00:00:18 <alise_> pikhq: I want to write a compiler now.
00:00:22 <alise_> Dammit life is awesome, have you ever noticed?
00:00:30 <oerjan> i think that deutsch was only gebrochen due to a missing suffix on the gebrochen
00:00:34 <alise_> You can DO SO MANY COOL THINGS in it.
00:00:57 -!- cpressey has quit (Quit: leaving).
00:01:02 -!- augur has quit (Ping timeout: 264 seconds).
00:01:10 <pikhq> Headphones are awesome.
00:01:19 <alise_> pikhq: But LIFE moreso!
00:05:07 <alise_> pikhq: My preferred tags for an album are completely at odds with the completely official, multiply-confirmed ones. You may now maul me
00:11:18 <alise_> pikhq: I'll say in /msg since it's irrelevant and long
00:12:42 -!- alise_ has changed nick to alise.
00:13:05 <pikhq> BTW, TV rips of series that are on DVD make baby Jesus cry.
00:33:24 <oerjan> <alise> x `seq` () `seq` x _probably_ == an evaluated copy of x <-- i think that's exactly equivalent to x by definition
00:33:55 <alise> pikhq: So does Nazism.
00:33:57 <oerjan> and as usual x will only be evaluated if the whole thing is
00:35:26 <oerjan> <alise> pikhq: i don't see why rnf can't just be a -> a
00:36:13 <oerjan> hm rnf is just one of the evaluation strategies right? now what if you wanted a strategy that did _not_ evaluate... it couldn't reasonably be a -> a
00:36:43 <alise> isn't it just deepseq's thing? maybe not
00:36:54 -!- FireFly has quit (Quit: swatted to death).
00:37:04 <oerjan> i recall this from Control.Parallel.Strategies
00:38:18 <oerjan> rnf is deepseq essentially, but there are others that evaluate stuff in parallel instead
00:41:44 <alise> coppro: HURRY UP WITH THAT MIDTERM
00:41:55 <alise> oerjan: this is part of DeepSeq though in pikhq's code
00:41:58 <alise> pikhq: Make your compiler MORE AWESOME
00:42:48 <alise> coppro: should i expect an email back from the mathNEWS people or will it just appear?
00:42:51 <alise> (in the next issue)
00:42:57 <oerjan> alise: oh right rnf is a method of DeepSeq? but it's also the right type for those strategies iirc
00:42:58 <coppro> alise: one or the other
00:43:05 <alise> coppro: that's so helpful
00:43:23 <lambdabot> Source not found. Take a stress pill and think things over.
00:46:04 <oerjan> <Gregor> Apparently Civilization (the original) had barbarian diplomats. <-- currently on front page of reddit: http://i.imgur.com/kWy5z.jpg
00:46:18 <oerjan> i guess that's from the newest version
00:46:45 <Gregor> Damn those barbarian paratroopers!
00:47:35 <alise> diplomatic paratroopers
00:47:49 <alise> coppro: wow, how do you guys even have a tv show
00:47:54 <alise> for a definition of tv equal to yt
00:48:00 <alise> also you guys = most vague use of tha tterm ever
00:49:39 <alise> coppro: my current perception of it is that about 10 people go to your university and they all write mathNEWS
00:49:48 <alise> do not attempt to disillusion me of this notion
00:51:38 -!- augur has joined.
00:52:10 <pikhq> Banshee. *It fucking does video metadata*.
00:53:04 <alise> pikhq: Imagine iTunes. Now imagine transplanting iTunes onto Mono.
00:53:24 <alise> pikhq: It most likely just stores it in a database.
00:53:34 <pikhq> That is revolting.
00:53:58 <pikhq> WHY CANT THERE BE A NICE VIDEO PLAYER
00:54:07 <alise> pikhq: *cough* You know what this is leading up to...
00:54:12 <alise> LET'S INVENT OUR OWN PROGRAMMING LANGUAGE AND GUI LIBRARY
00:54:15 <alise> ALSO VIDEO PLAYER, AFTER ALL THAT
00:54:39 <alise> "The chronology is sometimes associated with Young Earth Creationism, which holds that the universe was created only a few millennia ago." Umm... Wikipedia.
00:54:42 <alise> You mean 6000 years.
00:55:06 -!- Kordalien has joined.
00:58:29 <oerjan> well do _all_ YEC agree on the precise date?
01:01:25 <pikhq> alise: Also change the entire distribution infrastructure.
