bike> kmc: ooh, swirly
02:23:35 <kmc> http://upload.wikimedia.org/wikipedia/commons/f/fc/History_of_Tamil_script.jpg
02:23:40 <kmc> ADD MORE SWIRLS
02:32:56 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
02:46:04 <kmc> > 6 / pi**2
02:46:58 <Bike> some zeta sum, right?
02:48:26 <Bike> oh, one over zeta(two)
02:50:07 <kmc> it's the probability that two randomly chosen numbers are coprime
02:51:22 <Bike> 'randomly chosen number' is such a great concept
02:51:36 <pikhq> kmc: I assume "natural" or some such?
02:51:46 <kmc> yeah i think natural
02:51:53 <kmc> didn't know so i didn't specify :)
02:52:02 <Bike> "no, it includes reals, duh"
02:52:36 <kmc> it's cool that you can implement RSA in a page of code, even if it's terrible RSA nobody should actually use
02:52:50 <Bike> hm, that's not erdos-kac
02:52:56 <Bike> got my half-remembered number theory all confused, id o
02:52:59 <pikhq> Yeah, it's really quite nice that RSA is actually that understandable.
02:53:15 -!- zzo38 has joined.
02:53:30 <kmc> you need pretty deep math to actually prove Euler's totient theorem, right?
02:54:13 <Bike> " Euler's theorem can be proven using concepts from the theory of groups:" guess so
02:55:02 <kmc> well depends which concepts :)
02:55:22 <Bike> basel problem isn't too bad though
02:55:26 <kmc> oh actually I think this is an amount of group theory that I used to know
02:55:28 <shachaf> if it doesn't use invertibility then it's so easy
02:55:43 <shachaf> what does 'randomly chosen natural number' mean
03:06:03 <kmc> it seems really lucky that public key crypto is even possible at all
03:06:38 -!- Bike_ has joined.
03:06:38 -!- Bike has quit (Ping timeout: 255 seconds).
03:11:11 <Bike_> kmc: have you seen "A personal view of average-case complexity"? Impagliazzo's "five worlds" paper
03:11:21 <kmc> no, but that sounds really cool
03:11:27 <Bike_> it's really cool how that took me five minutes to look up
03:12:18 <kmc> do you have a link or should i google it
03:13:04 -!- Bike has joined.
03:13:40 <Bike> http://cseweb.ucsd.edu/users/russell/average.ps
03:13:47 <Bike> now if you'll excuse me i'll be destroying my ISP
03:13:57 <Bike> 20:11 < Bike_> it's really cool how that took me five minutes to look up
03:13:57 <Bike> 20:12 < Bike_> basically the guy lays out a few different ways the world would be like for various resolutions to P=NP
03:14:00 <Bike> 20:12 < Bike_> including whether trapdoor functions are real and stuff
03:14:44 <pikhq> Yeah, salt the earth;
03:14:57 -!- aloril has quit (Remote host closed the connection).
03:15:23 <Bike> it's weird how qwest is still in my hostname, considering qwest no longer exists, but eh
03:15:36 <pikhq> Right, yes, it's Century Link.
03:15:47 <pikhq> Salt the fucking earth or they'll come back with a new name.
03:15:59 -!- Bike_ has quit (Ping timeout: 246 seconds).
03:18:17 * kmc traceroutes the IP he had in high school, which he still has memorized
03:18:29 <kmc> 'hey i used to live here!'
03:18:42 <kmc> it seems to have moved to seattle
03:19:06 <shachaf> Hmm, I remember the DNS IPs but not my own address.
03:19:32 <shachaf> Well, it was before high school.
03:20:38 <zzo38> Do you know secret sharing cryptography? I know some things about this.
03:21:11 <Bike> do you mean private key cryptography aka "old-ass" cryptography
03:21:51 <kmc> maybe http://en.wikipedia.org/wiki/Secret_sharing
03:22:18 <Bike> hey look it's shamir again. dude gets around
03:22:22 <kmc> "Each secret share is a plane, and the secret is the point at which three shares intersect." wikipedia photo caption or 3am street person ramblings
03:22:24 -!- aloril has joined.
03:22:49 <Bike> aren't we all a plane, deep inside
03:23:47 <zzo38> Yes, like that, or a line, if there are two needed to make it, etc. I can think of things involving how to encode the data and how to figure out the numbers and so on, to improve the security.
03:34:54 <kmc> Secure multi-party computation is really cool and something I should learn about
03:38:04 -!- augur has quit (Remote host closed the connection).
03:40:01 <Bike> "The proof here is messy, but stupid."
03:42:53 <Sgeo> Pretty sure my old IP is somewhere on some wiki
03:43:30 -!- augur has joined.
03:45:27 <Sgeo> I remember when I wanted to use NetMeeting to connect to my computer from elsewhere, and thinking my IP address was 192.168.1.1
03:45:41 -!- augur has quit (Remote host closed the connection).
03:57:05 <Gregor> <Bike> aren't we all a plane, deep inside // But aren't we all snakes on a plane?
04:15:10 <kmc> Planesnake would be a good name for a band
04:17:14 <kmc> my #esoteric band names: Planesnake, Napier's Bones, KANGXI RADICAL FIGHT, Duff's Device, Intermezzo War Crime,
04:17:25 <kmc> Gütenbørg, BADNESS 10000
04:17:52 <mnoqy> lots of things would be good band names but i haven't thought to write any of them down....
04:18:00 <mnoqy> RIP / lost forever
04:18:12 <Bike> i know at least one person who collects band names in an irc-queryable list
04:18:22 <Bike> also does kangxi radical fight have to do with the emperor or...
04:18:49 <kmc> Too Close for Missiles, Jupiter and Beyond the Infinite, Berne Convention, Gopher Wood
04:19:03 <mnoqy> Bike: that's a good idea
04:19:26 <shachaf> zzo38: Do you have any band names for us?
04:19:30 <Bike> oh, kangxi radicals, compiled by the kangxi emperor
04:19:43 <kmc> mnoqy: you can use some of mine
04:19:54 <kmc> i have a surplus
04:20:53 <shachaf> band name surplus and band deficit((surminus??))
04:21:13 <mnoqy> that's one way to put it
04:21:15 <Bike> so should there be like `guitarsolo to spit out a band name
04:22:02 <shachaf> imo guitars are p. bad instruments
04:22:11 <Bike> why do you say that?
