00:00:28 <fizzie> (IIRC, Fedora includes it in their copy of bluez by default, in fact.)
00:01:02 <fizzie> I was about to mention that the Sixaxis Controller app does disable the normal Bluetooth as another negative thing about it, though.
00:12:54 -!- friesk has quit (Remote host closed the connection).
00:16:20 -!- WeThePeople has quit (Quit: Leaving).
00:37:32 <shachaf> monqy: do you know what limits are
00:37:47 <monqy> didn't you ask me that yesterday
00:38:18 <shachaf> also you're basically elliott right?
00:38:38 <shachaf> and elliott knows what limits are, which he didn't know yesterday
00:39:29 <shachaf> monqy: i'm trying to figure out what pullbacks are in a poset category?
00:39:50 <shachaf> i'm pretty sure they exist if it's a lattice
00:40:03 <shachaf> or maybe even just a semilattice....
00:40:22 <monqy> if i say ok enough will you figure it out? if i silently replace myself with eliza will you notice
00:49:09 -!- Taneb has joined.
00:49:28 <Taneb> Okay, in the end I did go to that UV rave
00:50:38 <Taneb> `quote dream about
00:50:47 <Taneb> `quote dreamt about
00:50:56 <HackEgo> 154) <zzo38> catseye: Please wake up. Not recorded for this timezone. The big spider is not your dream \ 239) <nddrylliog> back to legal tender, that expression really makes me daydream. Like, there'd be black-market tender. Out-of-town hug shops where people exchange tenderness you've NEVER SEEN BEFORE. \ 249) <Phantom__Hoover> Gregor, yeah, but P
00:51:04 <elliott> Phantom_Hoover: its like brainfuck
00:51:51 <Phantom_Hoover> Taneb, was ian bell at the uv rave (this is all i know about uv raves)
00:52:05 <Taneb> Is that close enough
00:52:57 <Taneb> A sink was destroyed
00:54:35 <Taneb> Is there anything that man cannot do
00:54:55 <zzo38> Well, some things are impossible.
00:55:20 <Taneb> I've drank alcohol. Everything is possible.
00:55:57 <zzo38> Is it possible to have a universe where thirteen is not a prime number?
00:56:31 <Taneb> Yes. Simplest way is to redefine 13 to equal, for example, what we call six
00:58:06 <Taneb> Then your universe sucks
00:59:32 <Phantom_Hoover> indeed i would strengthen that claim and propose that it sucks a big bag of cocks
01:02:00 <zzo38> Can there be the universe that time goes only sideways?
01:02:13 <Sgeo_> o.O kmc had a hand in this.... http://blog.ezyang.com/2011/11/how-to-read-haskell/
01:02:21 <Sgeo_> zzo38, have you read the Orthogonal series?
01:03:01 <kmc> yeah i offered feedback
01:03:17 <Sgeo_> zzo38, time acts geometrically in the universe of that series, so instead of x^2+y^2+z^2-t^2 = umimnotsureithinkaconstnat, it's x^2+y^2+z^2+t^2
01:04:17 <Phantom_Hoover> also er that pretty much automatically implies causality violation
01:04:50 <Phantom_Hoover> (s is proper time, i.e. time as observed by whatever's moving)
01:05:25 <Sgeo_> Well, there's no speed limit, some characters are trying to reach infinite velocity as determined by their home world
01:05:52 <zzo38> I don't think causality is a fundamental rule of the universe, though.
01:05:58 <Phantom_Hoover> i'm... pretty sure what you're describing is just s^2 = t^2
01:06:25 <Bike> sounds like really dorky sci-fi, anyway.
01:06:28 <Sgeo_> Time in the Orthogonal series's universe is not fundamentally a distinct direction from space
01:06:43 <Sgeo_> Bike, the author has all these... charts. I am not joking
01:06:56 <Sgeo_> Stops the narrative to have characters either learn or explain the science they're discovering
01:07:07 <Bike> And you read this of your own volition.
01:07:22 <Sgeo_> I read the first book of the series, going to buy the second soon
01:07:31 <Sgeo_> (It's a trilogy, but third book isn't out yet)
01:07:47 <Sgeo_> I have to admit to skimming over the science stuff a bit
01:07:50 <elliott> Bike: dont tell me you dont enjoy super dorky super hard scifi
01:07:54 <elliott> everyone does whether they admit it or not
01:08:04 <Sgeo_> I wish I understood it at a glance
01:08:04 <elliott> Phantom_Hoover: that was more a general statement
01:08:08 <Bike> Well, I do, but it doesn't usually go as far as charts.
01:08:09 <elliott> i in fact barely looked at what sgeo said
01:08:17 <Bike> ...okay, so I did like Expedition.
01:08:26 <Sgeo_> Phantom_Hoover, uh. lazy?
01:08:40 <Sgeo_> http://gregegan.customer.netspace.net.au/ORTHOGONAL/00/PM.html
01:08:51 <Phantom_Hoover> "here are a bunch of cool concepts, i will make the most cursory attempt to attach them to the narrative"
01:09:04 <elliott> going out on a limb and saying it's probably not lazy
01:09:06 <Bike> Haha of course it's Egan
01:09:24 <elliott> maybe i should actually read anything hes written beyond like the two short stories i have
01:09:42 <Sgeo_> elliott, I read The Clockwork Rocket
01:09:44 <Bike> I liked Schild's Ladder, though it probably warped my psyche some since I was like twelve.
01:10:31 -!- hagb4rd has joined.
01:10:53 <elliott> "Interstellar voyages taking longer for the travellers than for everyone else might just sound annoying"
01:11:09 <elliott> "dammit, I can't use this universe. it's just... annoying"
01:11:34 <Bike> http://tallcomics.com/?id=17 anyway this is my favorite sci-fi
01:11:41 <Sgeo_> Mind if I ROT13 some spoilers?
01:12:25 <shachaf> ROT26 it to be on the safe side.
01:12:48 <elliott> dont rot13 stuff in #esoteric
01:12:51 <elliott> unless you want me to read it
01:13:28 <Sgeo_> elliott, how much do you want to avoid spoilers for The Clockwork Rocket?
01:13:39 <zzo38> ROT(infinity) it if you know how.
01:13:53 <elliott> my caring is directly related to how good it is
01:14:00 <HackEgo> 263) <zzo38> Is anyone in here who knows cricket rules and has experience? <Slereah> What if I told you the baseball rules in a british accent? \ 328) <zzo38> Finally I found the wand of electric lightning now we can destroy any large object if it needs to be destroyed and is required to use a such a wand for that purpose. \ 439) <zzo38> elliott_:
01:14:02 <Sgeo_> elliott, I kind of liked it, but I'm a bad judge
01:14:26 <kmc> zzo38_ebooks
01:14:28 <shachaf> zzo38: Did you ever use the want of electric lightning?
01:14:41 * Sgeo_ decides not to paste the spoiler
01:14:44 <shachaf> kmc: I would follow that on Twitter!
01:15:12 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.3415
01:16:21 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.21081
01:16:54 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.24833
01:17:19 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.2696
01:17:21 <Bike> `pastelogs `pastelogs
01:17:22 <zzo38> Can you make another kind of pastelogs command which allow to specify start date/time?
01:17:33 <shachaf> zzo38: Are you on the moon?
01:17:43 <zzo38> No, I am still on the Earth.
01:18:31 <HackEgo> 467) <itidus20> well, you have bested me <zzo38> itidus20: Yes.
01:18:51 <Sgeo_> Oh god my first conversation in here was about stuff related to PSOX?
01:19:13 <Sgeo_> elliott, go tell Bike about PSOX
01:19:26 <Bike> Oh, I thought you said POSIX.
01:19:31 <Bike> Also "beerkills"
01:19:34 <HackEgo> qdbformat is: <nick> message; * nick action; two spaces between messages; all elisions marked with [...] other than irrelevant intervening messages; for messages separated by elision, one space on each side, not two
01:19:36 <Phantom_Hoover> istr that the first time i came in here i left about half an hour later because elliott yelled at me for spamming up the channel with HackEgo
01:19:48 <monqy> my first conversation was being asked what my business was in here. thanks elliott
01:19:58 <elliott> monqy: i still don't know what your business here is
01:20:03 <Bike> The first time I came in here was the first time that I came in here.
01:20:06 <Sgeo_> Bike, I was a very anti-alcohol person
01:20:09 <shachaf> wait monqy is here for business?
01:20:18 <Taneb> 677 is a CAD quote, I'm sure
01:20:19 <shachaf> i always thought it was pleasure........
01:20:38 <Sgeo_> Taneb, I'm sure it's a quote from something
01:20:40 <shachaf> monqy: what is your business here
01:21:01 <Taneb> Anyway, I need sleep
01:21:34 <HackEgo> 860) <kmc> i would subscribe to @zzo38_ebooks <zzo38> kmc: I have no ebooks which can subscribe to
01:21:36 <HackEgo> 677) <zzo38> When you die in Canada, you die in real life.
01:21:59 <elliott> it stays because zzo38 said it
01:22:08 <shachaf> zzo38 said a lot of things
01:22:14 <Sgeo_> 2005-10-31.txt:23:58:29: <GregorR> And now, for an interpretive piece I like to call "Sgep"
01:22:14 <Sgeo_> 2005-10-31.txt:23:58:33: <GregorR> <Sgep> Hi!
01:22:14 <Sgeo_> 2005-10-31.txt:23:58:39: <GregorR> <Sgep> brb
01:22:14 <Sgeo_> 2005-10-31.txt:23:58:47: <GregorR> <Sgep> Back.
01:22:14 <Sgeo_> 2005-10-31.txt:23:58:57: <GregorR> <Sgep> Got to go, see you later.
01:22:16 <shachaf> we can't addquote every single thing zzo38 says
01:22:17 <Sgeo_> 2005-10-31.txt:23:59:02: <GregorR> * Sgep has left freenode ()
01:22:25 <zzo38> Are you sure it is xkcd? I read it elsewhere
01:22:43 <elliott> maybe that was a reference but if it is i don't know what it was referencing
01:22:48 <zzo38> shachaf: Correct; that would make the quotation file too large; there are logs for that.
01:23:21 <elliott> well i guess its a matrix reference
01:23:27 <zzo38> However, I don't want deleting from the quotation file (whether it is mine or elsewise) (other than ones just written in error and should be deleted due to that); one thing it mess up the numbering but there are other problem too
01:23:27 <elliott> but the matrix isn't canada (afaik)
01:23:31 <Taneb> In CAD, Ethan did go to Canada at one point
01:23:40 <monqy> did he die in canada
01:23:46 <elliott> taneb why do you know things about cad
01:23:47 <Taneb> I do not believe so
01:23:56 <Taneb> elliott, I have a dark and shady past
01:24:13 <Phantom_Hoover> you can learn quite a lot about cad just from reading people mocking it
01:24:25 <monqy> i know things about cad for the reason Phantom_Hoover notes
01:24:25 <Taneb> I read quite a bit of CAD
01:24:38 <monqy> i've also watched the cad cartoons because they're awful
01:24:39 <Taneb> Up until the robot guy went on a quest to find something
01:24:41 <Bike> Will I learn things from watching people talk about learning about it from reading mockeries?
01:24:45 <elliott> tanebs knowledged seemed a bit `first hand'
01:24:57 <elliott> Phantom_Hoover: did you ever make a bf derivative
01:25:07 <monqy> im happy to have never read cad
01:25:08 <elliott> thats not strictly true i started programming by copying basic programs from a book into a computer (badly)
01:25:18 <elliott> then i forgot about programming until i was 8
01:25:29 <zzo38> You add quote if, that is what you like it particularly enough to add. It is no point adding everything which result in being the same as the full logs; read the full log instead if you want!
01:25:42 <elliott> life tip: don't start programming in php at 8 yrs of age
01:26:02 <Taneb> Start in Visual Basic at 12
01:26:20 <Bike> I wonder if that's actually trademarked...
01:26:50 <elliott> i found visual basic too complicated
01:27:03 <zzo38> Actually I do not think such things really are such problem. Sure it may not be very good programming languages, but, you can write a good program, regardless what one, and then learn other programming languages maybe is you liked better
01:27:34 <Bike> "It Worked For Me: Life Lessons from Colin Powell"
01:28:03 <Sgeo_> Can I continue to blame my inability to write GUIs on having learned Visual Basic when I was young?
01:28:08 <elliott> monqy: when did you / what language did you start programming with
01:28:30 <Phantom_Hoover> incidentally, apparently colin powell's name is pronounced 'cohlan'
01:28:40 <Taneb> shachaf, it's an acronym
01:28:58 <shachaf> does it stand for PHP: hi Phantom_Hoover
01:29:04 <zzo38> Phantom_Hoover: He doesn't want to be confused by other people having the same name, maybe
01:30:03 <Sgeo_> I wonder if my life would have taken a different path if I never shattered that ActiveX for Dummies CD when I was a kid...
01:30:19 <monqy> elliott: the details are kind of muddy but sometime around 8 with some combination of logo, calculator basic, and flowcharts (which were used to program the microcontrollers)
01:30:48 <Sgeo_> Would I have grown up to be a Windows lover, relied on IDEs for those... ID things that ActiveX uses?
01:30:59 <monqy> and also some other things like uhhhh
01:31:01 <Sgeo_> Would I be a better programmer because of more practice, even with horrible tools?
01:31:14 <Bike> this is a really boring alternate history i must say
01:31:49 -!- Taneb has quit (Quit: Goodnight).
01:32:09 <kmc> i don't like the IDEs = horrible tools implication
01:32:20 <kmc> i'm sure whatever was included with ActiveX for Dummies was horrible though
01:32:29 <kmc> since ActiveX is
01:32:43 <Sgeo_> By horrible tools I meant VBScript. And reliance on a language+environment where I needed the IDE in order to give me opaque blobs of information that need to go in the code
01:33:10 <Sgeo_> (The CLSID thing I think)
01:38:21 -!- augur has quit (Remote host closed the connection).
01:39:34 <shachaf> whoa, dude, this web page uses VRML
01:41:04 <kmc> that's a blast from the past
01:41:44 <shachaf> http://www.j-paine.org/scratch/wc_2013_2_17_1_37_5_368.html
01:41:49 <shachaf> I guess it doesn't actually "use" it.
01:41:53 <shachaf> But it generates VRML files?
01:42:39 <kmc> i remember a site that had a VRML model of a wienerschnitzel
01:42:54 <kmc> and when you clicked it would disappear and a voice would say "mmmm yummy schnitzel"
01:43:13 <copumpkin> kmc: what's the etymology of miuaf btw?
01:44:23 -!- DHeadshot has joined.
01:45:20 <kmc> initials of the name of my blog
01:45:23 <kmc> i'm bad at naming things
01:45:42 <kmc> copumpkin retweeted me, now a bunch of iPhone jailbreaking enthusiasts can see my halfbaked remarks :)
01:46:15 <copumpkin> need to get one of the real jailbreakers to retweet you
01:46:24 <elliott> them iphone jailbreaking people must be baffled by their celebrities' quirks
01:46:28 <elliott> "that haskell guy" "that nomic guy"
01:46:30 <copumpkin> a couple of them have more than 400k followers
01:46:34 <shachaf> copumpkin retweeted me once
01:46:45 <shachaf> I didn't even tweet anything. :-(
01:46:51 <copumpkin> comex is small fish compared to pod2g and MuscleNerd
01:47:57 <kmc> what's saurik's quirk
01:50:10 <elliott> unfortunately we have exhausted the space of iphone jailbreaky people i know
01:50:18 <elliott> so you will just have to wonder
01:51:38 <shachaf> I haven't heard of either of them.
01:51:42 <shachaf> I've heard of copumpkin, though!
01:52:12 <elliott> copumpkin is the celebrity of our hearts & that's all that matters
01:53:31 <shachaf> copumpkin: So Haskell doesn't have pullbacks?
01:55:05 <Sgeo_> Someone mentioned VRML and I wasn't paying attention?