01:01:41 <alise> oerjan: it's 5,000 -- 10,000 years basically
01:01:47 <alise> definitely not millennia
01:01:54 <alise> pikhq: Actually a nice GUI library for C would be nice.
01:02:02 <oerjan> um that _is_ millennia
01:02:20 <pikhq> alise: Say, make it so that all broadcast shows are sent out via an RSS feed alike, with a BitTorrent alike.
01:02:27 <pikhq> alise: With archives, of course, instantly downloadable.
01:02:35 <oerjan> millennium means 1000 years, millennia is the plural
01:02:53 <alise> pikhq: oh i thought you mean software distribution
01:03:10 <alise> iirc there was some work done on streaming bittorrent but i have no idea how that works
01:03:19 <alise> with it being out-of-order
01:04:06 <pikhq> alise: Wikimedia is currently testing it -- a combination of HTTP and BitTorrent, actually.
01:04:40 <pikhq> Use HTTP to fetch anything needed immediately, use BitTorrent to try and fetch stuff that's later in the stream.
01:04:49 <alise> pikhq: grotesque :)
01:09:55 <coppro> alise: I want to write an article, what should it be about?
01:10:15 <alise> coppro: write an errata for my errata
01:10:24 <alise> coppro: i can send you the full version
01:10:39 <alise> (note: my errata is pretty indisputably factually correct in every way that it actually tries to be factual, so this will be difficult)
01:11:04 <alise> coppro: "Actually, this IS valid Python code! Because I say so!"
01:11:04 <coppro> "errata: this article appears to be properly researched. In order to alleviate this concern, the following changes have been made"
01:11:28 <alise> coppro: I approve.
01:11:41 <alise> coppro: I have this sneaking suspicion that gmail's spam filter doesn't like my post, what with it being long and markuppy...
01:11:54 <coppro> hang on, I'll ask CorruptED to check right now
01:12:59 <alise> coppro: I am so glad the editors are corrupt.
01:13:14 <alise> coppro: I also sent a reply to that one correcting an error caused by gmail.
01:13:43 <coppro> hang on, the email account is busy
01:16:28 <alise> i am, however, insane enough to wait to see if it's received yet
01:16:31 <alise> because dammit this stuff is IMPORTANT!
01:16:51 <zeotrope> alise: what's the article about?
01:17:22 <alise> coppro: & the reply? okay.
01:17:24 <alise> now put it on the front page
01:17:55 <alise> zeotrope: Errata for the article "Python Implementation of ed" by *null, as printed in issue 114.1 of the University of Waterloo Faculty of Mathematics Student Newspaper mathNEWS.
01:18:10 <coppro> alise: can you send me a copy so that I can write errata to it?
01:18:41 <alise> coppro: certainly.
01:18:58 <alise> coppro: http://filebin.ca/wpeogf/mathnews-errata.html
01:19:13 <coppro> "Firstly, it appears that this article was written by a Brit. To correct for this, remove every usage of grammar."
01:19:18 <alise> (Yes, it's not HTML, but.)
01:19:30 <alise> coppro: the most solid point is the last one, btw
01:19:37 <alise> [[Printed in a newspaper with an incorrectly spelled name. While &mn; is an excellent newspaper, it unfortunately has a major blight against it: the incorrect spelling of its name as ``&mn;'', rather than the correct ``mathsNEWS''. This is the most severe flaw.]]
01:22:24 <alise> coppro: so yours will appear in the next issue?
01:22:30 <alise> or the same? <-- that would be super-ludicrous
01:24:20 <alise> zeotrope: bear in mind the original article is like six lines :P
01:24:30 <alise> i will squeeze every drop of blood from this stone
01:26:05 <zeotrope> how dare he fill such an esteemed paper with garbage
01:26:13 <alise> YES I KNOW the rest of it is so accurate
01:26:22 <alise> coppro: btw, i expect similar treatment of getting the gold master copy of yours before it's published
01:26:28 <alise> so i can publish my meta^2errata
01:26:39 <pikhq> alise: Y'know, fuck Hulu. It falls so short of what it could be.
01:26:49 <alise> pikhq: yeah. and fuck Hawaii too!
01:26:50 <augur> egg foo young is delicious D:
01:26:51 <pikhq> A set of RSS feeds attached to BitTorrent for me to gleefully download.
01:26:53 <alise> i leave you with this.