04:22:12 <shachaf> not as bad as drums but still p. bad
04:22:26 <shachaf> admittedly electric guitars are worse than non-electric guitars
04:22:49 <mnoqy> guitars can be used for good or bad
04:23:06 <shachaf> okay guitar users are the evil ones
04:23:09 <Bike> if you don't like spanish guitar i'm afraid we can't be whatever we are
04:23:21 <kmc> the electric guitar is one of the defining technologies of the 20th c.
04:23:49 <mnoqy> just how many defining technologies did the 20th c. have
04:24:37 <mnoqy> oh no!!who do i believe
04:24:55 <Bike> electric guitar, combine harvester, those cheapish well things
04:24:57 <Bike> that's about it
04:25:24 <shachaf> mnoqy: i was thinking of doing that thing that i did again, should i
04:25:48 <mnoqy> was it well-recieved
04:26:02 <Bike> maybe kalishnikovs. is that an invention
04:26:21 <mnoqy> more of a discovery imo
04:26:45 <Bike> right yeah your o is right
04:27:00 <Bike> wait shit what about those things you can put on dogs and then they translate dog language for you.
04:27:43 <Bike> that was 90s right? pretty defining of the 90s and the 20th century in general
04:27:54 <shachaf> the 90s didn't really exist
04:28:04 <Bike> Animal testing Baiting Breed-specific legislation Communication Dog attack Dog park Dog sports Dog walking Dog daycare Dog grooming Famous dogs Intelligence Therapy Training Fear of dogs Dog license Dog food Dogs in religion
04:28:35 <kmc> atomic bomb
04:29:04 <Bike> those things that were only used twice?
04:31:23 <kmc> and had a dominant effect on international relations in the remaining 50 years, yes
04:37:53 <Bike> i'm kind of automatically skeptical of technology being a super important factor in history though
04:40:08 <Jafet> Nuclear weapons have been deployed against: Japan, fishermen, Pacific islanders, Mongolian nomads
04:41:11 <Bike> who tested nukes in mongolia?
04:41:27 -!- augur has joined.
04:43:07 -!- augur has quit (Remote host closed the connection).
04:43:19 -!- augur has joined.
04:49:04 <Bike> i don't think mongolia ever had nuclear capability
04:49:47 <pikhq> That's because you are educated stupid.
04:51:26 <kmc> the CIA put some nuclear powered listening devices on a mountain in the Himalayas to spy on nuclear tests in China
04:52:25 <kmc> they fell down and ended up in the Ganges river
04:52:27 <Bike> it's kind of weird that mongolia didn't have more of a role in the cold war now that i think about it, i mean they were practically soviet and had had a whole lot of border conflicts with china
04:54:59 <pikhq> It's weird realizing that stuff like the Apple II's NTSC artifact color didn't work out of the US.
04:55:39 <pikhq> Particularly bad for the Apple II. That thing was actually designed entirely around that.
05:02:54 -!- nooodl has joined.
05:07:51 -!- nooodl has quit (Ping timeout: 272 seconds).
05:25:16 <zzo38> I think electric guitar and non-electric guitar can both be OK to play the specific music meant for those kind of guitar.
05:25:46 <pikhq> Electric guitar is awesome. Classical guitar is also awesome.
05:25:52 <pikhq> They're just very different instruments.
05:27:06 * Bike attempts to imagine what Anji would sound like on electric guitar, fails
05:27:45 <zzo38> pikhq: Yes. That is what I mean.
05:32:54 -!- TeruFSX has quit (Ping timeout: 256 seconds).
05:59:25 -!- carado has quit (Ping timeout: 246 seconds).
06:03:58 <kmc> heh, this paper on one way accumulators suggests (in passing) to use secure multiparty computation to bootstrap a new cryptosystem
06:04:25 <kmc> like if you need to pick some constants and don't want to just ask the NSA what the best ones are
06:04:36 <kmc> though there are advantages to asking the NSA as well as disadvantages
06:06:07 <Bike> clipper was great, right!?
06:13:43 <Fiora> http://en.wikipedia.org/wiki/Nothing_up_my_sleeve_number
06:15:35 <Bike> the way that's written sort of implies that you can just grab CK random numbers, nice
06:16:06 <Bike> i hope to one day cite A Million Random Digits with 100,000 Normal Deviates in a paper
06:17:03 <shachaf> _A Billion Random Digits with 100,000,000 Normal Deviates_, by Bike
06:17:28 <Bike> handwritten and published on 100% genuine papyrus
06:17:42 <kmc> in this case they're looking for numbers with a particular property
06:18:23 <Fiora> maybe they can list the property, and say, pick the first number in the digits of pi with that property?
06:19:34 <shachaf> But then you need to justify the index.
06:19:48 <Fiora> well, like, you justify it with the property
06:19:59 <shachaf> Oh, maybe I read that backwards.
06:20:09 <Fiora> "here is the property we want, here's a function that iterates through until it finds a number that fits it, we didn't pick anything weird beyond this property" I don't know if that works though
06:21:06 <Bike> Fiora: i think you may as well use a CSPRNG at that point?
06:22:58 <Bike> random number generator
06:23:00 <Fiora> I guess, but like, then you'd have to justify some numbers like te initialization values right?
06:23:15 <Fiora> like, you're having to convince people "I'm not doing anything weird here"
06:23:30 <Fiora> so if you're giving them some 64-bit seed or something you might have intentionally picked a seed that gave you some hidden property?
06:23:49 <Bike> I guess I'm thinking that they'd be able to come up with their own values easily if they wanted to.
06:23:51 <Fiora> um maybe this doesn't make any sense but I thought the whole thing was basically not having too much entropy in the constants
06:25:30 <Bike> maybe this is why we need secure multiparty computation
06:25:49 <Bike> that way we can hide the agreement on sanity in the math of the protocol, instead of actually talking about it, in which case nobody will ever agree
06:27:11 <kmc> now reading the zerocoin paper
06:27:52 <Fiora> zero-knowledge proofs still freak me out kind of
06:29:35 <Fiora> just the idea that you can prove something really complicated and significant without admitting any of your premises
06:30:38 <shachaf> But it's not actually "proving", generally.
06:39:01 -!- FreeFull has quit.
06:40:55 <kmc> zerocoin uses commitments, zero-knowledge proofs, *and* one-way accumulators
06:41:44 <Fiora> one-way accumulators?