01:55:08 <copumpkin> yeah, haskell doesn't have pullbacks and agda does
01:55:42 <shachaf> Because you need a type that expresses things like f(a)=g(b)
01:55:52 <Sgeo_> What's a pullback?
01:55:54 -!- nooga has quit (Ping timeout: 264 seconds).
01:55:55 <copumpkin> you can sort of do it at the type level
01:56:18 <copumpkin> Sgeo_: it's sort of like a fancier product
01:56:21 <shachaf> elliott: You're right, we should abandon Haskell for all the category things.
01:56:25 <shachaf> It doesn't even have pullbacks. :-(
01:56:37 <elliott> Sgeo_: it's the limit of the diagram X -> Z <- Y
01:56:49 <shachaf> I guess it doesn't have equalizers either.
01:56:53 <Sgeo_> ..that means nothing to me :(
01:57:06 <Sgeo_> To me product means tuple
01:57:29 <shachaf> Sgeo_: Don't worry, elliott is just trying to be intimidating.
01:57:40 <shachaf> Sgeo_: 24 hours ago he didn't know what limits were either!
01:57:56 <copumpkin> I think those are the most "down-to-earth" notion of pullbacks in common usage
01:58:02 <Sgeo_> shachaf, I know what a limit is, just not what a diagram is.
01:58:13 <shachaf> Sgeo_: It's a different sort of limit.
01:58:18 <elliott> this is probably a different kind of limit to the kind you're thinking of
01:58:26 <copumpkin> this is not the limit you are looking for
01:58:32 <shachaf> Hopefully http://flockdraw.com/upload/8kr07f6lb00s44k80c4.png will make things clearer!
01:59:09 <elliott> shachaf: imo show the other one that has my drawing talent too
01:59:24 <shachaf> copumpkin: Me, except for the black part.
01:59:48 <elliott> shachaf: http://flockdraw.com/upload/11rq0ta2ee1c8cog8c.png that one
02:00:10 <shachaf> That's a product expressed as a limit.
02:00:37 <copumpkin> shachaf: there, 24k iphone fanboys saw your artwork
02:01:02 <elliott> copumpkin: wow I feel discriminated against
02:01:16 <elliott> my work is being unfairly overlooked!!
02:01:26 <copumpkin> you aren't on twitter! or are you?
02:01:35 <elliott> but i do draw bad diagrams with shachaf
02:01:48 <HackEgo> Taneb atriq Ngevd Fiora nortti Sgeo ThatOtherPerson alot
02:01:49 <elliott> it's ok he gets all the credit
02:02:00 <elliott> that is but the place I am in
02:02:15 <elliott> shachaf: I like how you can't tell the V is a V
02:02:29 <shachaf> elliott: As opposed to a U?
02:02:31 <Sgeo_> False alarm. Or maybe Hussie changed something
02:02:38 <shachaf> It's actually supposed to be U
02:03:20 <shachaf> Darin Morrison (@darinmorrison) is now following you on Twitter!
02:03:48 <elliott> I hope you're paying copumpkin for these valuable followers
02:04:01 <shachaf> copumpkin should pay me for making my email notification thing beep.
02:06:18 <shachaf> Anyway the explanation worked.
02:09:48 <HackEgo> Taneb atriq Ngevd Fiora nortti Sgeo ThatOtherPerson alot
02:13:37 <Bike> Becaue the `list is mysterious and fickle.
02:16:57 <monqy> the s is for stupid
02:17:01 <Sgeo_> Phantom_Hoover, it is
02:17:14 <HackEgo> cuttlefish boily elliott Taneb HackEgo shachaf Sgeo monqy pikhq Sgeo_ tswett
02:17:27 <HackEgo> cuttlefish boily elliott Taneb HackEgo shachaf Sgeo monqy pikhq Sgeo_ tswett Phantom_Hoover
02:17:28 <Bike> You know not what you oh there it goes then.
02:18:00 <Sgeo_> Maybe someone can do something that blocks shachaf from being added?
02:18:16 <elliott> maybe shachaf can stop doing rm bin/list all the time
02:18:20 <monqy> why should shachaf get special treatment
02:18:21 <Bike> Or shachaf could just not run it?
02:18:39 <Bike> It was rather rude for you to be put on the list then.
02:18:45 <monqy> what if i want to be special too
02:18:54 <shachaf> monqy: you're already maximally special
02:18:54 <Bike> So just find that person and eat them.
02:18:57 <Bike> Problem solved.
02:18:57 <Sgeo_> `run sed -i s/ shachaf// bin/list
02:18:59 <HackEgo> sed: -e expression #1, char 2: unterminated `s' command
02:19:07 <shachaf> monqy: btw did that diagram clear things up
02:19:08 <elliott> shachaf is on the list according to the bin/list program
02:19:10 <Sgeo_> I don't actually know how to sed
02:19:14 <monqy> shachaf: clear what up
02:19:24 <shachaf> Sgeo_: I think this is a matter of not knowing how to bash.
02:19:42 <HackEgo> cuttlefish boily elliott Taneb HackEgo Sgeo monqy pikhq Sgeo_ tswett Phantom_Hoover
02:19:52 <monqy> um im upset shachaf isnt on the list
02:19:59 <monqy> if i whine enough can shachaf get put back on it
02:20:10 <shachaf> you're "too mature to whine"
02:20:16 <Bike> You're hypothetically very rude, Monqy.
02:20:18 <HackEgo> abort: unknown revision '`revert'!
02:20:32 <HackEgo> cuttlefish boily elliott Taneb HackEgo Sgeo monqy pikhq Sgeo_ tswett Phantom_Hoover
02:20:58 <kmc> maybe presentation slides need to exist in two versions
02:21:06 <kmc> the version with all the internet meme shit to keep the audience's attention
02:21:12 <kmc> and the version you post online with just the content
02:21:23 <monqy> how about presentation slides just dont contain the internet meme shit at all
02:21:24 <Bike> Oh I thought you were going to say and the version with cuttlefish.
02:21:30 <monqy> i hate it when that stuff happens
02:21:38 <kmc> i know i can't pay attention for more than 3 seconds without an advice animal of some kind
02:21:51 <Bike> Is there an advice cuttlefish?
02:22:05 <HackEgo> #!/bin/sh \ oldpwd=`pwd`; cd /var/irclogs/_esoteric; name=$(cat $(ls ????-??-??.txt | tail -1) | tail -1 | sed "s/[^<]*<//; s/>.*//; s/.*\* //; s/ .*//"); cd $oldpwd; fgrep -q "$name" bin/list || echo -n "$name " >> bin/list; echo cuttlefish boily elliott Taneb HackEgo shachaf Sgeo monqy pikhq Sgeo_ tswett Phantom_Hoover
02:22:10 <shachaf> elliott: Please stop being annoying.
02:22:18 <monqy> imo you stop first
02:22:31 <elliott> maybe if you start doing your edits with less than a sledgehammer
02:22:41 <Bike> Is it possible that this channel exists for people to irritate each other over pointless things for no reason?
02:22:42 <Phantom_Hoover> would you people either stop bickering or fill me in on the juicy details
02:22:54 <elliott> kind of ridiculous that you just wipe out files because of a tantrum
02:23:14 <monqy> Phantom_Hoover: there's a running joke where shachaf whines about this
02:23:16 <Sgeo_> monqy did as a joke
02:23:22 <Bike> http://www.quickmeme.com/meme/3pbhkr/ I suspect this is dumb.
02:23:27 * Sgeo_ sides with shachaf on this
02:23:46 <Bike> There are no sides. We are he as she is he and we are all together.
02:24:02 <Bike> And when I say "she" I mean noone.
02:24:53 <Bike> None of the cuttlefish I'm seeing on this meme site (why is there such a thing?) are very funny, but they are cute.
02:26:39 <Bike> Because they're cute. And at the head of the list.
02:27:37 <Phantom_Hoover> i feel like humanity should set cuttlefish up as our inheritors somehow
02:27:56 <Phantom_Hoover> with, idk, strategically-placed caches of cuttlefish-applicable technology
02:28:53 <Bike> They'd probably have to have significantly evolved morphology to use technology...
02:29:00 <Sgeo_> Hey, are Haskell records ever going to be fixed?
02:29:13 <coppro> no. you can only fix functions
02:30:54 <elliott> Sgeo_: http://lens.github.com/
02:31:20 * Sgeo_ goes to watch video
02:32:00 <Bike> I mean, we'd have to guess at what morphology they'd have in order to operate anything we gave them.
02:32:07 <kmc> Bike: i lolled
02:32:19 <Bike> These are important considerations kmc.
02:32:28 <Sgeo_> Maybe I'll read the tutorial instead
02:32:51 <Bike> Building something that could continue being usable milllions of years from now is also probably near impossible.
02:33:14 <Phantom_Hoover> i feel that designing technology for present cuttlefish is a task only the blinkeredly myopic would call 'impossible'
02:34:04 <elliott> can't we just evolve into cuttlefish
02:34:10 <Bike> That's a good word.
02:34:34 <Phantom_Hoover> elliott, no! we'd have to displace the cuttlefish then
02:34:48 <Phantom_Hoover> that wouldn't be fair unless we let them have something in return
02:35:53 <zzo38> Phantom_Hoover: I wouldn't call it 'impossible' but it might be difficult.
02:39:12 <Bike> All I'm saying is cuttlefish are notoriously handfree.
02:39:43 <zzo38> People sometimes use handsfree telephones too, though.
02:39:59 <Sgeo_> Why does everyone like cuttlefish so much?
02:40:29 <Phantom_Hoover> Bike, yes, an important challenge is overcoming handocentric human design
02:40:29 <Bike> They're cute, intelligent, and have interesting camouflage behavior, probably.
02:40:47 <Bike> Another good word.
02:41:04 <Bike> "manocentric" says my internal ignorant pedant, though.
02:41:26 <elliott> "manocentric" just sounds like a really silly term for sexism
02:42:05 <Bike> That, along with "centric" and "mano-" probably being from different languages, is why my internal pedant is such a dummy.
02:42:40 <Bike> They can't perceive color, but do perceive polarization. That might have interesting implications.
02:43:40 <kmc> cuttlefish can't perceive color?
02:43:41 <quintopia> you can do a wire-crossing in the plane with three xors right? is there a small wire-crossing circuit that does not use xor? (by small i mean less than 10 gates)
02:43:59 <kmc> but their camoflauge is color-sensitive isn't it
02:44:19 <kmc> not sure i know how we're deciding what they perceive
02:44:40 <zzo38> It may be done using glasses which have a picture for each eye you can then make them not only separate colors but also separate polarization, and so on, just so you know
02:44:49 <zzo38> Phantom_Hoover: By instinct, I guess?
02:44:58 <kmc> 'The statement about color blindness is due to a study in 'Sepia officinalis' that found no difference in camouflage in response to differences in color (see this paper). I think there are a number of holes that could be poked in the conclusions of that study, and it was only conducted on one species, so I think the statement that they are colorblind is rather premature.'
02:45:26 <kmc> anyway they are smart enough to learn simple directional cues from their environment
02:45:30 <Phantom_Hoover> http://hermes.mbl.edu/mrc/hanlon/pdfs/mathger_et_al_visres_2006.pdf
02:45:40 <kmc> so you cuold test whether they can perceive color in a behavioral sense
02:46:04 <zzo38> quintopia: Help with what?
02:46:22 <Bike> I forget, can they be clasically conditioned? The usual protocol for pigeons involves colored lights
02:46:28 <Phantom_Hoover> so apparently they don't know what colour they're camoflaging themselves as
02:46:43 <kmc> that says they have only one type of light receptor
02:46:47 <kmc> so i guess that's pretty definitive
02:47:58 <kmc> how carefully did they establish that the two patterns were lightness equivalent to a cuttlefish
02:48:10 <zzo38> Now you have "fi" ligatures
02:48:18 <kmc> fuck i've been thinking about human colorspaces all day and i can't even begin to fit cuttlefish into that picture
02:48:21 <zzo38> quintopia: I don't know the answer
02:49:15 <Phantom_Hoover> kmc, well they demonstrated that it used the same patterns for blue and yellow squares, uniform blue and uniform yellow
02:49:32 <Bike> kmc: Do we even know much about nonhuman colorspaces? Like in something easier than cuttlefish, like vertebrates.
02:49:55 <kmc> i don't know
02:50:47 <Phantom_Hoover> well i'm assuming someone's calculated colour spaces for the colourblind at some point so you can probably work it out for some mammals
02:51:09 <kmc> wikipedia says dogs have dichromatic color vision similar to deuteranopia in humans
02:52:27 <kmc> "Dogs have a temporal resolution of between 60 and 70 Hz, which explains why many dogs struggle to watch television"
02:53:02 <Phantom_Hoover> dogs struggling to watch television is an immensely entertaining concept for some reason
02:53:08 <Bike> yeah apparently a lot of animal experiments involve specially calibrated TVs
02:53:14 <kmc> i think this is my new favorite sentence on wikipedia
02:53:48 <Bike> What was the previous one?
02:54:04 <kmc> "Hagfish are usually not eaten owing to their repugnant looks, as well as their viscosity and unpleasant habits."
02:54:36 <Bike> pretty That's a sentence good.
02:54:37 <Phantom_Hoover> what about "Fukutsuru died in 2005, but his frozen sperm lived on for peoples benefit;"
02:54:45 <Bike> That's a clause.
02:54:50 <kmc> that's pretty solid [pun intended]
02:57:33 <kmc> "Due to the lack of required fare, the Staten Island Railway system is popular among destitute vagrants and heroin addicts."
02:57:42 <kmc> this is only funny because it was the caption on a totally normal photo of a SIR train
02:58:03 <Bike> That seems kind of weird, what was the article about, SIR?
02:58:19 <kmc> it was the file description on http://en.wikipedia.org/wiki/File:SIR_448_at_Great_Kills_Station.jpg
02:58:26 <kmc> basically just vandalism though
02:58:37 <kmc> it's a true enough statement but not descriptive of the photo
02:58:56 <kmc> hagfish are one of those weird not-quite-vertebrates
02:59:13 <kmc> "They are the only known living animals that have a skull but not a vertebral column."
02:59:32 <kmc> also they do incredible slimy tricks
02:59:49 <kmc> https://www.youtube.com/watch?v=RrPvMMkQkk0
03:00:14 <Jafet> Are there known undead animals with skulls but not vertebral columns
03:00:29 <Phantom_Hoover> "The hagfish is kept alive and irritated by rattling its container with a stick, prompting it to produce slime in large quantities. This slime is used in a similar manner as egg whites in various forms of cookery in the region."
03:00:56 <kmc> produces a huge volume of slime, suffocating its enemies, then ties its body into a knot and pulls itself through the knot to clean the slime off itself
03:04:04 <Bike> Phantom_Hoover: Don't look up bile bears sometime.
03:05:11 <Phantom_Hoover> but bears are cute! so eating their bodily fluids isn't as weird
03:08:21 <Bike> Oh, I thought you were referring to the methods.
03:09:35 <Phantom_Hoover> i'm not especially outraged at people irritating a hagfish, no
03:12:54 -!- azaq23 has quit (Quit: Leaving.).
03:13:38 -!- Phantom_Hoover has quit (Remote host closed the connection).
03:25:41 -!- dessos has left.
03:25:47 -!- dessos has joined.
03:47:16 <hagb4rd> i have a special gift for you to make.. and i guess it is finally ready for it's first flight! an inertia-driven assembly of heartbreaking-space-pilot-adventure music i have collected over the years..please tell me if you feel sth is still missin in this great list..and enjoy hab4rds zauberschatzliste(tm)
03:47:19 <hagb4rd> --> http://www.youtube.com/user/zauberschatzkiste
03:48:48 <shachaf> So what's with the notation for a pullback?
03:55:19 <shachaf> Why doesn't it mention f and g?
03:55:49 <shachaf> Right, but f and g matter.
03:56:01 <shachaf> elliott was complaining about this yesterday.