01:26:56 -!- alise has quit (Quit: Leaving).
01:27:45 <zeotrope> it is still technically piracy
01:28:14 <pikhq> zeotrope: Hulu is... Actually not piracy.
01:28:25 <zeotrope> I was talking about RSS + torrents
01:28:56 <zeotrope> which is definitely more convenient
01:29:27 <pikhq> Which is the point. :)
01:29:45 <zeotrope> some people frown on the piracy part
01:30:42 <zeotrope> and Hulu doesn't work outside of the USA
01:31:10 <zeotrope> so torrents are the only other option for many people
01:32:02 <zeotrope> country restricted content, what is this 1970?
01:32:48 <pikhq> As far as they're concerned, yes.
01:32:55 <pikhq> Oh, also: SCREW FLASH.
01:32:59 <pikhq> FLASH SUCKS FOR PLAYING VIDEO.
01:36:59 <pikhq> Flash sucks. Flash really, really, really sucks.
01:41:10 -!- Sgeo has changed nick to Flash.
01:41:18 -!- Flash has changed nick to Sgeo.
02:13:26 -!- iGO has quit (Ping timeout: 240 seconds).
02:45:26 <Sgeo> http://unitedcats.wordpress.com/2010/10/04/this-october-has-5-fridays-5-saturdays-and-5-sundays-all-in-one-month-it-happens-only-once-in-823-years/
02:45:35 <Sgeo> I hate humanity.
02:47:05 <Sgeo> Two of my friends "liked" a page claiming that
02:48:44 <oerjan> ok so the first part is actually correct
02:49:53 <oerjan> and the second obviously false for two different reasons i can think of
02:50:17 <Sgeo> What tipped me off was noticing that that's the same (or, at least, a superset of) months whose 1 is on a Friday and which have 31 days
02:50:36 <Sgeo> And I think you may be the second person I linked that to who didn't actually read the post
02:50:43 <oerjan> (1) it only depends on what day the first is on (2) the calendar _actually_ repeats every 400 years
02:50:53 <oerjan> oh i've started reading it
02:51:06 <oerjan> i'm just thinking about why it's false first :D
02:54:51 -!- cheater99 has quit (Ping timeout: 265 seconds).
02:56:53 <Sgeo> Well, it also depends on the month having 31 days
02:57:22 <oerjan> um i mean the year variation
02:58:15 <Sgeo> Well, depends on what day the 1st is, and whether or not it's a leap year, right?
03:00:13 <oerjan> _for october_ the leap year doesn't matter
03:00:58 <oerjan> i meant the first day of october, not of the year
03:01:00 <Sgeo> Are we talking Jan 1st's effects on October, or the day that Oct 1 is?
03:01:14 <oerjan> sheesh one really has to spell out everything here
03:09:00 -!- lament has joined.
03:11:17 <Ilari> MAJOR FAIL: (roughly translated) "The Baddie of cholesterol is LDL, sources of which for example include cheeses.".
03:12:12 <Ilari> There is only one cholesterol. LDL is not cholesterol. And Cheeses don't contain LDL.
03:15:45 <Ilari> Oh, the original is apparently (roughly translated): "The baddie of cholesterol is LDL, which is only contained in hard animal fats. For instance cheeses include those".". Just as much fail: Animal fats do not contain LDL.
03:17:18 <lament> 1) "the baddie of cholesterol" doesn't have to be cholesterol
03:17:29 <lament> 2) "source" doesn't mean "container"
03:19:16 <Ilari> LDL is made by liver. I have no idea if LDL survives digestion proceses. Probably not.
03:21:36 <Ilari> Nor are animal fats source of LDL...
03:21:57 <Ilari> Wow, there's lot more fail in this article...
03:23:45 <Ilari> Well, its based on recomendations by (national) heart disease association, so no wonder...
03:24:11 <lament> yeah those idiots know nothing about cholesterol
03:24:59 * Sgeo was taught that HDL was "good" and LDL was "bad"
03:26:13 <Ilari> There are subtypes of LDL. Some better, some sightly worse. Then there's real nasty kind that shows up as "LDL" (but isn't) on common panels.
03:27:27 <Ilari> Oh, and statins (cholesterol-lowering drugs) don't do much to that real nasty kind...
03:29:08 <Ilari> But then, the common way of determining LDL yields quite wild results that are often quite wrong.
03:29:33 <Ilari> (HDL is measured and total cholesterol is measured, those are fairly accurate).