06:43:51 * Bike googles, gets a bunch of pdfs. must be a common concept!!
06:44:15 <Bike> "A Decentralized Alternative to Digital Signatures" sounds pretty snazzy, though.
06:45:07 -!- nooga has joined.
06:45:27 <Bike> This paper describes a simple candidate one-way hash function which satisfies a quasi-commutative property that allows it to be used as an accumulator. This property allows protocols to be developed in which the need for a trusted central authority can be eliminated. Spaceefficient distributed protocols are given for document time stamping and for membership testing, and many other applications are possible.
06:45:47 <Bike> what's it accumulate? signatures of transmission middlemen?
06:47:25 <Bike> oh man, the original rsa paper
06:47:27 <Bike> «This has obvious applications in "electronic mail" and "electronic funds transfer" systems.»
06:50:52 <kmc> it accumulates whatever you want
06:51:07 <Bike> oh that's a good thing to accumulate
06:58:24 -!- Bike has quit (Quit: undie).
07:00:30 <kmc> oh i was going to link this in the earlier discussin of hardware random number generators https://www.youtube.com/watch?v=7n8LNxGbZbs
07:02:07 <Fiora> oh so basically it's a thing you can add sutff to, but can't remove things from?
07:05:31 <kmc> it lets you compute z from a bunch of y_i such that you can later demonstrate that your y was one of the ones that went into z
07:06:15 <coppro> anyone else seen the new doctor who?
07:06:34 <Fiora> kmc: but such that you don't have to say which y it is?
07:06:50 <kmc> no, you do say
07:07:12 <kmc> i think the main thing is that an adversary can't prove your y is in there if you've kept it secret
07:07:43 <kmc> also it takes up a constant amount of space no matter how many y are in there, unlike the simple 'concatenate all y and call that z' solution
07:08:10 <kmc> Fiora: zerocoin builds a zero-knowledge proof of membership though
07:08:21 <kmc> but that's not part of the idea of one-way accumulators themselves
07:08:26 <kmc> i think...
07:10:04 <kmc> so it's sort of like the value z is a public key that has 'signed' all the y, except there's no private key that can sign more things (and be stolen or coerced into doing so)
07:11:58 -!- Tritonio_ has quit (Ping timeout: 272 seconds).
07:30:20 -!- nooga has quit (Ping timeout: 260 seconds).
07:36:58 -!- Tritonio has joined.
07:41:42 -!- Tritonio has quit (Ping timeout: 276 seconds).
08:05:15 -!- nooga has joined.
08:10:40 -!- nooga has quit (Ping timeout: 246 seconds).
08:16:16 -!- epicmonkey has joined.
08:22:31 -!- zzo38 has quit (Remote host closed the connection).
08:42:12 -!- conehead has quit (Quit: Computer has gone to sleep.).
08:44:53 -!- mnoqy has quit (Quit: hello).
08:57:45 -!- ThatOtherPerson has joined.
09:09:33 -!- mnoqy has joined.
09:15:18 -!- epicmonkey has quit (Ping timeout: 276 seconds).
09:19:33 <fizzie> ThatOtherPerson: I hear it's relatively cheap to hire humans to do them for you.
09:20:38 <shachaf> fizzie: Yes, that's a relatively cheap move.
09:20:41 <ThatOtherPerson> fizzie: I probably should do that; I must not be human enough to be able to read them.
09:24:12 <fizzie> fungot: What'd you say about earning us some money by breaking a few CAPTCHAs?
09:24:12 <fungot> fizzie: je vais dir.... fnord
09:25:08 <olsner> does that actually mean "I want to say fnord"?
09:30:05 <olsner> fungot: what's that mean?
09:30:06 <fungot> olsner: work-safe but horrifying?" quiz for synapsid. this
09:30:47 <fizzie> fungot: I think you're pulling someone's leg here.
09:30:47 <fungot> fizzie: process 3 killed. it's a relative, not a shadowed variable. will do
09:38:35 <fungot> ThatOtherPerson: he was infamous for inflating time estimates, then delivering way ahead of the times
09:39:40 <fizzie> fungot: That's still no reason to KILL someone.
09:39:41 <fungot> fizzie: fnord was ega/ vga graphics and b800-bfff is text. and it provides a means to do unhygienic operations. that's just under the advanced line.
09:41:01 -!- Tritonio has joined.
09:43:40 <olsner> the hygienic/unhygienic/advanced spectrum of macro hygiene?
09:44:23 <olsner> vram hygiene might be a separate concept though
09:45:20 -!- Phantom_Hoover has joined.
09:49:28 <fizzie> olsner: I think it's related to how you can make macros that don't cause any visual noise on the monitor when expanded.
09:50:42 <ThatOtherPerson> @tell Taneb hmm, do you still think the grid cells should be changing size? It might make things tricky... I'll be working on other things for now
09:51:58 <fizzie> @ask Taneb also, do you think reversing the polarity of the tachyon matrix could streamline the efficiencies a bit?
09:56:30 -!- Phantom_Hoover has quit (Ping timeout: 264 seconds).
10:00:18 -!- ThatOtherPerson has quit (Quit: Page closed).
10:02:03 -!- Koen_ has joined.
10:12:47 -!- Phantom_Hoover has joined.
10:22:51 -!- mnoqy has quit (Quit: hello).
10:34:21 -!- sebbu2 has joined.
10:35:00 -!- sebbu2 has quit (Changing host).
10:35:00 -!- sebbu2 has joined.
10:37:12 -!- sebbu has quit (Ping timeout: 256 seconds).
11:02:02 -!- carado has joined.
11:05:00 <Sgeo> `olist NOT a new comic, but the website is back up
11:05:02 <HackEgo> olist: shachaf oerjan Sgeo FireFly
11:07:15 <shachaf> Why would you `olist when there's no new comic?
11:08:15 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: nolist: not found
11:22:57 -!- nooga has joined.
11:28:20 -!- nooga has quit (Ping timeout: 260 seconds).
11:34:24 -!- Gracenotes has quit (Ping timeout: 260 seconds).
12:05:07 -!- ais523 has joined.
12:08:49 -!- augur_ has joined.
12:09:55 -!- augur has quit (Read error: Connection reset by peer).
12:26:07 -!- ais523 has quit (Ping timeout: 272 seconds).
12:28:06 -!- ais523 has joined.
12:30:13 -!- nooga has joined.