03:56:38 <copumpkin> not everything needs to exist in the "type" though
03:57:03 <copumpkin> record Pullback (A B C : Set) : Set where field f : A -> C; g : B -> C
03:57:11 <kmc> nice, the UHDTV green primary is very close to a 532 nm green laser
03:57:18 <kmc> (ITU rec. 2020)
03:57:39 <kmc> resolution up to 7680 × 4320, framerate up to 120 fps
04:20:40 <zzo38> Do they need limits?
04:25:12 <kmc> well if you are designing equipment or software it's good to know what it will be expected to do
04:25:29 <kmc> consumers won't want to buy a "UHDTV" only to find out that all the stations broadcast in incompatible modes
04:26:41 <shachaf> kmc is an expert in ultra-high things
04:28:18 -!- augur_ has joined.
04:28:26 <zzo38> Well, that is true, that the equipment and station should specify what limit to use, but the protocol should not be limited in that way, and only limited in the cases of the two devices in this way that they should be *explicitly labeled* by what resolution/framerate is supported! I think that is better than just saying "UHDTV is up to" this and that.
04:30:01 -!- TeruFSX has joined.
04:35:03 <zzo38> Actually this is first time which I use __attribute__((constructor)) and in this case it is useful
04:35:46 <zzo38> Destructor does not seem as useful, but maybe in some case it is but I didn't know what it is.
04:36:33 <coppro> is hi monqy a running joke or something?
04:38:04 <zzo38> shachaf: What does what do, the __attribute__((constructor)) command in C?
04:38:09 <zzo38> Is that what you meant?
04:38:15 <shachaf> I meant ((destructor)), but sure.
04:38:22 <shachaf> By C I guess you mean GCCC.
04:38:51 <zzo38> Constructor means the function is called when the program started, destructor is done in the opposite order when the program stopped.
04:39:16 <kmc> i used it for a return-to-libdl attack once
04:39:16 <Bike> coppro: I think most of this channel is running jokes.
04:39:56 <zzo38> I am using __attribute__((constructor)) to allow all of the modules that are compiled in to be identified to the main program.
04:41:31 <zzo38> The function can be declared static but it still requires a name.
04:43:08 <kmc> do destructors of dlopen'd things get called when main returns? what about on exit()?
04:43:25 <zzo38> kmc: I don't know.
04:48:02 * Sgeo_ vaguely remembers a suggestion that main in Haskell should be required to have a type of IO a, I think
04:48:16 <Sgeo_> That either it infinitely loops or explicitly quits
04:48:32 <Sgeo_> I don't remember if the type to enforce that is IO a, or something similar
04:48:53 <Sgeo_> Although I guess it's really unenforcable, main = undefined
04:49:27 <shachaf> That counts as "infinitely loops".
04:50:32 <Jafet> Haskell computes that it is uncomputable
04:52:05 <coppro> Sgeo_: what is the other option?
04:52:19 <coppro> besides terminating and not terminating
04:52:29 <Sgeo_> terminating without being explicit about it
04:52:36 <Sgeo_> main = putStrLn "Hello world"
04:53:17 <Sgeo_> Hello world would become main = putStrLn "Hello world" >> exitSuccess
04:53:18 <shachaf> Sgeo_: Sounds like the sort of thing Ada would do.
04:53:20 <Sgeo_> Or something like that
04:54:13 <Jafet> main = return undefined
05:01:07 <kmc> hm why have i not thought about Fix IO before
05:04:57 <tswett> Sgeo_: twelve hours ago, I repealed a rule in Agora.
05:05:26 <shachaf> Is Fix IO particularly interesting?
05:05:33 <shachaf> FreeT f IO is probably more useful, I think.
05:06:28 -!- tswett has set topic: PACZKI (pronounced like "punshki") IS ENOUGH | http://codu.org/logs/_esoteric/.
05:07:00 <shachaf> E.g. newtype Stream m a b = Stream { runStream :: m (Either b (a, Stream m a b)) }
05:07:08 <Sgeo_> tswett, wait, as in, got people to vote to repeal a rule, or as in did it unilaterally?
05:07:10 <shachaf> Well, if you used IO there, anyway.
05:08:06 <tswett> Murphy incorrectly resolved the proposal repealing it; he said that it had passed, but it had actually failed.
05:08:26 <tswett> I was aware of this, but I kept quiet. As far as I know, nobody else noticed.
05:08:39 <tswett> So the resolution self-ratified twelve hours ago.
05:08:58 <FreeFull> shachaf: I don't like the tuple there
05:09:04 <Sgeo_> Was it your proposal?
05:09:10 <tswett> No, I think Pavitra submitted it.
05:09:13 <Sgeo_> Was there a particular motive for you to keep quiet?
05:09:22 <tswett> It repealed Rule 2386 "Belligerence".
05:09:29 <Sgeo_> What was that rule about?
05:09:34 <Sgeo_> It's been a while since I Agoraed
05:09:50 <tswett> It described a little subgame that nobody but me (and woggle, a little bit) was playing.
05:10:13 * Sgeo_ is now curious as to what that subgame was like
05:10:13 <tswett> There was no particular motive for me to keep quiet about it; I just changed my mind and decided I'd rather it be repealed.
05:11:08 <coppro> but your argument appears correct
05:11:13 <tswett> Funny thing. The proposal said 'Repeal the rule titled "Belligerence"'. If it had said 'Repeal Rule 2386' instead, then the self-ratification would have been ineffective.
05:11:21 <coppro> which means that, indeed, the rule would never be repealed no matter what
05:12:04 <coppro> title, text, and/or power
05:12:04 <tswett> Ratification can't cause a rule change, unless the text, title, and/or power of the rule is mentioned in the document ratified.
05:13:57 <zzo38> Maybe what there should be in Haskell though would be to call a IO as if it is its own program so the program will continue after it exit and have its own exit code as the result
05:15:47 -!- SDr has quit (Ping timeout: 246 seconds).
05:17:40 -!- Arc_Koen has quit (Quit: Arc_Koen).
05:24:26 <tswett> Maybe what we ought to do is have a comonad for I/O (called OI, of course), and a type representing entire programs.
05:24:57 <tswett> The Prelude should then have a value universe :: OI Program, and then your program has to define a value main :: Program.
05:25:03 -!- ogrom has joined.
05:26:56 <Sgeo_> comonads still confuse me :(
05:27:17 <monqy> i love them so???????
05:27:22 <Sgeo_> Yes, I get they have an extract and no return
05:27:26 <tswett> Everything will work perfectly. getLine can have the type OI () -> String, and, uh...
05:27:49 <shachaf> monqy = cobeaky???????????????????????
05:27:58 <tswett> Sgeo_: you know how you can think of a monadic value as a computation running within a certain context?
05:28:11 <shachaf> oh boy it's bad comonad analogy time
05:28:16 <shachaf> a refreshing break from bad monad analogy time
05:28:17 <tswett> A comonadic value is a computation *providing* a certain context.
05:28:58 <Sgeo_> So with a monadic value, you need to provide a context. With a comonadic value, there's a context built-in?
05:29:17 <tswett> Yeah, that sounds accurate.
05:29:49 <monqy> you know what's an easy comonad? env. you know reader? you know env.
05:30:22 <shachaf> monqy: you know Co? that's a good monad
05:30:31 <shachaf> Co w a = forall r. w (a -> r) -> r
05:30:32 <tswett> Yeah, looks like Env e a is equivalent to (e, a).
05:30:38 <zzo38> The environment comonad has the same Kleisli category as the reader monad.
05:30:51 <zzo38> And those things with Co are working too.
05:30:55 <tswett> So the context here is just a value of type e.
05:31:26 <tswett> shachaf: don't tell me it's possible to write a constructor called Coco that works in the opposite direction.
05:31:30 <tswett> (I'm reasonably sure it's not.)
05:31:33 <Sgeo_> And extract pulls out a
05:31:49 <shachaf> Hmm, tswett's monad and comonad analogies both don't make sense.
05:31:53 <kmc> cocoa-coated co-cones
05:32:04 <Sgeo_> Takes the value out of the context
05:32:09 <shachaf> kmc: There's such a thing as a cocone. :-(
05:32:13 <shachaf> Isn't it a bit ridiculous?
05:32:17 <Sgeo_> Hey, we should call taking quotes out of context extracting them!
05:32:25 <kmc> you mean they should call it a ne instead?
05:33:04 <zzo38> duplicate for trace comonad then means take two parameters which are combined using mappend, and extract for trace comonad means called the function with mempty, for one thing.
05:33:11 <Sgeo_> So, what's the co of bind, or whatever?
05:33:25 <Sgeo_> I know there's an operation analogous to bind, but don't know what it is, or its type
05:33:32 <tswett> Unrelated stupid word game: take each important word and insert "amp" before the stressed vowel, and then stress the "amp".
05:33:54 <tswett> Hampaskell. Esotamperic. Mamponad.
05:33:56 <zzo38> With a monad you have return,bind,join, while with a comonad you have extract,extend,duplicate.
05:34:08 <zzo38> (fmap is the same for both, since they are both being functors)
05:34:38 <shachaf> zzo38: What about a contravariant equivalent of Monad?
05:34:48 <monqy> does caleskell really not have comonad
05:34:52 <shachaf> zzo38: unjoin :: (p (p a) -> p a) -> p a
05:34:57 <zzo38> shachaf: That is difficult.
05:35:01 <lambdabot> `>>' (imported from Control.Monad.Writer),
05:35:03 <tswett> Sgampeo. Hampomestuck. Dampifficult.
05:35:09 <shachaf> @ty Control.Comonad.extend
05:35:11 <lambdabot> Control.Comonad.Comonad w => (w a -> b) -> w a -> w b
05:35:18 <monqy> shachaf: can you believe this
05:35:44 <Sgeo_> I think comonads also cannot have a universal-across-all-comonads operation that supplies a null context?
05:36:00 <Sgeo_> And that any such operation is specialized per comonad?
05:36:15 <shachaf> I think tswett's analogies have confused Sgeo_ beyond redemption.
05:36:23 <tswett> Sgeo_: I think that sounds right.
05:36:28 <shachaf> That's it. We'll have to throw you out and start with a new Sgeo.
05:36:29 <tswett> Given a value-with-context, you can remove the context.
05:36:46 <tswett> But given a plain old value, you can't generally create a context for it.
05:36:48 <Sgeo_> Ok, so can I get an explanation of extend?
05:37:06 <shachaf> extend f w = fmap f (duplicate w)
05:37:21 <Sgeo_> I said explanation not definition
05:37:35 <tswett> That's honestly a pretty decent way of "explaining" it.
05:37:39 <tswett> Then the question is just what duplicate does.
05:37:46 <tswett> It does not "create a copy of the context".
05:37:49 <zzo38> It is like a functor from the coKleisli category to the base category.
05:38:12 <zzo38> (Like how bind can be like a functor from the Kleisli category)
05:38:53 <tswett> duplicate x :: w (w a) is that value such that if you interact with the context, and then extract, and then interact with the context again, the result is the same as if you had just taken x and interacted with the context in both ways in sequence.
05:39:17 <shachaf> are you trying to be maximally confusing :'(
05:39:29 <tswett> So, you know that way of treating (->) Integer as a comonad?
05:39:36 <zzo38> Am I trying to be confusing?
05:39:40 <tswett> Where it's an infinite sequence that you're allowed to shift?
05:39:48 <zzo38> tswett: Maybe it should be ((->) (Sum Integer))?
05:40:05 <Sgeo_> Um, I'm somewhat aware of the view of a zipper as a comonad
05:40:27 <tswett> (shiftLeft 3 . extract . shiftRight 5 . duplicate) is equivalent to just (shiftLeft 2 . shiftRight 5). I think.
05:40:52 <shachaf> (e ->) is a comonad when e is a monoid.
05:41:05 <tswett> Uhh. To (shiftLeft 3 . shiftRight 5).
05:41:25 <shachaf> monoids are equivalent to easy things in the category of love
05:41:30 <Sgeo_> So given a context you can wrap the whole thing in another context?
05:42:03 <shachaf> Here, you should take this as your example of a comonad:
05:42:05 <tswett> I guess it's kinda like splitting the context into two pieces?
05:42:46 <Sgeo_> I guess extract for that just feeds the s into s->a?
05:42:57 <zzo38> (The MonadPlus instance for Either should be define in that way too, using the monoid; either by (Free (Const x)) or by (CodensityAsk ((->) x)) gives you the MonadPlus instance for free, assuming the certain way which gives you MonadPlus of Free, and Plus of Const)
05:43:06 <tswett> So, a Store s a is... well, yeah. It's a (s, s -> a).
05:43:17 <tswett> Store s (Store s a), then, is (s, s -> (s, s -> a)).
05:43:33 <Sgeo_> You could make a neutral context for that quite easily... but again, probably specific to Store
05:43:39 <shachaf> duplicate = fmap (return :: a -> State s a)
05:44:03 <tswett> I'm pretty sure that duplicate (c, f) = (c, \c' -> (c', f)).
05:44:29 * Sgeo_ finds tswett's easier to understand
05:44:39 <shachaf> tswett's has the problem that
05:44:46 <shachaf> duplicate (c, f) = (c, \c' -> (c, f))
05:45:21 <shachaf> also adjunctions are the future????
05:45:57 <tswett> So if you have a Store s a, it's like it has an s, and it's about to apply a function to it, but you're allowed to play with the s and the function before it does so.
05:46:01 <tswett> (shachaf begins crying.)
05:46:27 <tswett> If you "duplicate" that, then you have the same thing, except now you're allowed to do that twice.
05:48:16 <zzo38> (Store x) is also being like (Density (Const x))
05:52:50 <Sgeo_> Why wouldn't I be allowed to do it twice without duplicating it?
05:57:50 * Sgeo_ vaguely imagines a Squeak-like environment for Haskell
05:57:57 <Sgeo_> Seems... weird to think about, but
05:58:43 <Bike> late binding and static typing, what could possibly go wrong
06:00:09 <tswett> Sgeo_: you would be allowed to do that.
06:00:20 <tswett> But now if you have a function of the type (w a -> b), you can use it without losing access to the context.
06:00:53 <tswett> Or... something like that.
06:06:27 <Sgeo_> Is there a comonad corresponding to the Cont monad?
06:06:43 <shachaf> There's a comonad in Hask^op :-)
06:06:49 <tswett> If I remember correctly, the... yeah, that.
06:06:50 <Sgeo_> If the Cont monad is the mother of all monads, is the comonad for that the son of all comonads?
06:07:01 <shachaf> That would probably be Store
06:07:06 <tswett> There is such a comonad, but it's in the wrong category.
06:07:11 <shachaf> Store is related to Density in the same way that Cont is related to Codensity
06:07:22 <Sgeo_> I don't have the faintest idea what Density is
06:07:38 <shachaf> Density is the mother of all comonads. Or something.
06:07:44 <tswett> Hm. Is every monad in a category also a comonad in its opposite category?
06:08:07 <shachaf> tswett: I think that's the definition of a comonad. :-)
06:09:40 <tswett> Do the state comonads in Hask^op have anything in particular to do with the store monads in Hask?
06:11:36 -!- TeruFSX has quit (Ping timeout: 252 seconds).
06:18:13 -!- TeruFSX has joined.
06:38:49 -!- Nisstyre-laptop has quit (Quit: Leaving).
06:48:39 <shachaf> monqy: can we put spoilers in here
06:48:58 <shachaf> because in the end it turns out
06:55:21 -!- Nisstyre-laptop has joined.
07:12:34 -!- TeruFSX has quit (Ping timeout: 256 seconds).
07:15:29 <Sgeo_> I should try to sleep some get
08:08:17 -!- epicmonkey has joined.
08:11:44 <zzo38> Cont monad is not really mother of all monads; there is Codensity monad, and (Codensity (Const x)) makes (Cont x).
08:15:35 <shachaf> Yes, Codensity is the true MOAM.
08:26:15 -!- augur_ has quit (Remote host closed the connection).