03:31:59 <Ilari> That kind being "real nasty", one would expect that it would be quite strongly associated with heart disease. Indeed it is.
03:35:12 <pikhq> Sgeo: Nutrition is much more complicated than "X is good for you and Y is bad".
03:36:06 <pikhq> Well, except that modern, industrial processed foods tend to be bad for you. :P
03:37:53 <pikhq> Oh, yeah, and typical American portion sizes.
03:38:03 <pikhq> That's... Pretty obviously bad.
03:43:10 <Sgeo> Maybe I should start eating typical American portion sizes
03:43:43 <Ilari> But _WHY_ are typical american portion sizes so large?
03:44:57 <lament> but why is it pretty obviously bad?
03:45:38 <Ilari> Because it shouldn't be possible and indicates that something is badly wrong?
03:46:21 <lament> what shouldn't be possible?
03:46:42 <Ilari> The huge american portition sizes...
03:46:53 <lament> why shouldn't they be possible?
03:46:58 <pikhq> lament: Imagine a gigantic feast. We call it a "meal".
03:47:20 <lament> pikhq: i'm sure you realize that's total bullshit
03:47:20 <Ilari> Because very few should be able eat so much at once...
03:47:33 * Sgeo needs to eat more, so...
03:47:34 <lament> Ilari: nobody has any problems, really
03:47:52 <lament> just come to a restaurant and go crazy
03:48:09 <Ilari> Sgeo: "Just eat more and excercise less" is just as bad advice as "eat less and excercise more".
03:48:38 <pikhq> lament: 64 fl oz of soda (about 2L). As a single serving.
03:49:02 <lament> pikhq: that's not a standard serving size.
03:49:33 <lament> pikhq: i understand europeans perpetuating retarded stereotypes about americans, but you're american yourself and still doing that?
03:49:36 <pikhq> No, you're more likely to find 32 fl oz.
03:49:58 <Ilari> Here large bottle of soda is 1.5l... And the standard bottle (which is apparently meant as "serving" (ignore what manufacturers try to claim) is 0.5l.
03:50:01 <pikhq> With free refills.
03:50:41 * Sgeo tends to insist on small soda. If I get a large cup, I'll fill it up, then drink it all, and that's rather... uncomfortable
03:51:12 <Sgeo> Also, what, exactly, is wrong with eating more?
03:51:18 <Sgeo> For thin persons
03:51:44 <pikhq> lament: http://img.photobucket.com/albums/v226/ryany/BigMacExtraValueMeal.jpg
03:52:32 <lament> pikhq: yup. so? that's actually smaller than most meals at a restaurant.
03:52:41 <pikhq> That's 3,000 calories.
03:52:59 <Sgeo> What's the daily amount, 20,000? Or is it 2,000?
03:53:21 <pikhq> Well, that's what the FDA bases their recommendations on.
03:53:37 <Sgeo> So I could just eat that and not need to eat anything else for the day?
03:54:13 <Ilari> Well, taking in too many calories isn't good because high metabolism isn't good (and the alternatives are even worse).
03:54:21 <pikhq> Well, no, because it's fairly devoid of nutrients. It just has the caloric content that it was declared one needs.
03:54:30 <lament> pikhq: there're 1350 calories in a large big mac meal
03:54:39 <pikhq> lament: That's a super size big mac meal.
03:55:37 <Ilari> Second law of thermodynamics: "Calorie is not an calorie.". :-)
03:55:43 <pikhq> lament: Erm, sorry. I gave the wrong damned number.
03:55:49 <pikhq> lament: That's... 1580. My fault.
03:55:53 <pikhq> Still bloody absurd.
03:55:59 <pikhq> Just significantly less so.
03:56:08 <Sgeo> I think my pasta that I eat each night has over 1k calories
03:56:24 <lament> pikhq: 1580 for a larger than normal portion
03:56:45 <pikhq> For some people, it's larger than normal.
03:56:47 <Sgeo> Ilari, what are the chances of me being able to talk to a competent person somewhere in the US?
03:56:53 <pikhq> For others, that's smaller than normal.
03:56:57 <Sgeo> Get a nutrition plan worked out
03:57:35 <Ilari> Sgeo: Pretty slim (unless you pick carefully).
03:57:42 <pikhq> (granted, you're talking a small but fat portion of the population when you're discussing the people who actually do stuff like have multiple Big Macs in a sitting)
03:58:22 <Ilari> With soda and fries?