12:31:09 <lambdabot> ais523: You have 2 new messages. '/msg lambdabot @messages' to read them.
12:32:10 <ais523> @tell elliott filters 2 and 3 don't have specific warnings, but also don't have any consequences but preventing the attempt to edit; I don't know what it looks like but it would be safe to find out
12:34:02 <ais523> @tell elliott: haha, I /just/ triggered it by mistake: "This action has been automatically identified as harmful, and therefore disallowed. If you believe your edit was constructive, please inform an administrator of what you were trying to do. A brief description of the abuse rule which your action matched is: userpage contains no newlines and adds a link"
12:37:25 -!- nooga has quit (Ping timeout: 248 seconds).
12:41:46 -!- nooga has joined.
12:44:01 -!- carado has quit (Ping timeout: 246 seconds).
12:50:04 -!- nooga has quit (Ping timeout: 252 seconds).
13:13:25 -!- boily has joined.
13:16:39 <elliott> ais523: nice telling "elliott:"
13:16:40 <lambdabot> elliott: You have 2 new messages. '/msg lambdabot @messages' to read them.
13:17:01 <ais523> I wonder if lambdabot knew what I meant anyway
13:17:21 <shachaf> Did ais523 secretly resend the same message without the :?
13:17:46 <shachaf> That seems more likely than someone fixing lambdabot.
13:17:58 <ais523> perhaps it was never broken, so didn't need to be fixed
13:18:01 <lambdabot> shachaf: You have 1 new message. '/msg lambdabot @messages' to read it.
13:18:23 <shachaf> Does it work with commas too?
13:18:44 -!- Phantom_Hoover has quit (Read error: Operation timed out).
13:20:48 <lambdabot> Where d' all t' pirates come from?
13:21:03 <elliott> shachaf: now /nick shachaf,
13:21:23 <ais523> are commas even legal in nicks?
13:25:36 -!- elliott_ has joined.
13:25:37 -!- elliott has quit (Quit: Reconnecting).
13:25:48 -!- elliott_ has changed nick to elliott.
13:34:45 -!- Phantom_Hoover has joined.
13:44:05 -!- Phantom_Hoover has quit (Ping timeout: 248 seconds).
14:04:48 -!- Phantom_Hoover has joined.
14:16:06 -!- ThatOtherPerson has joined.
14:19:41 -!- ThatOtherPersonY has joined.
14:21:17 -!- ThatOtherPerson has quit (Ping timeout: 245 seconds).
14:21:25 -!- ThatOtherPersonY has changed nick to ThatOtherPerson.
14:40:11 -!- LVCIFER has joined.
14:43:30 <elliott> ais523: btw I sent you a message on esolang
14:45:29 -!- sirdancealot has quit (Ping timeout: 252 seconds).
14:49:07 -!- nooodl has joined.
14:50:01 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
14:51:13 -!- FreeFull has joined.
14:52:31 <elliott> http://en.wikipedia.org/wiki/Anti-Barney_humor
15:03:18 -!- ais523 has quit.
15:19:25 -!- MindlessDrone has joined.
15:25:07 <HackEgo> MindlessDrone: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
15:35:44 <fizzie> http://sprunge.us/jAUF why isn't our channel hip and popular enough to get this sort of stuff?
15:38:30 <fizzie> Well, you wouldn't want people to assume anything untrue.
15:41:10 <elliott> the #haskell spambots are a bit boringer
15:48:41 <elliott> my platform for op: i wouldn't op ThatOtherPerson
15:54:16 -!- ThatOtherPerson has quit (Quit: Leaving).
15:54:41 -!- nooga has joined.
16:05:13 -!- LVCIFER has left ("SVVM CVIQVE").
16:15:08 -!- augur_ has quit (Remote host closed the connection).
16:17:11 -!- nooga_ has joined.
16:18:38 -!- nooga has quit (Ping timeout: 255 seconds).
16:31:47 -!- nooga_ has quit (Ping timeout: 256 seconds).
16:32:03 -!- Taneb has joined.
16:32:33 <Taneb> elliott: I had a weird dream
16:32:34 <lambdabot> Taneb: You have 2 new messages. '/msg lambdabot @messages' to read them.
16:33:10 <Taneb> I think I was reading a programming tutorial
16:33:14 <Taneb> There was an analogy
16:33:32 <Taneb> It involved a French newspaper, the Chinese Communist Party, and a birdcage
16:34:05 <boily> I smell the inception of a new monad tutorial here.
16:37:52 -!- Phantom_Hoover has joined.
16:38:03 -!- Phantom_Hoover has quit (Client Quit).
16:38:19 -!- Phantom_Hoover has joined.
16:38:21 -!- conehead has joined.
16:38:56 <Taneb> elliott: I don't think so
16:39:24 <kmc> Phantom_Hoover: nice word
16:39:50 <elliott> would be a good band name™
16:39:58 -!- ThatOtherPerson has joined.
16:40:18 <Taneb> The chinese graphics problem looks worryingly fixed
16:40:19 <Taneb> More info coming soon
16:40:19 <Taneb> *graphics card problem
16:40:26 <Taneb> ThatOtherPerson: hi
16:40:35 -!- glogbackup has joined.
16:40:50 <Taneb> Chinese graphics card problem is NOT FIXED :(
16:41:55 <Phantom_Hoover> is the chinese graphics card problem when you try to work out the version of someone's graphics card by the remainders of its version number
16:42:04 <tswett> elliott: what's this about pressing s?
16:44:23 -!- Phantom__Hoover has joined.
16:47:02 -!- Phantom_Hoover has quit (Ping timeout: 252 seconds).
16:52:11 -!- Bike_ has joined.
16:53:28 -!- ais523 has joined.
16:56:07 -!- augur has joined.
16:57:09 -!- Phantom__Hoover has quit (Ping timeout: 256 seconds).
16:57:30 -!- nooga has joined.
16:58:08 -!- Bike_ has changed nick to Bike.
17:00:55 -!- Bike_ has joined.
17:02:44 -!- Bike has quit (Ping timeout: 255 seconds).
17:05:01 -!- nooga has quit (Ping timeout: 245 seconds).
17:27:46 -!- sebbu2 has changed nick to sebbu.
17:29:24 -!- carado has joined.
17:35:13 -!- NihilistDandy has joined.
17:36:11 <NihilistDandy> Does anyone know how Wolfram's numbering scheme for Turing machines works?