08:26:37 -!- augur has joined.
08:30:28 <HackEgo> Taneb atriq Ngevd Fiora nortti Sgeo ThatOtherPerson alot
08:32:13 -!- oerjan has joined.
08:38:22 -!- Halite has joined.
08:39:23 <Halite> > let doppleganger x = reverse x;
08:39:25 <lambdabot> not an expression: `let doppleganger x = reverse x;'
08:39:39 <Halite> > let doppleganger x = reverse x in doppleganger "Halite"
08:40:21 <oerjan> despite what people lambdabot may have told you yesterday, lambdabot is not quite ghci. > only takes expressions, @let only takes a declaration
08:40:44 <Halite> I know that now, oerjan.
08:41:10 <Halite> I wish they added data, so I could make a Nickname type.
08:41:26 <oerjan> also :t and :k work like in ghci, but only in channel for some stupid reason (@type and @kind work in private to do the same)
08:41:51 <oerjan> it's only recently that ghci itself got data working though
08:42:06 <Halite> I once attempted to make a Quantum Super-position type constructor.
08:42:41 <oerjan> (before you had to put that in a file)
08:45:09 <Halite> I'm still working on my Quantum type constructor. What I'd like is to be able to make SuperPosition x y and SuperPosition y x equivalent.
08:45:32 <Bike> does that actually have anything to do with quantum
08:47:10 <Halite> Bike, yes. A qubit (a Quantum Bit) can be 1 (true), 0 (false), or a super-position between 1 and 0 (SuperPosition x y).
08:47:34 <oerjan> i sense a disturbing lack of complex numbers hth
08:47:46 <shachaf> oerjan: did you see my diagram
08:48:05 <Bike> Halite: qubits have more than three states.
08:48:16 <shachaf> oerjan: I guess you'll be logreading, though.
08:48:22 <Halite> Bike, what are them states
08:48:38 <Bike> the various possible superpositions of |0> and |1>.
08:49:55 <Bike> I'm a fan of (1/2)|0> + (sqrt(3)/2)|1>, myself. I think that's a pretty good state.
08:50:11 <zzo38> Did you consider the ones with complex numbers?
08:50:19 <oerjan> Bike: nah, not enough i's
08:50:39 <shachaf> oerjan: "i'm a fan of (1/2)|0> + (sqrt(3)/2)|1>, myself. i think that's a pretty good state." -- better?
08:50:53 * oerjan swats shachaf -----###
08:52:06 <zzo38> Did you consider entanglement?
08:52:10 <shachaf> 00:51 <hackagebot> hackage-proxy 0.1.0.0 - Provide a proxy for Hackage which modifies responses in some way. http://hackage.haskell.org/package/hackage-proxy-0.1.0.0 (MichaelSnoyman)
08:52:24 <Bike> What's a good one with i's?
08:52:35 <shachaf> kmc: You should submit a patch to insert malicious code on the fly.
08:54:04 <Bike> Halite: also (1/2)|0> + (sqrt(3)/2)|1> isn't the same state as (sqrt(3)/2)|0> + (1/2)|1>, I'm not sure why you'd think that, and this has nothing to do with types.
08:54:15 -!- Bike has quit (Quit: leaving).
08:54:39 <oerjan> (1/2)i|0> + (sqrt(2)i/2 - 1/2)|1> perhaps
08:55:34 <oerjan> @tell Bike (1/2)i|0> + (sqrt(2)i/2 - 1/2)|1> perhaps, also no fair leaving just after asking a question
08:56:36 <zzo38> Maybe if you have multiple bits like sqrt(2)|0>|0> + sqrt(2)|1>|1> is that OK?
08:56:41 <Halite> everything computer-rendered (text, images, etc.) are all made of bits
08:57:14 <zzo38> Halite: I mean quantum bits
08:57:21 <Halite> if we change these bits to qubits, then we could have atleast one of the uncountably many superpositions
08:57:58 <zzo38> They will have to collapse once measured, though
08:58:17 <zzo38> (Although, you do not have to always measure in the same direction)
08:59:47 <Halite> of course, there are two superpositions in my type constructor relating to true-false booleans: true 0> and false 1>, then true 1> and false >0.
09:00:19 -!- nooodl has joined.
09:03:01 <zzo38> Do you even understand it? It just doesn't work that way.
09:06:24 <oerjan> @tell Phantom_Hoover sed -ie doesn't do what you think it does, the e is an argument to the -i
09:06:42 <HackEgo> rm: cannot remove `bin/liste': No such file or directory
09:07:50 <HackEgo> #!/bin/sh \ hg diff -c "$@" | patch -R
09:17:04 -!- carado has joined.
09:20:51 <mroman> If a FSM has access to another machine which produces completely random numbers
09:21:29 <mroman> That FSM is fed a program, that only terminates, if the random machine produces a certain sequence of numbers
09:21:58 <mroman> Is it a finite state machine if it has access to such a machine
09:23:36 <oerjan> @tell Vorpal <Vorpal> also my PS3 controller on my desk random came to life and blinked the 4 LEDS a couple of times <-- probably just the google cloud trying to print
09:25:30 <oerjan> mroman: it would be nondeterministic, and nondeterministic FSMs are equivalent in power to deterministic ones. although that's not really a proof.
09:26:29 <oerjan> since it might not be the right sense of nondeterministic.
09:27:38 <mroman> But the FSM in combination with the random machine might as well have not finite state anymore?
09:28:02 <oerjan> but the random machine has _no_ state.
09:28:24 <mroman> Not in the common sense.
09:28:26 <oerjan> what you get is a probabilistic FSM, i guess.
09:28:36 <mroman> (unless it were a PRNG machine)
09:28:54 <oerjan> which are pretty much what markov models use, right fungot?
09:28:54 <fungot> oerjan: that we told the power.
09:28:55 <mroman> but I'm going to assume that the random machine has no state other than the world it is in.
09:33:49 <mroman> That's a class of FSM where it is undecidable if it'll halt or not?
09:34:59 <oerjan> well given that it's probabilistic, you'd get a probability of it halting... but i would expect it's calculable
09:35:21 <mroman> Since it only has finite state it can't accept arbitarily large random numbers
09:35:26 <oerjan> so not really undecidable in a meaningful sense
09:35:32 <mroman> so there will be a calculable probabilty of halting.
09:36:17 <oerjan> in fact i'm pretty sure this can be calculated using eigenvalue methods based on perron-frobenius theory.
09:36:37 -!- carado has quit (Remote host closed the connection).
09:37:37 <oerjan> as in that new scoring method for bfjoust which i still don't think has been implemented.
09:38:15 <oerjan> it's a different problem. but still given by the same matrix.
09:38:53 -!- carado has joined.
09:39:38 <oerjan> oh hm no, it's really the same, it's just that the problem gives a state ("has found the substring") which you never leave once you enter it.
09:40:12 <oerjan> the only way you could avoid getting to that state is if the FSM has another set of states which you can never leave.
09:40:42 <oerjan> except that would not be a program to search for a substring. in fact if you search for a substring the probabiliy is obviously 1, really :P
09:41:36 <oerjan> but the more complicated problem of determining the probability of the FSM entering a given state for an arbitrary program would be solvable by such matrix methods.
09:42:47 <oerjan> now i just thought of this thing - what if you don't have _entirely_ random numbers, but instead get the output of a markov model encoded in another FSM (hi fungot)
09:42:47 <fungot> oerjan: the corporate and public. 10. from the " let the market will be best to wait. subject: power keeps on giving you a chance of the commission
09:42:59 <mroman> Fun thing is, that the probability of such a program to terminate can increase the longer it runs.
09:43:24 <oerjan> in that case you would have two FSM's interacting, which i think you could just merge into one.
09:44:35 <oerjan> mroman: hm... i was thinking this ran indefinitely. if it's a set number of steps, it gets a bit different. then you would calculate that power of the matrix, rather than looking for eigenvectors.
09:44:37 <mroman> If they are generated by a PRNG which is deterministic
09:44:58 <oerjan> in that case the PRNG itself is probably an FSM, no
09:45:37 <oerjan> i was assuming the markov model still had "true" randomness inside
09:45:44 -!- Taneb has joined.
09:45:53 <oerjan> just with variable probabilities dependent on state
09:46:32 <mroman> In the case of while(rnd() != 0); it's actually fairly easy to calculate the probability of it halting.
09:47:08 <mroman> (given that rnd() returns entirely random numbers in the range 0..n)
09:47:25 <mroman> (and all numbers are equally likely to occurr)
09:48:41 <mroman> how does that apply to programs that calculate infinite sums?
09:49:01 <mroman> 1/3 + 1/9 + 1/27 + ...
09:49:17 <mroman> and it halts if the sum reaches 1/2
09:49:32 <oerjan> i would expect you cannot do that in finite state
09:50:05 <mroman> I wouldn't now how to encode that
09:50:11 <oerjan> the sum of that is 3/2 iicc
09:50:51 <Taneb> x+1 = x*3; 2x = 1; x = 1/2
09:51:15 <mroman> oerjan: The sart value is 1/3
09:52:11 <mroman> that calculation is done by another machine coupled to my fsm
09:52:18 <mroman> my fsm machine polls the sum machine
09:52:33 <mroman> which returns 1 if it has reached 1/2 and 0 if it is not there yet.
09:53:11 <oerjan> actually you _could_ do that in finite state if you get it as a sequence of ternary digits
09:53:19 <mroman> while(otherMachineDone()) { /* Nope, still waiting */ }
09:53:56 <mroman> Does my fsm terminate or does it not?
09:54:20 <oerjan> it doesn't terminate, naturally, since that sequence is precisely on the boundary
09:55:01 <oerjan> it never gets a 0 ternary digit, so it cannot say that the sum never reaches 1/2; it never gets a 2, so it cannot say that it does
09:55:53 <oerjan> and any rational number boundary can be tested with an FSM. which may not halt if it is precisely on the boundary.
09:56:36 <mroman> So an argument that it'll reach 1/2 in infinite steps is just plain wrong?
09:57:00 <oerjan> well "infinite steps" is pretty much the opposite of halting, no? :P
09:57:17 <mroman> but it'll terminate if it runs for ever
09:57:31 <mroman> (if one can say that it'll reach 1/2 in infinite steps)
09:58:13 <oerjan> you're going to need to redefine terminate for that. also get a supply of limits and transfinite ordinals.
09:59:44 <mroman> "It doesn't terminate unless it runs forever" sounds very confusing to me.
10:00:44 <mroman> the guy who claimed there's no such thing as motion.
10:00:50 <oerjan> argh there's even more logs after midnight UTC :(
10:01:20 <oerjan> he would definitely have said you couldn't reach 1/2.
10:01:21 -!- zzo38 has quit (Quit: Is saying there is just one God like saying that all the water in the ocean is just one water?).
10:01:26 <olsner> ok, everybody shut up until oerjan catches up with the log reading
10:01:58 <oerjan> Halite: if you say so.
10:03:26 <mroman> If my FSM is infinetly fast (like math is)
10:04:43 <oerjan> <shachaf> monqy: i'm trying to figure out what pullbacks are in a poset category? <-- surely they're just meets?
10:07:36 <oerjan> i think the fact that a poset category has only one morphism between objects at most means that it's the same as the product of the two initial objects and the thirs is ignored.
10:07:48 <Halite> In quantum data, does 1 ket 0 + 0 ket 1 equal 0.5*2 ket 0 + 0.5-0.5 ket 1
10:08:15 <oerjan> @ask shachaf <shachaf> monqy: i'm trying to figure out what pullbacks are in a poset category? <-- surely they're just meets?
10:08:23 <oerjan> @tell shachaf i think the fact that a poset category has only one morphism between objects at most means that it's the same as the product of the two initial objects and the thirs is ignored.
10:08:51 <oerjan> Halite: your syntax is hard to understand.
10:08:56 <Halite> (In my code I am replacing SuperPosition with my new Ket function.)
10:09:34 <oerjan> if i understand it, the answer is yes, anyway
10:10:01 <monqy> | and > are easy to type
10:10:04 <Halite> oerjan, what do you understand it as
10:10:21 <Halite> monqy, the pipe character doesn't work because my keyboard is horrible.
10:10:29 <oerjan> 1 |0> + 0 |1> = 0.5*2 |0> + (0.5-0.5) |1>
10:10:45 <monqy> keep it on your clipboard
10:10:47 <Halite> oerjan, that's what I meant. thank you for helping.
10:11:04 <Halite> monqy, it was already on my clipboard for some reason :o
10:11:28 <monqy> that's what happens when a common character doesn't work, it just ends up there somehow??
10:11:40 <Halite> monqy, but other keys don't work either such as the solidus (slash) and arrow keys, that's why I can't use question mark
10:13:01 <monqy> maybe you should get a new keyboard
10:13:41 <Halite> monqy, I have a laptop
10:14:27 <monqy> maybe you should get a new laptop
10:14:42 <Halite> monqy, but meh data and meh Ubuntu
10:14:58 <Halite> monqy, the laptops you get are Macintosh or Windows nowadays
10:14:59 <monqy> a usb keyboard and type on it
10:15:15 <Halite> how about a USB computer
10:15:27 <Halite> (the screen coming apart too)
10:15:28 <monqy> get a windows laptop and overwrite it; that's what i do
10:15:49 <Halite> laptops cost too much monet
10:18:02 <monqy> if your laptop isn't from “nowadays„ you might be able to find a cheap laptop that's better than whatever you're using
10:18:28 <Halite> my laptop is close to nowadays
10:19:14 <Halite> I need the slash to join #haskell
10:20:06 <oerjan> haskell without | is going to be awkward.
10:39:08 <lambdabot> oerjan asked 30m 52s ago: <shachaf> monqy: i'm trying to figure out what pullbacks are in a poset category? <-- surely they're just meets?
10:39:08 <lambdabot> oerjan said 30m 45s ago: i think the fact that a poset category has only one morphism between objects at most means that it's the same as the product of the two initial objects and the thirs is
10:39:26 <shachaf> Arrows in limit diagrams pretty much don't matter because commutativity is trivial.
10:39:36 <shachaf> (When you only have at most one arrow between objects.)
10:49:57 <oerjan> @tell Bike <Bike> "manocentric" says my internal ignorant pedant, though. <-- "manu-" hth
10:53:41 -!- monqy has quit (Quit: hello).
10:57:48 <shachaf> hello to you too, dear monqy
10:58:29 <shachaf> You're acting a lot like a troll.
10:58:42 <shachaf> In #esoteric that's one thing, but #haskell and #agda are a different matter.
11:09:50 -!- Taneb has quit (Ping timeout: 252 seconds).
11:16:04 -!- Phantom_Hoover has joined.
11:29:16 -!- DHeadshot has quit (Ping timeout: 256 seconds).
11:31:13 -!- nooga has joined.
11:38:38 <shachaf> Halite: Would you stop wasting everyone's time?
11:42:03 <Jafet> Welcome to the international hub of #haskell pundits
11:42:21 <oerjan> yay finished the logs!
11:43:24 <shachaf> or is it like open and closed sets
11:44:11 <oerjan> i think open and closed sets might be dual, so... i guess?
11:59:42 -!- Halite has quit (Remote host closed the connection).
12:13:10 <fungot> 59 126 58 33 42 40 41 97 94 83
12:13:38 -!- hagb4rd has quit (Ping timeout: 252 seconds).
12:14:59 <oerjan> > sort [(ord x, x) | x <- ";~:!*()a^S"]
12:15:01 <lambdabot> [(33,'!'),(40,'('),(41,')'),(42,'*'),(58,':'),(59,';'),(83,'S'),(94,'^'),(9...