03:58:45 <pikhq> Ilari: Well, of course.
03:59:17 <Ilari> Wasn't the "big mac guy" in "Supersize Me", well... Not fat?
03:59:34 <pikhq> I'm quite perplexed about that as well.
04:00:19 <pikhq> Though it does demonstrate that absurd amounts of food won't automatically make you fat.
04:01:01 <Sgeo> My step-mother constantly tells me that I'm not eating enough
04:01:05 <Ilari> Well, its about fat in - fat out. Not calories in - calories out (unless you define the latter quantities to make it a tautology).
04:01:06 <Sgeo> It's probably true, but still
04:01:15 <Ilari> (and that's at storage boundary).
04:01:28 <Sgeo> If I want advice, I can't get it from her, no matter how competent, or not, she may be, because she'll drive me up the wall
04:01:42 <pikhq> Ilari: Well, calories are nothing more than a rough estimate of how much energy you get out of that food, so... Yeah.
04:01:45 <Sgeo> She does drive me up the wall
04:02:27 <pikhq> (incredibly rough when you consider that there's freaking sugar water as a common beverage.)
04:02:33 <Ilari> If it was fat_absorbed - fat_burned, then it would be just plain _wrong_.
04:03:05 <Ilari> Because there's fat synthethis term as well, and it certainly can be nontrivial.
04:06:22 <Ilari> pikhq: Not to mention, same macronutrients can provode different amounts of energy depending on amounts of other macronutrients...
04:07:44 <Ilari> Oh, and then there is stuff like adjustable efficiency of ATP production...
04:08:06 <pikhq> Ilari: Shit's more complex than burning the dry food and seeing the change in temperature -- who would've thought!
04:08:22 <pikhq> Erm, change in temperature of the water above.
04:08:51 <Ilari> Not even all fats are equivalent calories. Let alone fats, proteins and cabohydrates...
04:09:07 * Sgeo would love to just be able to inject something that could deliver ATP to all his cells
04:11:12 <Ilari> Too bad that with what all cells can use (glucose), even few extra grams injected (fast) would be toxic...
04:12:32 <Ilari> IIRC, something like 3 grams in fast injection would cause your blood sugar to go to toxic levels...
04:12:32 <lament> a grassy mount Or set with two feet Gules winged Sable and in base a bar wavy Sable inscribed with zeros and ones Or
04:13:19 <Ilari> Or maybe it would take about 4 if blood sugar is slightly low.
04:14:52 <Ilari> (Amount of glucose circulating in blood of normal adult under normal conditions: About 5 grams).
04:16:48 <Ilari> Or, one could inject fatty acids. Considerably less toxic (but not all cells can use those... OTOH, body can produce glucose for those cells that absolutely need it).
04:17:26 <pikhq> Eh, just bring about the singularity. That'll solve it.
04:18:30 -!- yorick has quit (Remote host closed the connection).
04:19:02 -!- olsner has quit (Ping timeout: 264 seconds).
04:26:32 <pikhq> No, you stupid Flash video, you do not need to keep buffering, there is already a 5% of the video buffer and increasing.
04:27:02 <pikhq> Whoever wrote this: you suck at programming and you should be ashamed of yourself.
04:27:31 <pikhq> Oh, and Adobe: you suck at programming and you should be ashamed of yourself.
04:32:15 -!- olsner has joined.
04:33:02 -!- Kordalien has quit (Ping timeout: 240 seconds).
04:39:53 <Ilari> With "why are they eating such huge porititons" I meant things like why the satiety isn't kicking in? Why doesn't the body downregulate hunger and upregulate energy consumption to keep the weight in check?
04:44:52 <flippo> Lack of exercise. The metabolism has nothing to calibrate with.
04:47:42 <Sgeo> Does running to school/running for the bus count as excersize?
04:47:55 <Ilari> The amount of excercise required for that is very small (one pretty much has to go out of one's way to get that little)...
04:48:57 <pikhq> Ilari: I presume that means "you need to get out of bed and walk to the kitchen every now and then" kind of "little".
04:49:21 <Ilari> pikhq: Well, within that order of magnitude...
04:49:51 <pikhq> Ilari: Point being that you'd have to make a point of it to not do that.
04:50:07 <Ilari> Not to mention, excercise isn't important for weight control. One can also see this by comparing energy expended in NEAT vs EAT(a.k.a. excercise).