17:36:34 <NihilistDandy> I'm not sure how it relates to the specification of the machine other than "this is just the number for this one"
17:37:21 <Bike_> what, like the state/color thing, or...?
17:37:25 -!- Bike_ has changed nick to Bike.
17:39:08 <NihilistDandy> So, for instance, Wolfram's (2,3) "UTM" is given here http://www.wolframscience.com/prizes/tm23/technicaldetails.html
17:40:04 <NihilistDandy> Where (2,3) is (state, color) and 596440 is the particular machine out of the 2985984 possible
17:41:03 <Bike> well, i guess it encodes the transition function
17:41:57 <Bike> the docs for mathematica's TuringMachine seem a bit vague about it though
17:43:03 <NihilistDandy> Yeah, that's what I thought. I gathered the same about it encoding the transition function, I just have no idea if there's a way to find them myself.
17:46:31 <Bike> the actual TuringMachine function can just take a transition graph too, so
17:48:49 <NihilistDandy> I guess that's true. Maybe there's a function to work it out from the graph. I'll have to do more reading. I have to imagine that this stuff is published *somewhere*
17:49:20 <Bike> is it important?
17:50:03 <NihilistDandy> I'm playing around with a universal Turing machine implementation in Haskell and I wanted to get around constructing the transition functions by hand
17:50:18 <NihilistDandy> It seemed like convenient notation. Not particularly important, just a thing to do
17:50:54 <Bike> i think you'd probably be better off just specifying the transition rules with some data structure
17:52:30 <NihilistDandy> Unfortunate, because I already have it working so nicely with the numbers. Oh, well, guess I'll just have to do things a little differently.
17:53:27 <NihilistDandy> Oh, maybe it's http://en.wikipedia.org/wiki/Description_number
17:54:00 <Bike> you'd just be decoding it into the data structure anyway, probably
17:54:37 <NihilistDandy> I'll be back later. Class is starting. Thanks for bouncing ideas around :)
17:54:41 <Bike> as the article says the fact that you can encode machines like that is usually more important than actually doing so
17:54:50 -!- NihilistDandy has quit (Quit: Textual IRC Client: www.textualapp.com).
17:59:15 -!- mnoqy has joined.
18:00:13 -!- Taneb has quit (Ping timeout: 256 seconds).
18:04:03 -!- zzo38 has joined.
18:08:46 -!- epicmonkey has joined.
18:10:28 -!- nooga has joined.
18:12:46 -!- oerjan has joined.
18:14:35 -!- MindlessDrone has quit (Quit: WeeChat 0.4.0).
18:16:14 <oerjan> taneb has this nasty habit of being away when you want to tell him something.
18:17:00 -!- nooga has quit (Ping timeout: 256 seconds).
18:20:39 -!- Taneb has joined.
18:22:23 -!- Lumpio- has quit (Ping timeout: 255 seconds).
18:23:41 <oerjan> ThatOtherPerson: YOU CAN TRY AGAIN NOW
18:23:59 <Taneb> UTF-8 Appreciation Day?
18:24:28 <ThatOtherPerson> <oerjan> taneb has this nasty habit of being away when you want to tell him something.
18:24:43 <Taneb> I have a crap connection here
18:25:02 <Taneb> My computer can connect to wifi pretty well, but my laptop's... poorly maintained
18:25:10 <Taneb> And my computer has the whole Chinese Graphics Card proplem
18:27:22 -!- copumpkin has quit (Ping timeout: 252 seconds).
18:27:59 -!- copumpkin has joined.
18:33:31 <zzo38> UTF-8 has some good ideas, such as being compliant with principle of extended ASCII, and that the sorting is still ordered by code points, however it has disadvantages such as different lengths for different characters, and all the stupid stuff that Unicode has.
18:34:11 <zzo38> However, it could be used with any character set which is extending ASCII and up to 36-bit code point numbers.
18:34:39 -!- Lumpio- has joined.
18:36:06 <zzo38> Also, not all programs that use UTF-8 even follow the principle of extended ASCII.
18:37:59 <zzo38> Some programs can use any extended ASCII encodings and it won't care, such as TeX (which by default uses single-byte encoding, although they are not included in the standard fonts, but you can write macros to support UTF-8 and other extended ASCII encodings).
18:38:25 <oerjan> <Bike> http://en.wikipedia.org/wiki/Reverse_migration_(birds) how could anyone not like biology, i ask you <-- well clearly biology is killing birds, here
18:43:10 <zzo38> As another example, my program VGMCK is mostly ASCII, although comments can use any extended ASCII coding whatever you want, and GD3 tags are in UTF-8. (GD3 tags are actually UTF-16; the program will do the conversion.)
18:44:19 <zzo38> Are there programs that will allow trailing spaces only if encoded using UTF-8 overlong encodings?
18:44:26 <zzo38> (VGMCK is one of them.)
18:46:09 <zzo38> Do you know of any others? Did you write any such programs?
18:50:28 <Koen_> Taneb: ThatOtherPerson: so how's that deadline going?
18:51:03 <ThatOtherPerson> I have confidence that we'll have something finished by then
18:53:08 <zzo38> Because this program only reads UTF-8 for the purpose of converting into UTF-16, CESU-8 is automatically also supported (although it will also correctly convert astral characters encoded in proper UTF-8).
18:58:59 -!- augur has quit (Read error: Connection reset by peer).
18:59:22 -!- augur has joined.
18:59:38 -!- oerjan has quit (Ping timeout: 245 seconds).
19:00:00 -!- oerjan_ has joined.
19:00:15 -!- oerjan_ has changed nick to oerjan.
19:01:38 -!- Phantom__Hoover has joined.
19:10:11 -!- Phantom__Hoover has quit (Ping timeout: 272 seconds).
19:11:28 <Koen_> ThatOtherPerson: ohhhh I thought you were doing the compo
19:11:36 <Koen_> it ended last night I believe
19:12:15 <oerjan> <fizzie> http://sprunge.us/jAUF why isn't our channel hip and popular enough to get this sort of stuff? <-- i vaguely think we've had those?