12:15:12 * oerjan swats lambdabot -----###
12:15:49 <Jafet> > sort ";~:!*()a^S"
12:17:22 <shachaf> > unwords [show (ord x) ++ [x] | x <- sort ";~:!*()a^S"]
12:17:24 <lambdabot> "33! 40( 41) 42* 58: 59; 83S 94^ 97a 126~"
12:18:20 <Jafet> > unzip $ sort [(ord x, x) | x <- ";~:!*()a^S"]
12:18:21 <lambdabot> ([33,40,41,42,58,59,83,94,97,126],"!()*:;S^a~")
12:21:28 <shachaf> is thanx the dual of swatting
12:21:35 <shachaf> you need some ascii art for it
12:22:18 <shachaf> oerjan: what category thing should i figure out next
12:22:43 * oerjan doesn't understand them, but is sure they are cool
12:23:16 <shachaf> mr.hird thinks (∞,n)-categories are the answer to all of life's problems
12:23:16 <oerjan> cartesian closed categories, then?
12:23:45 <oerjan> maybe they are, but i don't know what they are
12:24:23 <oerjan> i suppose you could ask at the n-category café
12:24:29 <shachaf> I guess I should work out ends.
12:24:51 <oerjan> all is good that ends well
12:25:29 <oerjan> also, i think you may have passed my CT knowledge level already
12:26:14 <oerjan> unless you weer off into homology stuff, which requires algebra on top
12:26:36 <shachaf> I don't even know what homology is. :-(
12:28:01 <shachaf> I bet those are pronounced the same way in Norwegian.
12:28:14 <shachaf> Norwegian is the best language.
12:28:25 <shachaf> I mean, it has the empty set in its alphabet.
12:30:30 * shachaf vanishes in a puff of smoke
12:30:31 <oerjan> homology is basically a sequence of functors defined from topological spaces into groups. and then you generalize it to modules over rings, and stuff.
12:31:14 <shachaf> Is cohomology just mplicated?
12:31:18 <oerjan> and this tells you things about the topological space, or the rings.
12:31:39 <lambdabot> Phantom_Hoover: You have 1 new message. '/msg lambdabot @messages' to read it.
12:31:55 <oerjan> well cohomology means you take the dual modules at one step of the construction
12:32:21 <oerjan> Phantom_Hoover: well for example it gives you proofs of things like the jordan-brouwer theorem
12:33:20 <oerjan> which includes the jordan curve theorem as a special case (which has other known proofs)
12:34:37 <oerjan> basically, if you deform a sphere arbitrarily but still continuously, it will still always separate a surrounding euclidean space (of one more dimension) into two pieces.
12:43:40 -!- epicmonkey has quit (Read error: Operation timed out).
12:58:37 -!- epicmonkey has joined.
13:23:10 -!- oerjan has quit (Quit: leaving).
13:37:29 <Phantom_Hoover> hey, i found another terrible swede http://www.reddit.com/r/math/comments/18mh1p/my_favorite_song_about_doing_calculus/c8gm0ii
13:41:32 -!- azaq23 has joined.
13:41:46 -!- azaq23 has quit (Max SendQ exceeded).
13:42:08 -!- azaq23 has joined.
14:11:16 <mroman> How is it that books containing "An introduction" are usually completely useless
14:11:38 -!- esomimic has joined.
14:11:39 <esomimic> fizzie: say 'mimic X' to mimic X; 'who' for list; anything else for a reply
14:12:09 <mroman> esomimic: mimic fizzie
14:12:09 <esomimic> mroman: okay, mimicking fizzie
14:12:10 <fizzie> I forgot to mention you need to say it to im.
14:12:25 <mroman> esomimic: do your work
14:12:26 <esomimic> mroman: i've used a lot in gnu canything only included in the possibility.
14:12:49 <fizzie> Hm, I wonder if the "canything" is a bug.
14:13:06 <mroman> esomimic: mimic esomimic
14:13:06 <esomimic> mroman: sorry, don't know 'esomimic' well enough
14:13:16 <mroman> esomimic: mimic mroman
14:13:16 <esomimic> mroman: sorry, don't know 'mroman' well enough
14:13:23 <esomimic> fizzie: ais523 asiekierka augur bsmntbombdood CakeProphet coppro cpressey Deewiant elliott *fizzie fungot Gregor ihope itidus21 kallisti kmc lament monqy nooga oerjan oklopol olsner Phantom_Hoover pikhq quintopia Sgeo shachaf Taneb Vorpal zzo38
14:13:23 <fungot> esomimic: that will assume the first is the pulmonary injury by deal count by trader and product for the existing state plans of a message on the its board of directors on dec. 21.
14:13:24 <esomimic> fungot: it's following screens, sdsm for.
14:13:25 <fungot> esomimic: his and your management have been a few that could make the decision on the number of our best to keep the current state of the street. tullman, ( 888) 593-4771 or ( 212) 686-6808, email the new bond language has an offer.
14:13:26 <fungot> esomimic: 1. i in room eb1938 to formalize an already executed more than that of the market that would please be soooo busy lately. terminated 12.1 e prime, inc. on a thursday that the last two and a state of my and john
14:13:27 <fungot> esomimic: an internet exchange for chemicals that cause of the it central website. rick causey, buy and deal entry.
14:14:02 <fizzie> At least fungot has the loop protection.
14:14:02 <fungot> fizzie: he recalled the critics want. settlement.
14:14:12 <esomimic> Phantom_Hoover: okay, mimicking itidus21
14:14:34 <fizzie> I think that's an "ICE BURN".
14:14:41 <fizzie> esomimic: Really, now.
14:14:41 <esomimic> fizzie: areas of an encoding of an efficient potatoes it, and concentrate, and link
14:14:57 <esomimic> Phantom_Hoover: im not asking pretty cool
14:15:19 <mroman> esomimic: Gimme your lunch money or else...!
14:15:36 <mroman> esomimic: thats where you call the police? We'll see about that!
14:16:04 <fizzie> I don't really know why it's so much more incoherent; it's done with SRILM, though.
14:17:11 <esomimic> Phantom_Hoover: okay, mimicking zzo38
14:17:24 <esomimic> Phantom_Hoover: +one of a subroutine greater and you can refer that, also
14:17:25 <mroman> just give him an omegle backend
14:20:59 <esomimic> fizzie: okay, mimicking elliott
14:21:05 <fizzie> esomimic: What would elliott do?
14:21:16 <fizzie> esomimic: Sounds vaguely dirty.
14:21:21 <esomimic> fizzie: i am not sure which wiki/functions how can i become a "game comes from the +or sth?"
14:21:28 <fizzie> (That model is a giant, it takes a while to load.)
14:21:41 -!- Taneb has joined.
14:21:56 <fizzie> esomimic: Greet your guy.
14:22:15 <Taneb> esomimic, what's doing?
14:22:35 <Taneb> esomimic, seen anything cool lately?
14:22:35 <esomimic> Taneb: i am aware the don't have interesting
14:23:04 <mroman> He might need some tweaks.
14:23:12 <Taneb> esomimic: you're right! The don't do have interesting
14:23:23 <Taneb> mroman, it's perfect
14:23:37 <fizzie> esomimic: mimic ais523
14:23:37 <esomimic> fizzie: okay, mimicking ais523
14:23:42 <fizzie> esomimic: Try to sound sensible for a while
14:23:43 <esomimic> fizzie: next time you are using clc-intercal
14:23:56 <fizzie> esomimic: ...then what?
14:23:57 <esomimic> fizzie: unfortunately i haven't submitted (i haven't modified well context is ridiculously slow for that sort of thing, they keep)
14:24:38 <esomimic> Taneb: +it could be fun to shrink too, as i remember who it was? i guess it's a really bad puns is translated into but suboptimal but not the accumulator, it becomes whole shape drawing with a it with a stupid sure or +it isn't wikimedia-connected behaviour same as the form
14:24:55 <Taneb> esomimic: mimic itidus20
14:24:55 <esomimic> Taneb: okay, mimicking itidus21
14:25:01 <Taneb> esomimic, how's it going
14:25:02 <esomimic> Taneb: billion spend i have a mass c is fun to make. and ((s x i am already mentioned great deal be more of i saying "an image, and i thought human then it is better shot bastards"))
14:25:20 <fizzie> All the +s are probably from the old freenode capab identify-msg thing.
14:25:25 <Taneb> <Taneb> esomimic: mimic itidus20
14:25:25 <Taneb> <esomimic> Taneb: okay, mimicking itidus21
14:25:36 <fizzie> It has a bit of nick-merging.
14:25:58 <Taneb> esomimic, mimic atriq
14:26:02 <Taneb> esomimic, mimic AnMaster
14:26:07 <Taneb> esomimic, mimic kallisti
14:26:07 <esomimic> Taneb: okay, mimicking kallisti
14:26:11 <Taneb> esomimic, mimic CakeProphet
14:26:11 <esomimic> Taneb: okay, mimicking CakeProphet
14:26:17 <Taneb> esomimic, mimic RocketJSquirrel
14:26:17 <esomimic> Taneb: sorry, don't know 'RocketJSquirrel' well enough
14:26:24 <fizzie> It's somewhat limited.
14:26:26 <Taneb> esomimic, mimic cpressey
14:26:26 <esomimic> Taneb: okay, mimicking cpressey
14:26:40 <esomimic> Phantom_Hoover: an what i was thinking ah ha i'm out constructing extremely but see, hurd interesting at (which may be a word as a technical them probably seems very and english make the asylum.)
14:28:44 <fizzie> It might be less grammatik than fungot due to srilm actually using the backoffs when generating. Maybe.
14:28:44 <fungot> fizzie: at the time, on the same in the file that the modifications. light http://www.brookings.edu/ press/ review/ rev_des.htm book event: politics prose, 5015 connecticut ave., suite of energy
14:29:05 <fizzie> Well, that's not such a great showing either.
14:30:08 <esomimic> fungot: i complicated s i want i if your mediocre brainfuck many of the out
14:30:09 <fungot> esomimic: they do not have any authority on an urgency of the money.) bombay.
14:30:09 <esomimic> fungot: a kind of behaviour, it to entries. something. i got the normal mechanism something specify probably be friend i apologize that
14:30:09 <fungot> esomimic: see the attached. 99-14563 doing that." wagner rates enron " keep you in the loop on the below: index: gas daily, find.
14:30:10 <fungot> esomimic: to complete the form of the meeting.??
14:30:10 <esomimic> fungot: there's got a a dir, but we'll see languages was a significant "planets"
14:30:10 <fungot> esomimic: go to: http://www20.cera.com/ client/ company/ account what locations on 21 will not let
14:30:10 <esomimic> fungot: if i did, for example, if you ask... that's copying is (full!)
14:30:14 <esomimic> mroman: ais523 asiekierka augur bsmntbombdood CakeProphet coppro *cpressey Deewiant elliott fizzie fungot Gregor ihope itidus21 kallisti kmc lament monqy nooga oerjan oklopol olsner Phantom_Hoover pikhq quintopia Sgeo shachaf Taneb Vorpal zzo38
14:30:22 <mroman> esomimic: mimic fungot
14:30:22 <esomimic> mroman: okay, mimicking fungot
14:30:22 <fungot> mroman: is the spot only the state alabama alaska arizona arkansas and lsu.
14:30:22 <fungot> esomimic: in that case you will put the issue of the talk" face="verdana, helvetica, arial, helvetica, arial, helvetica, sans-serif"bfont color="cc0000" size="3"br all sales employees.
14:30:23 <fungot> esomimic: they pay the esp. to weather i've seen in a spread. commercial power representative and any representative of the price. +44 ( 20) the worst of california's/ utilities -2: not the first and will assume the first to the main source of the davis plan, giving the state
14:30:23 <esomimic> fungot: miizzzzaaaaaaaarrrrrrrrrrrrrrrrddddddddddddddddddddddddddddddddxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx way to the ocean palace of irc* iwcs jargon lovecraft nethack pa qwantz sms speeches ss wp youtube
14:30:23 <fungot> esomimic: that is not on any of the questions on the espeak site urging the company's) plan."
14:30:24 <fungot> esomimic: what do we know the number of questions." granted the right of the as gets exercised and processed.
14:30:24 <esomimic> fungot: to, uhhh.......... colours i time!
14:30:43 <Taneb> fizzie, you may want to fix that
14:31:02 <mroman> esomimic: What are you doing?
14:31:03 <esomimic> mroman: hey fhjlkdshfkjasdasd hjlkdshfkjasdasd 0
14:31:15 <Phantom_Hoover> i assume we were saved there by fungot's rate limiting
14:31:16 <fungot> Phantom_Hoover: that we fail the credit and the fun of our earnings this quarter we have attached the new with every time we utilized our partnership with the major power plants.
14:32:22 <Taneb> esomimic, mimic shachaf
14:32:23 <esomimic> Taneb: okay, mimicking shachaf
14:32:27 <Taneb> esomimic, what's a lens
14:32:35 <Taneb> esomimic, I do mean that good
14:32:36 <esomimic> Taneb: i think it haskell get a so you're.
14:32:46 <Taneb> (it's not wrong, per se)
14:32:58 <Taneb> (its grammar is a tad dodgy)
14:33:15 <elliott> `addquote <esomimic> fungot: begrudging pat
14:33:15 <fungot> elliott: the risk of an approved site
14:33:19 <HackEgo> 964) <esomimic> fungot: begrudging pat
14:35:44 -!- Frooxius has joined.
14:37:20 <fizzie> "there's also a duck are you actually want to get around it." Yeah, "tad dodgy".
14:37:53 <Taneb> Does it work on similar lines to Fungot?
14:38:16 <fizzie> It's very similar, except it uses an existing toolkit (SRILM) to do all the work.
14:38:39 <fizzie> And speaker-dependent models, of course.
14:39:26 <elliott> Taneb: Do you think that person in #haskell is telling the truth?
14:39:48 <Taneb> I doubt anyone has that many hyphens in their name
14:40:17 <Taneb> But you should be prepared to help anyone at any time!
14:41:14 <Taneb> Halite reminds me of Kankri an a weird way
14:44:08 -!- esomimic has quit (Remote host closed the connection).
14:44:20 -!- esomimic has joined.
14:44:20 <fizzie> Possibly it no longer the fungot.
14:44:38 <Taneb> esomimic, mimic fungot
14:44:39 <fungot> Taneb: of the deal will be at the airport. they
14:44:39 <fungot> esomimic: all of the mail the original. " i lost my balance of the 20 years."
14:44:39 <esomimic> fungot: selected style: lovecraft (h. p. lovecraft's writings)
14:44:39 <fungot> esomimic: yet the lights on.????????
14:44:40 <fungot> esomimic: all we have had the option of going on.??????
14:44:40 <esomimic> fungot: (((cathedral to the west?... yes!! you brought back my cat! thank you, crono!!! help!)))
14:44:40 <fungot> esomimic: we should know that the summer of 2000.
14:44:41 <esomimic> fungot: helps oerjan: oerjan: oerjan: ct (chrono trigger game script)
14:45:34 -!- esomimic has quit (Remote host closed the connection).
14:45:47 -!- esomimic has joined.
14:45:49 <fizzie> esomimic: mimic fungot
14:45:49 <esomimic> fizzie: okay, mimicking fungot
14:45:49 <fungot> fizzie: that is not to... .a bunch of construction within an iso market? interface. come support the houston cougars as they held stock to the authority that is important that we
14:45:57 <esomimic> Phantom_Hoover: you be in cloud ventral been a time see a grave to the north. it's a great place for a picnic! heard that magus's place...
14:46:29 <fizzie> esomimic: How did you know it was about Chrono Trigger?
14:46:33 <Taneb> > map fromEnum "fungot"
14:46:34 <lambdabot> mueval: recoverEncode: invalid argument (invalid character)
14:46:37 <Taneb> > map fromEnum "fungot"
14:46:38 <lambdabot> mueval: recoverEncode: invalid argument (invalid character)
14:46:51 <fizzie> It's just a zero-width space.