04:50:46 <Ilari> Well, if one has shortage of energy, then the excercise levels could drop a lot...
04:50:53 <pikhq> Well, I'd imagine it helps somewhat, in that it does actually expend energy.
04:51:45 <Ilari> When one is talking about energies expended by ten hour walk (or was it run?)...
04:52:40 <Ilari> The mechanisms body uses to waste energy easily waste order of magnitude more than one could burn by excercising.
04:53:35 <pikhq> And anywas, all the exercise in the world won't help you if you use more energy than you ingest.
04:54:09 <pikhq> Though if you use more energy than you ingest, you might be having trouble as well. :P
04:55:37 <Ilari> There are some dangerous chemicals sometimes used for weight loss. If one overdoses on them, the primary toxic effect is that they make body literally cook itself producing energy for use.
04:56:12 <pikhq> That's... Frightening.
04:56:30 <Ilari> pikhq: Oh, your appetite should go down a lot as well when weight goes up...
04:56:37 <Sgeo> Any nice chemicals for weight gain?
04:57:07 <pikhq> Sgeo: "Force-feeding", I believe is the term.
04:57:20 <pikhq> Sgeo: Though it's rather unlikely you actually need weight gain.
04:58:03 <Ilari> Sgeo: Well, I think I know couple, but they are not nice ones...
04:58:05 <pikhq> Presuming reasonable health and a lack of relevant mental disorders, of course.
05:00:56 <Sgeo> As a kid, Supposedly, in an attempt to get me to eat on my own, suggested by a doctor, my parents didn't try to force me to eat, hoping I would eat on my own.
05:01:44 <pikhq> Okay, you may have a disorder then.
05:02:21 <Gregor> Who wants to read/edit a silly game story for meeeee?
05:02:46 <pikhq> Gregor: Does it involve Americans being fatasses?
05:03:01 <Gregor> It doesn't, but it /could/.
05:03:54 <Gregor> ANYWAY, I wrote a silly but appropriately-over-the-top opening sequence for ZEE: http://codu.org/tmp/story.txt
05:04:10 <Sgeo> I just asked my dad, he said he thinks I might have imagined being told that
05:04:15 <Sgeo> FUCK FALLIBLE MEMORY
05:04:25 <Sgeo> FUCK IT IN THE ASS
05:04:36 <Gregor> Sgeo: So your disorder is just delusions, not dietary problems.
05:04:56 <Sgeo> Gregor, since when is being human a disorder?
05:05:05 <Sgeo> And my dad may be mistaken :3
05:05:18 <Gregor> Sgeo: Since roughly 1.5 million years ago, I'd say.
05:07:00 <pikhq> Gregor: Quite nice.
05:07:40 <Sgeo> Gregor, that's awesome
05:08:09 <pikhq> Gregor: Also, you need more time.
05:08:20 <Gregor> I CAN'T MANUFACTURE TIME
05:08:26 <Gregor> For ZEE, what I need is more workforce!
05:08:35 <pikhq> Get cracking on the singularity! That'll solve it!
05:09:46 * oerjan is sensing a pattern here
05:10:48 -!- pineal_aenimal7 has joined.
05:12:41 <Gregor> Slereah: ... I love you.
05:12:54 <pikhq> unsafePerformIO suicideBomb
05:12:56 -!- pineal_aenimal7 has changed nick to pineal_aenimal.
05:13:24 <Sgeo> That could also be a point about people thinking this is esoterica
05:13:40 <Sgeo> </dont-explain-even-half-of-a-multifaceted-joke>
05:14:09 <Gregor> pikhq: Unsafe indeeeeeeeeeeeeeeeed!
05:17:12 -!- GreaseMonkey has joined.
05:17:12 -!- GreaseMonkey has quit (Changing host).
05:17:12 -!- GreaseMonkey has joined.
05:17:27 -!- Kordalien has joined.
05:29:02 -!- oerjan has quit (Quit: Later).
05:32:22 -!- pineal_aenimal has left (?).
05:41:05 <pikhq> It rips CDs. And is sufficiently paranoid about it to ensure a perfect rip if at all possible.
05:41:50 <pikhq> And does nothing else. It just dumps WAVs to disk.
05:42:37 <pikhq> Sgeo: It can't fix a disc made 100% opaque.
05:43:28 <Sgeo> I'm having some trouble parsing that. You can't mean "literally unreadable by anything", can you?