19:13:14 <Koen_> ohhh that's the difference
19:17:07 <kmc> been learning about zero knowledge proofs
19:17:29 <elliott> kmc: that's the kind of proof you make right
19:17:51 <kmc> here's a demo of zkp for 3-colorings of a graph http://web.mit.edu/~ezyang/Public/graph/svg.html
19:18:32 <oerjan> <NihilistDandy> Where (2,3) is (state, color) and 596440 is the particular machine out of the 2985984 possible
19:19:46 <kmc> we agree on a graph. i want to prove to you that i know a 3-coloring of this graph. i publish my colorings for each edge, encrypted separately. you pick an edge and i decrypt it and demonstrate that the colors were different
19:20:06 -!- Gracenotes has joined.
19:20:07 <elliott> oh zero-knowledge proofs aren't actually 100% certain?
19:20:24 <elliott> i assumed they were and now they are at least 5x less cool
19:20:25 <oerjan> @tell NihilistDandy Knowing how Wolfram numbered his CA's, I expect 596440 should be interpreted as a length 2*3 base 2*3*2 number
19:20:26 <kmc> yeah, we iterate this game some number of times until you are convinced to your satisfaction
19:20:49 <kmc> and i can permute the colors each time, so there's no way for you to assemble a full coloring from my answers
19:21:10 <kmc> (also it's not really 'encrypt' you want but a digital committment scheme, however i think encryption is close enough to demonstrate why the algo works)
19:21:21 <elliott> should be called zero-knowledge evidence obvs
19:21:36 <Gracenotes> elliott: given that they about probability distributions of challenge->response transactions, it does necessitate a sampling from that distribution to prove things about them.
19:21:59 <elliott> Gracenotes: yes it makes sense of course
19:22:12 <kmc> elliott: apparently they use 'zero-knowledge argument' to mean a ZKP which can be faked by an unboundedly powerful cheating prover but not by a polynomial time one
19:22:18 <elliott> I am just wholly disappointed because I knew nothing about them but my assumptions made them sound impossible and cool!
19:22:28 <kmc> elliott: sorry 2 disappoint
19:23:22 <Gracenotes> and there is no encryption algorithm with a fixed-length key that's unbreakable, quite tragic.
19:23:48 <elliott> someone should get on that.
19:26:56 <tromp> very stuck on robozzle 4005 http://www.robozzle.com/puzzle.aspx?id=4005 ...
19:27:49 <zzo38> Are there such quantum algorithms though?
19:32:49 <Fiora> isn't quantum encryption basically about a secure way to agree on a one-time pad?
19:33:04 <Gracenotes> (as well, not as though they are usable by anyone who's more than a kelvin above absolute zero)
19:33:04 <Fiora> kind of a bit like diffie hellman I think, except immune to a mitm?
19:33:39 <elliott> just realised what a badass name hellman is
19:34:08 <zzo38> Fiora: I guess so.
19:34:24 <Gracenotes> there's also Hellmann, which is mayonnaise.
19:34:40 <kmc> Fiora: http://en.wikipedia.org/wiki/Quantum_key_distribution protests that it is only the most well known of the quantum cryptographic primitives
19:35:28 <Fiora> oooh, so there's others
19:35:34 <kmc> also there are practical quantum crypto systems in the real world
19:35:39 <kmc> 'Quantum encryption technology provided by the Swiss company Id Quantique was used in the Swiss canton (state) of Geneva to transmit ballot results to the capitol in the national election occurring on October 21, 2007'
19:35:48 <kmc> i don't know if they're above 1 K or not ;)
19:36:37 <Gracenotes> the best thing to do for distributed secure elections is mixnets
19:40:35 <kmc> apparently most MRI magnets are at 4.2 K (boiling point of helium)
19:40:50 <zzo38> Another thing that might be possible to use with quantum encryption might be teleportation?
19:41:35 <kmc> quantum crypto will be broken through implementation bugs and operational fuckups
19:41:38 <kmc> just like classical crypto
19:42:01 <kmc> i was pleased to see the wikipedia page already lists a number of these
19:44:54 -!- oerjan has quit (Quit: Good night).
19:46:26 <kmc> philosophical consequences of ZKP are slightly weird
19:47:06 <kmc> your belief that i have a 3-coloring is justified and true, but you can't convey it to a third party, so is it knowledge or not
19:47:42 <Taneb> Isn't ZKP that one that's equivalent to the axiom of choice
19:47:49 <Fiora> what was the definition of knowledge...
19:48:08 <Fiora> "I belive X to be true." "If X were not true, I would not believe X to be true." "X is true."
19:48:12 <elliott> kmc: you can convey it by proxying right
19:48:20 <Fiora> I think ZKP satisfies that definition?
19:48:23 <kmc> even if you show the third party a full transcript of our conversation, they could suspect that we colluded to pick random values that aren't really random
19:48:28 <Fiora> at least that was the one we learned in philosophy class
19:48:40 <kmc> your justification in your belief is rooted in your belief that your random choices were truly random
19:48:41 <elliott> kmc: proof that it isn't knowledge: if a zero knowledge proof gave knowledge, then contradiction; QED
19:48:42 <Fiora> you believe it to be true, it is in fact true because it's been proven to you, and if it were not true, it could not have been proven to you
19:48:49 <Fiora> so even though you can't pass the knowledge on, you still know it?
19:49:12 <Fiora> so I think at least by that definition it's still knowledge (?)
19:49:35 <kmc> i think it fits the philosophy 101 definitions of knowledge (which is all i know) but it lacks some property people expect of something called 'knowledge'
19:49:43 <kmc> properties*
19:49:47 <Fiora> http://en.wikipedia.org/wiki/Robert_Nozick#Epistemology this was the one we learned I think
19:51:04 <Taneb> I disagree with the second statement of Fiora's definition of ZKP
19:52:01 <Gracenotes> proving that something is a probabilistic proof is the easy part. you also have to prove it's zero-knowledge.
19:52:22 <kmc> i think Fiora was giving a definition of knowledge not a definition of ZKP
19:54:01 <zzo38> http://adl.sourceforge.net/ seems to have many similarities with Z-machine.
19:54:33 -!- Phantom_Hoover has joined.
19:55:27 <Fiora> I was trying to mentally match up the ZKP thing with the nozick knowledge definition
19:55:30 <Fiora> to try to follow kmc's logic
19:55:37 <Gracenotes> zzo38: there's a port of Zork in that which I've read through
19:56:16 <zzo38> Maybe it is even possible to compile into Z-machine format.
19:56:34 <Gracenotes> For a LISP it has a hell of a lot of syntax, which is especially annoying when you have something so dense.