14:47:39 <esomimic> elliott: sorry, don't know 'lambdabot' well enough
14:47:54 <fizzie> I didn't teach the other bots. :/
14:48:47 <esomimic> elliott: sorry, don't know 'optbot' well enough
14:49:04 <esomimic> Taneb: i grow so tired and stiff
14:49:06 <esomimic> Taneb: ais523 asiekierka augur bsmntbombdood CakeProphet coppro cpressey Deewiant elliott fizzie *fungot Gregor ihope itidus21 kallisti kmc lament monqy nooga oerjan oklopol olsner Phantom_Hoover pikhq quintopia Sgeo shachaf Taneb Vorpal zzo38
14:49:12 <Taneb> esomimic, mimic Phantom_Hoover
14:49:12 <esomimic> Taneb: sorry, don't know 'Phantom_Hoover ' well enough
14:49:28 <esomimic> Phantom_Hoover: okay, mimicking Phantom_Hoover
14:49:32 <fizzie> Maybe it should strip some spaces.
14:50:08 <esomimic> elliott: okay, mimicking zzo38
14:50:11 <elliott> esomimic: i'd like to see you try
14:50:12 <esomimic> elliott: i am using the command-line mean.
14:50:17 <esomimic> elliott: i thought it has amber is for those, the server than....
14:50:40 <fizzie> esomimic: You're not quite like the real deal.
14:50:41 <esomimic> fizzie: did you write it using the correct in category theory a disk and then assume there is a bad, so i have no.
14:50:52 <elliott> fizzie: can you make it preserve caps/punct
14:51:54 <fizzie> elliott: In theory I could just skip the preprocessing. Though then the models might be a bit more sparse.
14:52:09 <fizzie> Admittedly it's not very good mimicry when it normalizes that stuff.
14:52:29 <fizzie> esomimic: mimic fizzie
14:52:29 <esomimic> fizzie: okay, mimicking fizzie
14:52:30 <fizzie> esomimic: But why are you talking in all lowercase? I don't do that!
14:52:31 <esomimic> fizzie: you two 8-ball say "?"
14:53:42 -!- Arc_Koen has joined.
14:58:46 <fizzie> esomimic: Try to be like the 2002 fizzie.
14:58:47 <esomimic> fizzie: obviously haven't the gnu think anyone was doing the whole lot more impressive in this dates the point in a bus.
15:00:28 <nooga> how does this mimicking work?
15:03:40 -!- DHeadshot has joined.
15:04:25 <Gregor> Who is this esomimic, and why does he mimic someone who spams X_X
15:04:46 <fizzie> I haven't seen any spammed "X_X"s.
15:05:44 <Gregor> I'm about to spam filet knives to your jugular vein.
15:06:47 <elliott> esomimic: mimic Phantom_Hoover
15:06:47 <esomimic> elliott: okay, mimicking Phantom_Hoover
15:06:49 <esomimic> elliott: something called try to won you don't think sense drama.
15:06:52 <esomimic> elliott: i assume it call/cc interaction so is why would you mean the computable reals, his tower.
15:06:58 <esomimic> elliott: he's arguing; internet talk about 5 'austerity come to think of it.'
15:07:04 <elliott> Phantom_Hoover: this isnt very good
15:07:07 <elliott> can you be more entertaining in future
15:08:20 <esomimic> elliott: okay, mimicking cpressey
15:08:26 <elliott> fizzie: does this include his seminal work as ZOMGMODULES?
15:09:12 <elliott> fizzie: can I have at least 20 refunds
15:09:17 <elliott> he was catseye like half the time!
15:09:26 <fizzie> You can have twenty times zero monies.
15:09:39 <elliott> esomimic: this better be good
15:09:39 <esomimic> elliott: yes. for indirection.... will vm i want for us and but and directory other formats on that languages have syntax channels stuff oh oh
15:09:41 <elliott> esomimic: this better be good
15:09:42 <esomimic> elliott: or if it... mostly sending you expect
15:09:42 <elliott> esomimic: this better be good
15:09:44 <elliott> esomimic: this better be good
15:09:44 <esomimic> elliott: hm can build better comparison. ok
15:09:45 <elliott> esomimic: this better be good
15:09:45 <esomimic> elliott: you have you ever get it crash why do you mean by dismissing you know. i say you have to comic strip almost always
15:09:51 <esomimic> elliott: okay, mimicking monqy
15:09:52 <elliott> esomimic: this better be good
15:09:53 <elliott> esomimic: this better be good
15:10:07 <fizzie> People are a sometimes food.
15:10:16 <esomimic> elliott: surely that, i think its do that
15:10:18 <esomimic> elliott: those the make when fuck indentation because it's not doing what i want to find it
15:10:28 <elliott> fizzie: these don't quite have fungot's charm
15:10:28 <fungot> elliott: with that in mind the following ( in dth's) off of cng, cgas
15:10:41 <Gregor> esomimic: mimic esomimic
15:10:41 <esomimic> Gregor: sorry, don't know 'esomimic' well enough
15:10:53 <fizzie> I probably should've simply used fungot's code to do the whole thing.
15:10:54 <fungot> fizzie: option as ceo of this right? -----original message----- from: steve venturatos.
15:11:04 <elliott> fizzie: IMO it should be able to take a nickname regexp and then generate a model for mimicking that on the fly
15:11:10 <elliott> so I can make hideous chimerae of people
15:11:17 <Gregor> I would be CEO of Steve Venturatos.
15:11:47 <fizzie> fungot: I suppose you're still doing the Enron thing?
15:11:47 <fungot> fizzie: will we. hollings will confer in connection with any decision i think we have two set on the language of the) lines not working on i
15:12:06 <elliott> `pastelogs steve venturatos
15:12:23 <fungot> Available: agora alice c64 ct darwin discworld enron* europarl ff7 fisher fungot homestuck ic irc iwcs jargon lovecraft nethack pa qwantz sms speeches ss wp youtube
15:12:23 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.15070
15:12:33 <elliott> fizzie: I hope you didn't update the irc style
15:12:49 <fizzie> I have not changed anything.
15:12:51 -!- DHeadshot has quit (Ping timeout: 252 seconds).
15:13:00 <elliott> you need reliable things in this e'er-changing world
15:13:26 -!- DHeadshot has joined.
15:13:27 <elliott> fizzie: so wait is that enron thing not new
15:13:40 <fungot> Selected style: enron (subset of the Enron email dataset)
15:13:46 <fungot> Phantom_Hoover: also on the price of 1 ( limit 2). i
15:13:49 <nooga> i'm reading SICP again and playing with a scheme interpreter
15:14:16 <elliott> 23:41:27: <Vorpal> fizzie, Enron?
15:14:18 <elliott> 23:41:57: <Vorpal> fizzie, the electricity company?
15:14:32 <nooga> uh, what a nice way procrastinate
15:14:53 <mroman> any python guy in here?
15:15:10 <elliott> if "python guy" means know python then yes. if it means like python then no
15:15:31 <mroman> is there a nice way to convert a string to a list of char?
15:15:36 <mroman> other than map(lambda a:a,str)
15:15:52 <fizzie> elliott: Oh. Well, I mean, I have not changed any of the things that already existed. I did add that.
15:16:34 <fizzie> In most cases you can just enumerate over the string, though.
15:20:48 <nooga> because i don't like it
15:20:51 -!- ais523 has joined.
15:20:59 <elliott> http://www.cs.cmu.edu/~enron/DELETIONS.txt
15:21:01 <mroman> God I hate file modes.
15:21:06 <elliott> this just makes me intensely curious as to what those emails were
15:21:15 <mroman> Is there a shut up just give me a file handle open
15:22:13 <ais523> lambdabot: what about yesterday? I wasn't here then either
15:22:20 <mroman> r+ seems to do the trick.
15:22:37 <fizzie> elliott: http://sprunge.us/MIVE is one of them.
15:24:16 <fizzie> elliott: sent/205 seems reasonably similar, except to some other girl.
15:24:49 <elliott> fizzie: I sense a pattern.
15:25:23 <fizzie> For some reason, gay-r/sent/74. is just an email about scheduling a conference call discussing someone's termination notice.
15:25:35 <fizzie> (Don't know why to remove that.)
15:26:35 <elliott> fizzie: I guess the gay-r ones are still available for download, since it says removed 2011 but the latest archive is from 2009.
15:26:41 <elliott> So the skilling-j one is the real mystery.
15:26:57 -!- TeruFSX has joined.
15:26:58 <mroman> and is there a better way than reduce(str.__add__,foo)
15:28:20 <fizzie> elliott: The copy at work is missing skilling-j/1584 too, so can't help there.
15:28:59 <fizzie> elliott: And the "Agust 21, 2009" download link is actually to enron_mail_20110402.tgz so I think it actually doesn't have the gay-r ones.
15:29:34 <mroman> My python is so rusty.
15:29:54 <elliott> fizzie: So that's what Agust means.
15:29:56 <mroman> But I'm doin' Array Manipulation Stuff and I'm even poorer at that with haskell than python
15:31:14 -!- carado has quit (Ping timeout: 256 seconds).
15:31:28 <ais523> "Triforce is identical to [[brainfuck]] while purely using triangles, and with the addition of an operator that resets a memory cell to 0. It is not yet possible to input and store a character."
15:31:42 <ais523> this… somehow manages to do worse than the average pointless BF derivative
15:31:49 <fizzie> But it uses triangles!
15:31:52 <elliott> ais523: that langauge was made as a threat, AIUI
15:32:01 <ais523> elliott: who was it threatening?
15:32:19 <elliott> 17:15:49: <Halite> SaltScript
15:32:19 <Taneb> ais523, when Phantom_Hoover tried to write a Tumblr post about it, my computer crashed
15:32:19 <elliott> 17:16:34: <Halite> what should I add to SaltScript
15:32:19 <elliott> 17:18:59: <Halite> ok, I'll make another BF-like language if you don't listen to those developing better languages
15:32:34 <elliott> (Phantom_Hoover or someone told him not to make a BF derivative)
15:32:43 <Phantom_Hoover> Taneb, so are my thoughts on it never going to come to light
15:33:12 <ais523> Taneb: you own PH's Tumblr, right?
15:33:35 <Taneb> Phantom_Hoover, we shall see
15:33:38 <Phantom_Hoover> is 'irthening bad' never going to be given the rightful place in a sentence that it deserves
15:33:56 <Taneb> I was thinking of using disappointirth
15:34:26 <elliott> ais523: btw, you should bug me to update the esowiki soon
15:34:33 <elliott> it's several updates behind and some of them are even security-related
15:35:15 <elliott> ais523: well, I don't feel like doing it now :)
15:35:31 <ais523> so next time I remember and you seem like you're in a mood to do something major?
15:35:42 <elliott> no, I won't feel like doing it later, either
15:35:51 <elliott> which is why I'm telling you to bug me, see
15:36:02 <elliott> but I don't want you to bug me /now/, because I don't want to do it
15:36:15 <Taneb> elliott, doo eeeeeeet
15:37:46 <Taneb> I will when elliott updates the wiki!
15:48:51 <mroman> http://codepad.org/nyIPsxbn
15:49:45 -!- SDr has joined.
15:52:03 <mroman> Threads must manipulate each other to do stuff
15:53:02 -!- ais523_ has joined.
15:53:28 <ais523_> it's nice to have an OS such that your computer can overheat, crash, and restart before you ping out from IRC
15:53:30 -!- ais523 has quit (Disconnected by services).
15:53:32 -!- ais523_ has changed nick to ais523.
15:57:35 -!- DHeadshot has quit (Ping timeout: 251 seconds).
16:07:31 -!- oerjan has joined.
16:23:16 -!- ogrom has quit (Quit: Left).
16:29:43 -!- efm has quit.
16:29:52 -!- efm has joined.
16:33:34 <oerjan> elliott: update the wiki O KAY
16:34:05 <oerjan> O KAY if you don't mind it breaking all over the place
16:35:45 <oerjan> also, why are there 83 people here
16:36:06 <Taneb> Because I have been ADVERTISING
16:36:43 <Taneb> (I have not been advertising)
16:36:54 <Arc_Koen> hello guys, I'm here thanks to Taneb!
16:36:56 <Taneb> (well, I have, but I've caused no-one to stick)
16:37:03 <Taneb> Arc_Koen, you are!?
16:37:07 <Arc_Koen> and I have a great idea for a new language! it has 8 symbols!
16:37:24 <Taneb> Is it a MIBBLLII derivative?
16:37:38 <oerjan> Arc_Koen: i shall assume it's an underload derivative without ) hth
16:37:58 <Arc_Koen> Taneb: do you really think I understand BCKW??
16:38:09 <Gregor> Is there a BF alternative for which the symbols are a, A, á, à, Á, À, ą and Ą?
16:38:17 <Taneb> I find BCKW easier than SKI
16:38:46 <Gregor> Or maybe a, á, à, ä, ą, ā, ã and å.
16:39:14 <Arc_Koen> Gregor: I believe you should make one
16:40:36 -!- sebbu has quit (Ping timeout: 248 seconds).
16:40:51 -!- sebbu has joined.
16:41:21 <Gregor> I believe I should not.
16:41:39 <Arc_Koen> some would argue you already did :(
16:43:03 <elliott> Gregor: please do, I have a use for BF derivatives
16:43:27 <Gregor> Arc_Koen: I haven't specified the mapping.
16:43:47 <elliott> Gregor: so you specified as many BF derivatives as there are possible mappings
16:43:57 <Arc_Koen> elliott: do you wedge unstable furnitures with bf derivatives??
16:44:25 <Gregor> Oh wow, I specified 8 factorial BF derivatives without even realizing it!
16:44:34 <Arc_Koen> Gregor: you specified 40320 mappings!
16:45:03 <Arc_Koen> that's gotta set a record - I don't believe anyone else has accidentally specified so many bf derivatives
16:45:21 <Arc_Koen> (I wish I hadn't been so slow)
16:45:42 <Gregor> Actually, every sequence of unique characters in existence specifies mappings!
16:46:02 <Gregor> The alphabet specifies 26*25*24*23*22*21*20*19 mappings
16:46:28 <Arc_Koen> please tell us it's no longer accidental
16:46:43 <Taneb> > product [109990..110000]
16:46:45 <lambdabot> 28516904589671371494881311400913125079881049855168000000
16:46:51 <Taneb> That's a lot of bf derivatives
16:46:58 <Taneb> (approx number of characters in Unicode)
16:47:08 -!- mekeor has joined.
16:47:16 <Gregor> As well as a featured language, the site should have a Brainfuck derivative of the day X-D
16:47:19 <Arc_Koen> and that's assuming we only use one character
16:47:58 <Gregor> It could even be automatic. Just choose at random from the BF derivatives category and barf it onto the main page.
16:48:03 <Gregor> We could call it the box of shame!
16:48:06 <Taneb> If you also have some encryption scheme such that each character is encoded by a different cipher...
16:49:23 <Taneb> Any sufficiently long block of text could be a programming encoding any algorithm
16:51:23 <ais523> hey, this has already been done better
16:51:37 <ais523> something like TMMDLPTEALPAITAFNFAL
16:51:47 <ais523> (that was from memory, I probably don't have it exactly right)
16:51:53 <ais523> also it's automatically better due to not being a BF derivative
16:52:08 <ais523> and I looked it up, it's TMMLPTEALPAITAFNFAL
16:52:20 <ais523> I was only one letter off :)
16:52:31 <tswett> I just defined |V| programming languages.
16:54:43 <tswett> I defined so many programming languages that ZFC suddenly became inconsistent with reality.
16:55:04 -!- DHeadshot has joined.
16:56:41 -!- TeruFSX has quit (Ping timeout: 245 seconds).
17:00:49 -!- carado has joined.
17:03:20 <Taneb> Well, that'll explain why my graphics card hasn't arrived yet
17:03:24 <Taneb> It just got dispatched
17:07:45 -!- nooodl has quit (Ping timeout: 252 seconds).
17:15:41 <Taneb> (a -> b) -> (a -> c) -> (a -> b & c)
17:15:57 <Taneb> (a -> c) -> (b -> c) -> (a | b -> c)
17:20:59 <oerjan> Taneb: you might want to look up sequent calculus...