05:43:43 <Sgeo> That's too trivial
05:44:03 <Sgeo> Wait, I guess portions could be fully opaque? Or am I just confused today
05:44:06 <Gregor> Sgeo: Too trivial for "if at all possible"
05:44:16 <Gregor> <Sgeo> Give me a corner case.
05:44:23 <Gregor> <Sgeo> No, not that one.
05:48:38 <pikhq> Oh, yeah, it also often functions as a circumvention device.
05:50:14 <pikhq> (some CD copy protection schemes function by introducing intentional errors that a dedicated CD player will ignore)
05:51:35 <Gregor> Lots of people have single-purpose CD players nowadays.
05:55:21 <pikhq> You'll note that CD copy protection schemes kinda fell out of favor after a few years. :)
06:01:50 -!- sshc_ has joined.
06:04:55 -!- sshc has quit (Ping timeout: 252 seconds).
06:12:46 -!- zzo38 has joined.
06:12:52 <zzo38> Please help me. I tried to compile TeX but it just made a lot of errors, and it won't compile.
06:12:54 <zzo38> There is the errors: http://sprunge.us/SEhT
06:14:23 <zzo38> Can you please tell me what I did wrong?
06:20:39 <Gregor> Even the esoteric topics in computing channel cannot save you from Pascal.
06:21:31 -!- Kordalien has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.10/20100914125854]).
06:22:12 <zzo38> Save me from Pascal? I don't generally program in Pascal, but this is a program that is Pascal.
06:23:58 <zzo38> Is there some command-line parameter I need to add?
06:24:09 <Gregor> Haven't a clue, never built TeX myself.
06:28:26 <zzo38> Which options of GPC cause it to emulate Pascal-H?
06:40:40 -!- oerjan has joined.
06:43:12 -!- zzo38 has left (?).
06:43:15 -!- zzo38 has joined.
06:48:20 -!- Mathnerd314 has quit (Ping timeout: 255 seconds).
06:51:16 <zzo38> Do I need to redefine alpha_file (in section 25)? I probably also have to redefine othercases
06:54:47 -!- cheater99 has joined.
07:01:22 -!- Sgeo has quit (Ping timeout: 245 seconds).
07:08:08 -!- myndzi has quit (Ping timeout: 255 seconds).
07:12:06 <zzo38> I got it to compile with no errors, but now it complains about 'his.tex'
07:13:22 <zzo38> I cannot find anything in the program about what 'his.tex' is supposed to be.
07:17:13 -!- myndzi has joined.
07:18:51 <zzo38> (I cannot find on Google or Wikipedia about what 'his.tex' is supposed to be, either.)
07:22:19 -!- Mathnerd314 has joined.
07:24:05 <zzo38> What could probably be done, although it would take a very long time, is looking at the source-codes of TeX and rewriting it in Enhanced CWEB, changing things as necessary as you are moving along.
07:25:08 -!- wareya_ has changed nick to wareya.
07:25:35 -!- FireFly has joined.
07:28:18 -!- Mathnerd314 has quit (Ping timeout: 272 seconds).
07:28:30 <zzo38> Are you willing to collaborate on this project?
07:34:53 -!- myndzi has quit (Ping timeout: 276 seconds).
07:34:53 -!- zeotrope has quit (Ping timeout: 276 seconds).
07:45:27 -!- tombom has joined.
07:54:14 -!- zzo38 has quit (Remote host closed the connection).
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:00:47 -!- lament has quit (Ping timeout: 255 seconds).
08:01:59 -!- sftp has quit (Remote host closed the connection).
08:13:38 -!- Mathnerd314 has joined.
08:20:59 -!- lament has joined.
08:27:55 -!- wareya_ has joined.
08:31:02 -!- wareya has quit (Ping timeout: 264 seconds).
08:56:53 -!- tombom has quit (Quit: Leaving).
09:01:22 -!- oerjan has quit (Quit: Good night).
09:01:42 -!- lament has quit (Ping timeout: 240 seconds).
09:01:59 -!- FireFly has quit (Quit: swatted to death).
09:03:35 -!- FireFly has joined.
09:21:14 -!- MigoMipo has joined.
09:31:13 -!- coppro has quit (Remote host closed the connection).
09:46:30 -!- iGO has joined.
09:52:50 -!- sebbu has quit (Ping timeout: 255 seconds).
09:54:11 -!- MigoMipo_ has joined.