19:56:47 <Gracenotes> then again, I shouldn't be talking, whatwith Haskell's combinator parties
19:57:26 <zzo38> Yes, it does have a lot of syntax that is not like LISP.
19:59:34 -!- Taneb has quit (Ping timeout: 256 seconds).
20:01:20 -!- Taneb has joined.
20:02:37 <Vorpal> I need a couple of new 120 mm case fans. Mine are not sounding so good any more. Anyone know a good quiet brand? Good quality but most of all quiet.
20:02:54 <Vorpal> Are there different mountings? Power connectors?
20:03:38 <olsner> usually each fan has its own mounting and power connector
20:05:35 -!- Phantom_Hoover has quit (Ping timeout: 252 seconds).
20:05:37 -!- augur has quit (Read error: Connection reset by peer).
20:05:47 <Vorpal> olsner, what about good quiet fans?
20:06:06 -!- augur has joined.
20:06:13 <olsner> I think they work like other fans, just better and quieter
20:06:27 <Vorpal> right, but any recommendations?
20:06:35 <Vorpal> elliott, you usually know this stuff, right?
20:10:26 <elliott> iirc uhh nexus and noctua are well-liked, if I remember the names
20:10:50 <Vorpal> elliott, Thanks a lot :)
20:13:02 -!- Phantom_Hoover has joined.
20:17:59 <Taneb> Well, my computer's doing the other problem again
20:18:53 <Vorpal> Taneb, what other problem?
20:19:22 <Taneb> Crashes just after Ubuntu's splashy thing
20:19:27 <Vorpal> yeah that sounds like a bit of a bummer
20:19:43 <Vorpal> Taneb, what if you disable splash and just display the console?
20:19:57 <Taneb> Can't be bothered to play with it right now
20:20:10 <Vorpal> Taneb, what are you using atm then if not your computer?
20:20:38 <Taneb> (how would I use my computer if it doesn't start up)
20:22:19 <Taneb> Phantom_Hoover: I've had worse luck with Windows than I've had with Ubuntu
20:22:36 <Taneb> In terms of my luck with OS's it goes (roughly)
20:23:01 <Taneb> Solaris, Gentoo, Debian, Arch, Windows, Ubuntu, Mac OSX
20:23:23 <Taneb> Oh, and Haiku between Windows and Ubuntu
20:23:32 <Taneb> You're reading it backwards
20:23:55 <Bike> maybe you should just use an abacus
20:23:57 <Taneb> I've had the best luck with OSX because someone else set it up and I didn't do anything with it
20:24:03 <Taneb> Bike: couldn't get wifi
20:24:08 <Taneb> Same problem as Haiku
20:29:41 -!- Phantom__Hoover has joined.
20:30:48 <Vorpal> elliott, it is such a pain trying to find the product from Swedish resellers. :/
20:31:07 <Vorpal> most of the fans I can only find different variants of
20:31:21 <Taneb> How can I force Ubuntu to not skip the grub menu
20:32:17 <ais523> Taneb: you can edit the grub config
20:33:31 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
20:33:41 <Taneb> It SOMETIMES skips it
20:36:49 -!- Phantom__Hoover has quit (Ping timeout: 268 seconds).
20:37:51 -!- nooodl_ has joined.
20:41:14 -!- nooodl has quit (Ping timeout: 258 seconds).
20:42:29 -!- augur has quit (Remote host closed the connection).
20:44:02 -!- augur has joined.
20:45:26 <Vorpal> Taneb, set a longer timeout?
20:47:27 <fizzie> If it's Ubuntu, /etc/default/grub has the settings it puts in grub.cfg regarding the menu.
20:47:44 <fizzie> Grub itself has that "hidden timeout" thing and many settings.
20:49:59 -!- copumpkin has quit (Ping timeout: 252 seconds).
20:50:37 -!- copumpkin has joined.
20:57:16 <kmc> Bike: http://cosmarxpolitan.tumblr.com/
21:00:17 -!- Phantom_Hoover has joined.
21:04:40 -!- Phantom_Hoover has quit (Ping timeout: 260 seconds).
21:08:12 -!- Taneb has quit (Ping timeout: 256 seconds).
21:10:42 * ThatOtherPerson vows to hunt down whoever is responsible for Taneb's poor connection and tell them how disappointed he is in them
21:11:24 <Koen_> he only gets a bad connection two days a year
21:12:34 -!- Taneb has joined.
21:16:59 -!- Phantom_Hoover has joined.
21:24:32 -!- augur has quit (Read error: Connection reset by peer).
21:24:53 -!- augur has joined.
21:31:23 -!- Phantom_Hoover has quit (Ping timeout: 255 seconds).
21:33:25 -!- Phantom_Hoover has joined.
21:36:42 -!- Taneb has quit (Ping timeout: 245 seconds).
21:37:36 -!- Taneb has joined.
21:39:21 -!- augur has quit (Read error: Connection reset by peer).
21:39:49 -!- augur has joined.
21:41:48 -!- augur has quit (Read error: Connection reset by peer).
21:42:18 -!- augur has joined.
21:47:34 -!- Taneb has quit (Ping timeout: 256 seconds).
21:53:18 -!- Taneb has joined.
22:03:53 -!- nooodl_ has quit (Ping timeout: 245 seconds).
22:05:42 -!- boily has quit (Quit: Poulet!).
22:16:39 <Taneb> elliott: I am increasingly worried that we are the same person
22:17:14 <elliott> Taneb: 7 is the sacred number of hexham
22:17:33 <Fiora> wouldn't... wouldn't it be 6?
22:17:33 <elliott> Bike: hexham talk. you wouldn't understand.
22:17:35 <Taneb> You'd think 6 would be the sacred number of Hexham
22:17:41 <Taneb> What with the hex and the ham
22:17:48 <elliott> yes that is a common misconception.
22:17:48 <kmc> 6 little piggies went to market
22:18:00 <elliott> those who truly understand the nature of hexham will find their minds clear of such nonsense
22:19:31 <Bike> imo fuck hexham
22:20:03 <kmc> more like sexham (sex with a ham)
22:20:16 <Taneb> Bike: you should move to Hexham
22:21:24 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
22:22:49 <Bike> because fuck it
22:24:16 <Taneb> We've got some churches and an old gaol
22:24:21 <Taneb> It's called The Old Gaol
22:24:47 -!- Phantom_Hoover has joined.