17:21:17 <Taneb> At the moment I'm looking up accents of the english language
17:21:24 <Taneb> I'll add sequent calculus to the queue
17:21:32 <FreeFull> Time for me to check pipes out
17:21:53 <Taneb> Unix pipes, Haskell pipes, or plumbing pipes?
17:22:57 <oerjan> Taneb: hm you weren't brought to accents from wikipedia's picture of the day by any chance?
17:23:14 <Taneb> I was talking to someone in the American South about it
17:23:31 <oerjan> because i just had "Transatlantic accent" open
17:25:44 <FreeFull> I've been pretty familiar with unix pipes for a while now =P
17:33:14 <kmc> hash pipes
17:37:17 <Taneb> Cigars are much nicer
17:37:27 <Taneb> Actually, I've never smoken hash or a pipe
17:37:53 <Taneb> Cigarettes are nasty
17:41:30 <Phantom_Hoover> a boatload of crappy side effects for... getting relaxed?
17:43:00 -!- Bike has joined.
17:47:03 -!- TeruFSX has joined.
17:47:08 -!- TeruFSX2 has joined.
17:51:06 -!- const has changed nick to constant.
17:52:04 -!- Slereah has quit (Ping timeout: 256 seconds).
17:52:23 -!- Slereah_ has joined.
17:53:28 <ion> Finnish LotR. I hadn’t even heard about it existing. :-D http://youtu.be/Koj0V7G46fs?t=5m52s
17:57:31 <Taneb> I read that as "Finished [reading] LotR"
17:57:43 <Taneb> Which changes the meaning entirely
18:01:21 <fizzie> ion: What, you didn't know about that show?
18:01:54 <fizzie> It has an impressive opening.
18:07:15 -!- nooodl has joined.
18:14:03 -!- mp98 has joined.
18:15:02 -!- mp98 has quit (Remote host closed the connection).
18:31:42 -!- Nisstyre-laptop has quit (Read error: Connection reset by peer).
18:45:42 -!- Nisstyre-laptop has joined.
18:46:04 -!- impomatic has joined.
18:46:51 <Gregor> ion: I would watch that if the subtitles weren't in Comic Sans.
18:47:34 <oerjan> idea: comic sans for spoiler tags
18:57:17 -!- dessos has left.
19:06:12 -!- Taneb has quit (Ping timeout: 256 seconds).
19:22:34 -!- Gregor has set topic: Y'ALL DON'T NEED NO PRIVATE KEYS TO PUSH TO THIS REPO | http://codu.org/logs/_esoteric/.
19:40:32 <lambdabot> BS.foldr' (imported from Data.ByteString),
19:42:38 -!- monqy has joined.
19:47:22 -!- TeruFSX2 has quit (Ping timeout: 256 seconds).
19:47:22 -!- TeruFSX has quit (Ping timeout: 256 seconds).
20:02:33 -!- AnotherTest has joined.
20:03:14 -!- Taneb has joined.
20:03:29 <nooga> is it bad that I don't understand a thing from this: http://www.willamette.edu/~fruehr/haskell/evolution.html#categorical ?
20:03:54 -!- epicmonkey has quit (Read error: Operation timed out).
20:05:11 <AnotherTest> actually I don't understand a thing from it either
20:05:37 <monqy> what's not to understand about it
20:06:03 <AnotherTest> Well, as a junior haskell programmer (probably not even) it's pretty hard
20:06:17 <AnotherTest> I think the idea actually is that you don't understand it, until you are at the said level
20:07:08 <lambdabot> No instance for (GHC.Show.Show (t0 -> t0))
20:07:15 <lambdabot> ASetter s t a (Maybe b) -> b -> s -> t
20:07:32 <kmc> AnotherTest: nah, look at the last one
20:07:49 <nooga> arising from a use of `M26650
20:07:57 <nooga> yeah, that's so meaningful
20:08:01 <AnotherTest> but I think someone not knowing haskell also does
20:09:04 <elliott> > M.empty & ix "abc" ?~ Just 123
20:09:07 <nooga> haskell is weird and its algebraic type system is even weirder
20:09:20 <elliott> > M. & at "abc" ?~ Nothing
20:09:22 <lambdabot> <hint>:1:4: parse error on input `&'
20:09:22 <elliott> > M.empty & at "abc" ?~ Nothing
20:09:25 <Taneb> > "hello" & _head ?~ 'Y' & id <>~ "w"
20:09:26 <lambdabot> Couldn't match type `GHC.Types.Char'
20:09:37 <elliott> > M.singleton "abc" 123 & at "abc" ?~ 123
20:09:38 <AnotherTest> nooga: I guess some stuff is weirder than Haskell though
20:09:43 <elliott> sometimes i forget i'm an idiot
20:10:17 <nooga> elliott: good that you remember now ;>
20:10:26 <oerjan> on this "evolution" thing, i'm disappointed by the "teaching Haskell to freshmen" comment on the tenured professor. without it, it would be a better joke of "no longer has to be pretentious", i think.
20:11:06 <monqy> nooga: um i think you'll find that it's things that aren't haskell that are weird
20:11:37 <nooga> i didn't say that haskell is the weirdest thing i saw
20:12:03 <elliott> oerjan: that sounds like wishful thinking :P
20:12:38 <AnotherTest> like in C++, you can write function() try { } catch { }
20:12:53 <AnotherTest> (note you'd have to write catch(something))
20:13:08 <nooga> C++ does not exist for me :D
20:13:27 <oerjan> god dammit i just start looking at something slightly brain-requiring (lenses), and like clockwork the housemate makes a phone call.
20:13:50 <nooga> yeah, because C++ is almost that bad, AnotherTest
20:13:56 -!- DHeadshot has quit (Ping timeout: 248 seconds).
20:14:15 <Taneb> oerjan, you are the only man alive who can program in Fueue, you should be able to understand lenses
20:16:24 <elliott> oerjan: just play edwardk's lens video really loud
20:16:25 <oerjan> i can. but the house has to be quiet.
20:16:47 <kmc> AnotherTest: huh? can you elaborate on that syntax?
20:17:20 <oerjan> <elliott> oerjan: that sounds like wishful thinking :P <-- i thought that was sort of the point of tenure, at least the american version...
20:17:41 <AnotherTest> kmc: It's a shorthand for function { try { } catch(something) { } }
20:17:54 <elliott> oerjan: I think I was trying to make some joke about you wanting to believe you could stop being pretentious with tenure or something
20:18:15 <AnotherTest> that only happens when it's in a constructor
20:18:34 <olsner> oerjan: the house has to be quiet?
20:18:36 <AnotherTest> so for a normal function, it will return when going out of scope, and for a constructor it will rethrow
20:18:58 <oerjan> olsner: if i am to think properly, yes.
20:19:19 <nooga> try that when living with fiance
20:22:32 <olsner> oerjan: http://hackage.haskell.org/packages/archive/lens/3.8.5/doc/html/Control-Lens-Internal-Bazaar.html
20:23:28 <olsner> I'm sure that's the bestest starting point for learning about lenses
20:24:05 <oerjan> i'm sure you are joking, as the alternative would require me to kill you now.
20:24:19 <olsner> I'm glad to see that Bazaar has a Bizarre instance nowadays
20:24:27 <elliott> that one is edwardk's fault
20:24:47 <Phantom_Hoover> what does that class <stuff> | <other stuff> thing mean
20:24:58 <Phantom_Hoover> i tried finding out the other day and failed miserably
20:25:08 <elliott> Phantom_Hoover: functional dependencies
20:25:12 <olsner> elliott: the type/typeclass or the lens library?
20:25:15 <elliott> a b c -> d e f mean that the choice of a,b,c determine d,e,f
20:25:44 <nooga> i don't even try to visualize what are you talking about
20:32:47 <nooga> functional equivalent of reference
20:34:23 * oerjan suddenly realizes the pun in "Bazaar". i guess because it was written out explicitly in that link.
20:34:44 -!- mekeor has quit (Quit: goo nigh).
20:35:08 <lambdabot> s -> Getting (First a) s t a b -> Maybe a
20:35:39 <tswett> Seriously, what is that. What are Getting and First?
20:37:35 <tswett> All right, it's the "Maybe monoid returning the leftmost non-Nothing value".
20:38:06 <tswett> (Look, a functor from Set to Mon!)
20:38:27 <oerjan> aka "why the heck isn't that the Monoid instance for Maybe itself, anyway?"
20:38:52 <tswett> Presumably because there's also Last. Not that that's a great reason.
20:39:05 <tswett> Hey, I bet there's a natrual transformation from First to Last.
20:39:14 <lambdabot> (Foldable t, Monoid m) => t m -> m
20:39:34 <shachaf> tswett: Because the correct Monoid instance is Semigroup a => Monoid (Maybe a)
20:39:59 <shachaf> Anyway: type Getting r s t a b -> (a -> Const r b) -> s -> Const r t
20:40:45 <oerjan> Const r b is r, wrapped.
20:41:13 <tswett> So it's equivalent to r.
20:41:24 <tswett> Isomorphic, even, isn't it...
20:41:29 <oerjan> it's used here because it makes Getting a special case of the type of Lens
20:41:33 <Taneb> > mconcat [Just "hello ", Nothing, Just "monoids!"]
20:41:58 <tswett> I'm starting to think something silly may be going on.
20:42:07 <oerjan> (also, because it's how you use a Lens as a Getter, which is sort of equivalent)
20:43:00 <elliott> 20:34:23 * oerjan suddenly realizes the pun in "Bazaar". i guess because it was written out explicitly in that link.
20:43:11 <elliott> oerjan: lens' puns aren't quite... puns
20:44:16 <elliott> oerjan: well there is also Market and Exchange
20:45:27 <Taneb> There's a market in Hexham and there used to be an exchange but now it's a cafe thing
20:45:47 <tswett> Is there also a type called Cafe?
20:46:17 <Bike> the cafe s t a b
20:46:17 <lambdabot> Bike: You have 2 new messages. '/msg lambdabot @messages' to read them.
20:48:23 <lambdabot> Cons (->) f s s a a => LensLike' f s a
20:48:46 <Bike> That's a pretty good one, oerjan.
20:49:32 * oerjan isn't sure whether Bike is referring to _head or his message
20:53:48 <Taneb> _head is just a traversal
20:54:12 <Taneb> Which means (^.) needs a monoid
20:54:17 <Taneb> :t [] ^. _head :: String
20:54:21 <Taneb> > [] ^. _head :: String
20:54:26 <Taneb> > [] ^. _head :: Int
20:54:27 <lambdabot> No instance for (Data.Monoid.Monoid GHC.Types.Int)
20:56:34 <lambdabot> (Applicative f, Cons (->) f t t b b) => t -> f t
20:56:54 <oerjan> > _head %%~ pure $ [1,2,3]
20:56:56 <lambdabot> No instance for (GHC.Show.Show (f0 [t0]))
20:57:07 <oerjan> > _head %%~ (:[]) $ [1,2,3]
20:58:04 <oerjan> oh right, that "always returns the same number of elements" thing
20:58:46 <lambdabot> Overloading p q f s t a b -> p a (f b) -> q s (f t)
20:59:09 -!- madbr has joined.
20:59:12 <Taneb> I'm pretty sure that's just id
20:59:18 <elliott> oerjan: (%%~) = traverseOf = id
20:59:41 <Taneb> > _head (:[]) "hello"
20:59:42 <elliott> > traverse %%~ (:[]) $ [1,2,3]
20:59:47 <elliott> > traverse %%~ (:[]) $ [1,2,3]
21:00:03 <Sgeo_> "This paper also shows how Scalas type system conspires with implicits to enable, and even surpass, many common extensions of the Haskell type class system, making Scala ideally suited for generic programming in the large.
21:00:08 <Sgeo_> I should probably read this
21:00:14 <oerjan> elliott: thank you for answering a question i'm not caring about
21:00:24 <elliott> Sgeo_: ask monqy about scala's many benefits over haskell
21:00:34 <Sgeo_> http://lambda-the-ultimate.org/node/4039
21:00:35 <elliott> oerjan: well I don't see how this is related to "always returns the same number of elements"?? i'm confused
21:00:35 <monqy> fuck scala fuck scala fuck scala
21:00:45 <Sgeo_> monqy, um, hm? Why?
21:00:55 <madbr> does anyone know if it's easy in a C copiler to reorder the operations to group as much memory read/writes together as possible?? like, group all the math operations and conditionals together, and then once in a while do a whole group of just memory accesses
21:00:55 <elliott> its just too powerful for monqy
21:00:59 <elliott> he needs good old typeclasses
21:02:48 <monqy> Sgeo_: have you ever used scala
21:02:52 <madbr> I'm working on a RISC cpu design with no cache(!)
21:03:03 <Sgeo_> I've read some Scala documentation, never really used Scala though
21:03:06 <monqy> im sure you havent, because if you had, you'd know the True Heck that is scala
21:03:22 <Sgeo_> What, the piles and piles of syntax?
21:03:36 <elliott> sorta like how esolangs are all about adding weird syntax
21:03:54 <lambdabot> s -> Getting (First a) s t a b -> Maybe a
21:04:05 <Sgeo_> I like that there's special syntax for delimited continuations.
21:04:12 <elliott> haskell has that too its called do notation
21:04:38 <madbr> and the problem is that when you have to do memory accesses, they have to do them on another mem page than where the instructions are, which has a penalty (3 cycles RAS/CAS access instead of 1 cycle CAS only), and then once the read/write is done, you have to load up the next opcode, which is also on another memory page and also gets the 3 cycle instead of 1 penalty
21:04:41 <Sgeo_> do notation is ugly compared to shift/reset imo
21:04:46 <oerjan> > foldMapOf [1,2,3] (:[]) _head
21:04:48 <lambdabot> Couldn't match expected type `p0 a0 (Control.Lens.Internal.Getter.Accessor
21:04:59 <Sgeo_> https://github.com/urso/embeddedmonads
21:05:03 <lambdabot> Profunctor p => Accessing p r s t a b -> p a r -> s -> r
21:05:08 <lambdabot> Profunctor p => Accessing p r s t a b -> p a r -> s -> r
21:05:27 <FreeFull> I don't get those types with profunctors
21:05:29 <oerjan> > foldMapOf _head (:[]) [1,2,3]
21:05:38 <madbr> so if I did just a standard load/store RISC cpu, the alu/jmp ops would take 1 cycle but the memory IO would take a massive 6 cycles
21:05:55 <FreeFull> > foldMapOf _tail (:[]) [3,4,5]
21:06:20 <lambdabot> Could not deduce (Control.Lens.Tuple.Field1 [t1] t0 a b0)
21:07:40 <lambdabot> (Applicative f, Indexable Int p) => LensLike (Control.Lens.Internal.Indexed.Indexing f) s t a a -> Int -> p a (f a) -> s -> f t
21:08:26 <FreeFull> What hapenned to simple types like [a] -> Int -> a
21:08:57 <elliott> it's ghc's fault that the types look complex because it likes to reduce away type synonyms for no reason
21:09:14 <oerjan> > foldMapOf (elementOf id 2) (:[]) [1,2,3]
21:09:55 <oerjan> oh wait it doesn't do anything to the type...
21:09:57 -!- copumpkin has quit (Ping timeout: 252 seconds).
21:10:37 -!- copumpkin has joined.
21:11:04 <Sgeo_> monqy, what's wrong with Scala?
21:11:38 <monqy> oh god so many things
21:11:40 <monqy> i suggest you try it for yourself
21:11:40 <oerjan> > foldMapOf (elementOf traverse 2) (:[]) [1,2,3]
21:11:54 <elliott> monqy: that doesnt really fit into the sgeo language workflow
21:11:54 <kmc> scala's type system is the illegitimate child of haskell's and java's
21:13:12 <oerjan> > foldMapOf (element 2) (:[]) [1,2,3] -- had an abbreviation
21:13:28 <Sgeo_> elliott, oh come on. I've written real Haskell and Tcl programs.