22:25:11 <kmc> it's what it says on the tin
22:25:14 <elliott> Bike: imo if you came to hexham we could have a hexham #esoteric meetup
22:26:04 <kmc> built 1330
22:26:14 <Bike> couldn't you have that anyway?
22:26:21 <elliott> yes but it wouldn't be the same without a bicycle.
22:26:34 <Bike> i'm not the only bicycle in the world
22:26:52 <elliott> you're the only one that matters... to me......
22:27:10 <Fiora> they just want my genderswap :|
22:28:24 <Phantom_Hoover> Bike, if you go to hexham you can be the village bicycle
22:28:44 <Bike> isn't it a town
22:29:31 <fizzie> Is "Fiora" the term for a ladies' bicycle, or did I misunderstand something?
22:29:46 <Fiora> ... sorry, it's just an old in-joke
22:29:57 <Bike> Fioras are actually pegasi iirc
22:30:17 <elliott> fizzie: you misunderstood THE VALUE OF SPEECH RECOGNITION
22:30:19 <fizzie> We are cleaning up the Bike storage in this building, it's got a load of old cruft in it.
22:30:39 <elliott> how mean. Bike isn't old cruft.
22:30:44 <Fiora> pegasi? I think I prefer unicorns
22:30:48 <Fiora> but alicorns are pretty nice too
22:31:10 <fizzie> "Uniikki unikorni olikin korni koni."
22:33:11 <elliott> Phantom_Hoover are you sure you want to be making fun of people's names
22:33:14 <elliott> have you forgotten your name
22:33:50 <Taneb> He's named after a ethereal vacuum cleaner
22:34:00 <Fiora> Bike: http://fioraaeterna.tumblr.com/tagged/gpony seeeeeeeeeeee
22:34:03 <elliott> the mmnmhmsdnhfmhgdfmghsmhmfgh one
22:34:10 <Bike> well i can't argue with tumblr
22:34:10 <elliott> god dammit Phantom_Hoover what was it again
22:38:15 <Bike> wasn't that an alien race in Star Control
22:38:51 <elliott> Phantom_Hoover: ok but seriously what is it
22:39:00 <elliott> i'm too lazy to grep the logs to find where fizzie revealed it
22:39:09 <Taneb> It was something normalish followed by keyboard mashing
22:39:14 <Taneb> harryfjwe[uiofgheapiugh or something
22:40:24 <elliott> Phantom_Hoover is also named after a search engine
22:43:22 -!- Phantom___Hoover has joined.
22:44:17 -!- Phantom_Hoover has quit (Ping timeout: 246 seconds).
22:46:44 -!- Phantom__Hoover has quit (Ping timeout: 252 seconds).
22:47:03 <fizzie> Adahn. No, that was from Torment.
22:48:08 <elliott> Phantom___Hoover: how can you actually bear to have that name
22:48:39 <zzo38> It doesn't seem so bad to me.
22:49:03 <elliott> Phantom___Hoover: also how do i know you're not lying
22:49:08 <Phantom___Hoover> when i meet people it's basically "oh what's your name" "domhnall" "donald?" "domhnall" "dougal" "domhnall" "donor?" etc. ad infinitum.
22:50:30 <zzo38> I have a talent agency form that someone else was filling up. It specifies the HST rate which must be used. Does this mean the contract is void if the HST rate changes? Anyways, there is no more HST at this time (there was no HST at the time the form was sent, either).
22:51:32 <Bike> huh i thought "domhnall" was welsh
22:51:44 <Bike> instead of... irish. ok
22:52:15 <zzo38> Under the skills list, it lists both "dj" and "disk jockey", and they misspelled "instrument" twice in two different ways; they also misspelled "secretary"; many things are missing.
22:52:38 <Phantom___Hoover> although i thing the whole domhnall-donald etymology is just goidelic or sth?
22:53:48 <zzo38> It also says "rocking climbing gear" under "props", as well as juggling props; if you have balls that you can juggle, but you don't know how to juggle, does that count? (Anyways, some people juggle knife, fire, etc, too; not only balls)
22:54:53 <fizzie> That's some rocking climbing gear you've got there.
23:00:00 -!- augur has quit (Read error: Connection reset by peer).
23:00:05 -!- Phantom___Hoover has quit (Ping timeout: 272 seconds).
23:00:32 -!- augur has joined.
23:00:50 -!- Phantom___Hoover has joined.
23:07:15 -!- nooga has joined.
23:07:40 <shachaf> mnoqy: can i have some "monochrom wisdom" plz
23:08:30 <shachaf> it's like mnoqy wisdom except instead of mnoqy it's monochrom
23:10:52 -!- epicmonkey has quit (Ping timeout: 256 seconds).
23:12:13 -!- Taneb has quit (Quit: Leaving).
23:12:52 -!- nooga has quit (Ping timeout: 256 seconds).
23:19:45 <Bike> are you resting peacefully
23:26:55 <Sgeo> Putting out fires is fun
23:27:09 <Sgeo> Although, I think, if I hadn't noticed and fixed it, someone else would have eventually
23:40:55 <Sgeo> It was not a literal fire.
23:41:49 <Sgeo> 'putting out fires' IS an expression that can be used for things that are not literally fires, right? I'm not just losing my grasp of the English language?
23:42:23 <Bike> usually you'd offer some kind of context
23:42:28 <Bike> so that we don't assume actual fires
23:42:33 <Phantom___Hoover> well see normally the thought process would be "well he would have done something that ridiculous → reëvaluate as humour → aha, that's what he meant"
23:43:14 <Sgeo> The build process was frozen during testing
23:43:23 <Sgeo> The build was going on for almost 30 minutes
23:43:24 <Phantom___Hoover> whereas in this case it went "well he wouldnt have something that ridiculous → would he → would he... → sgeo..."
23:44:17 <Sgeo> Before I aborted it and uploaded a patch, to tell the testing thing the thing it was choking on
23:44:30 <Sgeo> *to ignore the thing it was choking on
23:50:48 <mnoqy> well aiui he saved the day so ya
23:51:03 <mnoqy> do you get a party for yr heroic deed
23:51:39 <Sgeo> Phantom___Hoover, there were no tests for the code, it's third party code. Also, I tried to make sure someone else knows about it to investigate
23:54:17 <Sgeo> I do want to know why the tests broke
23:54:26 <Sgeo> Erm, the entire testing framework broke