21:14:01 <Sgeo_> Should have used Python or Haskell for that one though
21:14:34 <oerjan> > foldMapOf (elements even) (:[]) ['a'..'z']
21:14:40 <elliott> well 3 is a smaller number than a really big number
21:14:53 <Bike> Also smaller than four.
21:15:06 <shachaf> Bike: I'm not sure about that.
21:15:17 <shachaf> > comparing length "three" "four"
21:15:33 <Bike> I'm talking about four, not "four".
21:15:49 <elliott> > comparing length "three" four
21:15:52 <lambdabot> Perhaps you meant `Data.Traversable.for' (imported fr...
21:15:53 <elliott> > comparing length "three" four
21:15:57 <lambdabot> Perhaps you meant `Data.Traversable.for' (imported fr...
21:16:03 <elliott> 21:15:46 <elliott> @let "four" = four
21:16:04 <elliott> 21:15:48 <lambdabot> <local>:5:10:
21:16:04 <elliott> 21:15:48 <lambdabot> Not in scope: `four'
21:16:04 <elliott> 21:15:48 <lambdabot> Perhaps you meant `Data.Traversable.for' (imported from Data.Traversable)
21:16:16 <Bike> So they're incomparable.
21:16:27 <Bike> lambdabot is inconsistent.
21:16:38 <shachaf> Inconsistent *and* incomplete.
21:16:44 <shachaf> What kind of deal did we get here, anyway?
21:17:18 <oerjan> > elements even %~ toUpper $ ['a'..'z']
21:17:57 <nooodl> that reminds me of `WeLcOmE
21:18:34 <HackEgo> TeSt: 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.)
21:19:05 <shachaf> > elements even %~ toUpper $ "irc.dahl.net"
21:19:25 <oerjan> wait, WeLcOmE also downcases the others, i think
21:19:51 <nooodl> > elements odd %~ toUpper $ ['a'..'z']
21:20:13 <elliott> > traversed %@~ if' even toUpper toLower "ABCdef"
21:20:15 <lambdabot> Perhaps you meant f' (imported from Debug.SimpleReflect)
21:20:23 <elliott> > traversed %@~ (\i -> if even i then toUpper else toLower) "ABCdef"
21:20:25 <lambdabot> Couldn't match expected type `a0 -> b0'
21:20:28 <elliott> > traversed %@~ (\i -> if even i then toUpper else toLower) $ "ABCdef"
21:22:02 <nooodl> > traversed %@~ (cycle [toUpper,toLower] !!) $ "ABCdef"
21:22:35 <oerjan> i have a hunch zipWith ($) is more efficient in that case
21:22:50 <shachaf> Sadly list indexing is really expensive.
21:23:00 <shachaf> It allocates a bunch of tuples and annoying things like that.
21:23:11 <elliott> should make it use unboxed tuples
21:23:22 <elliott> I suspect a simple use of traversed + (%@~) gets inlined
21:23:35 <shachaf> Yes, but traversed is expensive even if you don't use the index.
21:24:15 <elliott> I mean I suspect the allocation gets inlined away.
21:24:54 <shachaf> It allocates a tuple for every element of the list.
21:25:07 <shachaf> Remember, indexing is implemented by traversing it with something like State Int.
21:26:07 <elliott> I mean when you're monomorphic on the container type.
21:27:38 <shachaf> What does it change when you're monomorphic over the container type?
21:27:57 <elliott> Why does (traversed %@~ (\i -> if even i then toUpper else toLower)) have to allocate any tuples?
21:28:14 <elliott> If you use it as String -> String.
21:28:18 <elliott> As in fixing the Traversable instance.
21:29:11 <shachaf> It's still going to traverse it with State Int to figure out the indices.
21:29:20 <elliott> But traverse can be inlined.
21:30:18 <shachaf> Note that you have to traverse it with *lazy* State.
21:30:32 <shachaf> You can't use an unboxed tuple there because then it fails with infinite lists and such.
21:35:08 -!- zzo38 has joined.
21:36:20 -!- AnotherTest has quit (Quit: Leaving.).
21:43:50 <kmc> http://www.reddit.com/r/IAmA/comments/18pik4/i_am_astronaut_chris_hadfield_currently_orbiting/
21:55:32 <olsner> how funny would it be if it turned out we have people in space
21:57:58 -!- Taneb has quit (Read error: Connection reset by peer).
22:00:17 <lambdabot> (MonadState (b, t) m, Cons Control.Lens.Internal.Review.Reviewed Identity s t a b) => m t
22:16:19 <lambdabot> Functor f => f (a -> b) -> a -> f b
22:17:07 <FreeFull> That got printed out weirdly o.o
22:17:14 -!- EgoBot has quit (Read error: Connection reset by peer).
22:17:31 <kmc> Phantom_Hoover: it would be funny if he made it sound like the space station was under attack by alien monsters
22:17:43 -!- EgoBot has joined.
22:18:11 <kmc> maybe i should watch that terrible movie where (spoiler alert) it turns out that moon rocks are alien monsters
22:18:36 -!- Nisstyre-laptop has quit (Quit: Leaving).
22:18:38 <FreeFull> I like how the type basically tells you what the function does
22:19:16 <kmc> i think in this case it actually does tell you
22:19:27 <lambdabot> <hint>:1:3: parse error on input `]'
22:19:35 <kmc> that is, there's only one function of that type obeying the laws
22:19:41 <kmc> that happens sometimes with functors
22:21:06 -!- Nisstyre-laptop has joined.
22:21:07 <kmc> > ((++) ?? "foo") "bar"
22:21:21 <kmc> (??) is a generalization of flip :)
22:21:41 <lambdabot> Functor f => f (a -> b) -> a -> f b
22:22:57 <kmc> hm i shouldn't have said "obeying the laws" because that's the job of the Functor instance and not the function using Functor
22:23:26 <kmc> perhaps I'm thinking of free theorems and the fact that they only work on law-abiding instances
22:25:14 <FreeFull> (??) feels to me like something from applicative
22:28:20 -!- aloril has quit (Ping timeout: 276 seconds).
22:29:38 <FreeFull> It happens to be flip for functions
22:30:34 <Phantom_Hoover> i was trying it with (->) and getting immensely confused
22:31:42 <FreeFull> I wonder how it works with ((,) a)
22:32:43 <elliott> (c, a -> b) -> a -> (c, b)
22:33:53 <shachaf> kmc: Well, you are assuming the laws when you say that.
22:34:05 <shachaf> E.g. without the laws, you could fmap id a few times and get a different result.
22:34:06 -!- aloril has joined.
22:35:25 <elliott> shachaf: You mean "law" singular.
22:35:51 <shachaf> I mean laws. Though you do get the other one for free here.
22:36:08 -!- madbr has quit (Quit: Radiateur).
22:36:21 -!- Slereah_ has changed nick to Slereah.
22:36:25 <elliott> I think there are many laws you get for free that we just take for granted.
22:36:30 <elliott> (So we should probably just take them all for granted.)
22:37:20 <shachaf> lambdabot doesn't like you.
22:37:58 <oerjan> > ifoldMap replicate "testing"
22:38:07 <oerjan> > ifoldMap replicate "testing"
22:38:25 <FreeFull> I think lambdabot is exploding
22:38:28 * oerjan hits lambdabot with the saucepan ===\__/
22:42:50 <elliott> apparently it is lispy's fault
22:42:55 <elliott> all the cpu is used up by ld
22:43:02 -!- carado has quit (Ping timeout: 256 seconds).
22:47:16 <oerjan> > ifoldMap replicate "testing"
22:47:45 <lambdabot> (Monoid m, FoldableWithIndex i f) => (i -> a -> m) -> f a -> m
22:52:13 <oerjan> > imapM replicate "abcd"
22:52:27 <oerjan> > imapM (replicate.succ) "abcd"
22:52:30 <lambdabot> ["abcd","abcd","abcd","abcd","abcd","abcd","abcd","abcd","abcd","abcd","abc...
23:01:02 <tswett> I wonder how difficult sustained terrestrial microgravity would be.
23:01:28 <tswett> I.e. build an evacuated magnetic levitation tube going all the way around the earth, and accelerate a vehicle within it to the appropriate speed.
23:07:51 <tswett> Presumably, you'd want the tube to go over as little water as possible.
23:27:28 <lambdabot> Couldn't match expected type `GHC.Types.Int'
23:27:46 <lambdabot> No instance for (GHC.Show.Show (a0 -> [a0]))
23:30:03 <lambdabot> Could not deduce (GHC.Num.Num (a0 -> GHC.Types.Int -> a -> [a]))
23:31:37 <nooga> > replicate 200 $ replicate 200 "wee"
23:31:38 <lambdabot> [["wee","wee","wee","wee","wee","wee","wee","wee","wee","wee","wee","wee","...
23:32:03 <elliott> Phantom_Hoover: ps its (-1)
23:32:33 <nooga> only 384 wees to go
23:33:54 <elliott> replicate -1 2 is (replicate - 1) 2
23:34:37 <Phantom_Hoover> i'm just surprised i hadn't heard of it before, it seems Kind of a Big Deal
23:35:36 <Sgeo_> This is why J > Haskell
23:36:05 <nooga> haskell has almost no syntax
23:36:18 <FreeFull> No, haskell has plenty of syntax
23:36:35 <nooga> and it makes weird things with operators and infix context
23:37:25 <nooga> FreeFull: haskell looks the same but with indentation instead of parens
23:37:45 * Sgeo_ thinks that macros count as syntax
23:37:58 <kmc> lisp has plenty of syntax
23:38:00 <Sgeo_> Syntax that can be added, but still syntax
23:38:02 <coppro> Haskell has little syntax
23:38:07 <coppro> compared to most language
23:38:09 <kmc> is it (let (x 3) (y 4) ...) or is it (let ((x 3) (y 4)) ...) ?
23:38:22 <kmc> hint: CL does one and Scheme does the other
23:38:32 <elliott> i thought that was a sincere question
23:38:32 <kmc> that's syntax
23:38:35 <Sgeo_> elliott, that was just to point out that that's a syntax question
23:38:44 <kmc> abstract syntax is still syntax
23:38:45 <nooga> who uses common lisp anyway
23:38:51 <kmc> that concludes your canned rant
23:38:57 <nooodl> the more i read about common lisp the more i hate it
23:39:11 <nooga> scheme looks more pleasant
23:39:12 * Sgeo_ wonders if there's anything that kmc could say that I would disagree with.
23:39:22 <kmc> nooga: obviously you're just an idiot drone because CL programmers are 50x more productive and sexy and the proof of this is hiding... right over here
23:39:37 <nooodl> scheme is more pleasant... but there's less stuff in it? "who uses lisp anyway"
23:39:40 <kmc> i hear the second life programming language is the worst
23:39:48 <Sgeo_> kmc, can't disagree with that.
23:40:02 <Phantom_Hoover> "the only good thing to come from activeworlds was shamus young"
23:40:04 <ais523> Sgeo doesn't like any programming languages
23:40:05 <kmc> i knew a guy who did behavioral econ experiments in SL and that was his conclusion
23:40:11 <ais523> nor does elliott, but possibly for different reasons?
23:40:34 <nooodl> i think elliott likes haskell!!
23:40:37 <elliott> sometimes I like some programming languages
23:40:43 <FreeFull> There is no programming language that has 0 syntax
23:40:53 <kmc> if lisp has almost no syntax then lisp is just a language for specifying trees, not a language for specifying programs
23:40:54 <Sgeo_> FreeFull, sure there is? Well, hmm.
23:41:06 <Sgeo_> The language where every program means do nothing.
23:41:28 <Sgeo_> All possible strings are legal programs in this language.
23:41:31 <nooodl> then the syntax is "every program does nothing"
23:41:31 <Sgeo_> Or does that count as syntax?
23:42:02 <Phantom_Hoover> nooodl, how is that different from a language with 0 syntax
23:42:14 <kmc> i'm not going to claim that there's a perfectly sharp line between syntax and semantics, but i think any definition of "semantics" which includes the number of parens in let is designed for winning arguments rather than understanding things
23:42:47 <Sgeo_> Esolang idea: Esolang where that sort of thing is... now I'm confused
23:42:56 <nooga> i'd like a language with minimal syntax that could change itself freely in runtime
23:42:59 <Sgeo_> Something to push it into more distinctly semantic territory... somehow
23:43:08 <Sgeo_> nooga, I've seen something like that
23:43:08 <kmc> i think i prefer scheme's way, because a) it seems more regular, and b) it allows an implicit begin/progn
23:43:11 <kmc> nooga: TeX
23:43:20 <nooga> tex is distorted lisp
23:43:49 <FreeFull> TeX is a language where a string can do everything
23:43:52 <shachaf> kmc: Which one of those is supposed to be Scheme?
23:44:00 <shachaf> It looks to me like it's the latter for both of them.
23:44:01 <kmc> shachaf: (let ((x 3) (y 4)) ...)
23:44:05 <shachaf> (I know that's beside the point.)
23:44:09 <elliott> kmc: it's interesting that scheme has the implicit progn, for being the less imperative language
23:44:11 <kmc> sorry, "both of them"?
23:44:23 <nooga> Sgeo_: i think that Ian Piumarta showcased something like that
23:44:28 <shachaf> Both CL and Scheme use the latter
23:44:47 <kmc> elliott: yeah although i'm not sure if scheme is "really" less imperative or if it's just that most people who have learned any Scheme learned it in intro CS while most people who have learned CL have written a "real" program in it
23:45:05 <kmc> oh i may be mistaken
23:45:16 <kmc> i think i'm thinking of clojure for the former
23:45:29 <Sgeo_> Clojure is (let [a 1 b 2] ...)
23:45:30 <kmc> oh clojure does (let [x 1 y 2] ...)
23:45:35 <kmc> even differenter
23:45:58 <Sgeo_> I kind of like it, the [] is expected to be used in different places and with a different vague meaning than ()
23:45:59 <nooodl> Phantom_Hoover: well, it's a rule that tells you something about which programs are valid
23:46:15 <kmc> Sgeo_: is the difference observable past the reader stage?
23:46:23 <shachaf> Why do Scheme/CL do the double-parentheses thing, anyway?
23:46:45 <kmc> so there's some predicate which tells you whether a given quoted s-expr was made with [] or ()?
23:47:05 <elliott> it's gross to not have two parens
23:47:10 <elliott> you're sacrificing structure
23:47:15 <elliott> and making it implicit instead
23:47:22 <nooodl> a language with 0 syntax would be one where valid programs don't exist at all, maybe. but arguably "valid programs don't exist" is also a syntax rule
23:47:31 <Sgeo_> kmc, yes, there exists such a predicate, but the difference goes beyond that
23:47:32 <elliott> which means e.g. it's harder to see the structure from the parens (the literal whole point), indentation/alignment/etc. become less simple
23:47:40 <Sgeo_> [+ 1 2] does not do (+ 1 2)
23:47:41 <elliott> program trnasformation is harder
23:48:16 <Sgeo_> [] gets read as a vector, which is a different data structure, and there are predicates to find out exactly what data structure
23:48:49 <Sgeo_> <lazybot> ⇒ [#<core$_PLUS_ clojure.core$_PLUS_@1284047> 1 2]
23:49:30 <kmc> i read "literal white point" and was confused
23:49:50 <kmc> vectors have fast random access yeah?
23:50:09 <kmc> it seems weird to say "function argument should be specified by a structure with fast random access" solely for syntactic reasons
23:54:50 <kmc> Florida Man Too Fat For Jail
23:54:55 <kmc> guess what he was arrested for
23:55:19 <kmc> stealing food
23:57:03 <kmc> http://blogs.miaminewtimes.com/riptide/2010/04/florida_man_too_fat_for_jail.php
23:57:22 <kmc> this is courtesy https://twitter.com/_FloridaMan of course
23:57:46 <kmc> "Jolicur avoided jail because it would be too expensive to house him... The cost of bringing him to trial alone would be thousands of dollars."
23:58:21 <kmc> Florida Man Accidentally Dials 911 While Trying To Call Tech Support, Gets Busted For Marijuana Grow Operation