09:57:08 -!- MigoMipo has quit (Ping timeout: 272 seconds).
10:31:06 -!- MigoMipo_ has changed nick to MigoMipo.
10:55:01 -!- GreaseMonkey has quit (Quit: Welcome honored guest. I got the key you want! would you like onderves. of Yourself).
11:15:40 -!- ais523 has joined.
11:38:56 -!- sebbu has joined.
11:54:24 -!- ais523 has quit (Remote host closed the connection).
12:17:17 -!- Sgeo has joined.
12:20:01 -!- iGO has quit (Remote host closed the connection).
12:25:36 -!- iGO has joined.
12:42:10 -!- webquint has joined.
12:45:47 -!- webquint has quit (Client Quit).
12:53:13 -!- BeholdMyGlory has joined.
13:05:42 -!- myndzi has joined.
13:10:43 -!- myndzi\ has joined.
13:12:14 -!- myndzi has quit (Ping timeout: 276 seconds).
13:43:05 -!- ais523 has joined.
13:43:49 <ais523> yay, my paper was accepted
13:44:19 <ais523> hmm, now what's going on?
13:44:29 <ais523> IRC is working fine, but I don't seem to be able to create new Internet connections
13:44:33 <ais523> not web or email, anyway
13:44:49 <ais523> oh, started working again
13:44:52 <ais523> it was just being really slow
14:25:28 -!- alise has joined.
14:30:12 <ais523> ugh, the Internet here is crazy
14:30:25 <ais523> it sometimes doesn't form new connections for something like a minute, even though existing connections work fine
14:30:36 <ais523> (at least, for web and email)
14:32:09 <alise> 19:23:45 <Ilari> Well, its based on recomendations by (national) heart disease association, so no wonder...
14:32:09 <alise> 19:24:11 <lament> yeah those idiots know nothing about cholesterol
14:32:24 <alise> i'm glad our channel op is such a fine, upstanding citizen who believes everything the american government tells him.
14:32:54 <alise> better check your BMI
14:33:03 <alise> ais523: is that at uni?
14:34:22 <alise> ais523: take over the networking department
14:34:39 <alise> ais523: err, can Wooble actually do that while he's on hold?
14:34:54 <alise> oh hm or is deputisation one of those things even non-players can do...
14:35:57 <ais523> being on hold hardly stops anything
14:36:00 <ais523> it's not like BlogNomic
14:36:11 <ais523> mostly, it just lets people oust you from offices, and makes you ineligible for certain things
14:36:30 <Sgeo> Such as winning when everyone else is winning
14:36:33 <Sgeo> </still-bitter>
14:38:57 <quintopia> i looked at that once. seems awfully complicated (as any nomic that has been in existence for years upon years must)
14:39:18 <quintopia> but i GUESS it at LEAST has fewer rules than certain versions of DnD...
14:39:49 <alise> Ilari: you should publish a book about nutrition :-)
14:40:32 <ais523> <sysctl> Here is a quick summary of Java build systems: Q: What's the difference between Ant and Maven? A: The creator of Ant has apologised.
14:40:41 <ais523> I still don't see why they don't just use make, it seems to work fine
14:40:57 <ais523> meh, find | xargs javac works fine if you don't care about minimal recompiles
14:41:06 <Sgeo> Wait, seriously? Someone apologized for making widely-used software?
14:41:38 <ais523> Sgeo: it actually had a source for that statement, http://blogs.tedneward.com/2005/08/22/When+Do+You+Use+XML+Again.aspx
14:41:50 <alise> Sgeo: ant is awful
14:44:22 <Sgeo> That sounds like an apology for an aspect of Ant
14:45:34 <Vorpal> <ais523> it sometimes doesn't form new connections for something like a minute, even though existing connections work fine <-- maybe some firewall issue?
14:46:04 <alise> 05:43:49 <ais523> yay, my paper was accepted
14:47:08 <ais523> one about hardware compilation
14:47:33 <ais523> to be precise, we generalised a previous result from "programs that type in SCI can be compiled into hardware" to "programs that type in ICA can be compiled into hardware, with a few exceptions"
14:47:51 <ais523> which is pretty nice, because unlike SCI, ICA doesn't have a bunch of arbitrary restrictoins
14:48:17 <ais523> it was really much the same as a TC-ness proof, showing one lang compiles into another...
14:49:45 <alise> ais523: how's the Complex Systems publication going? ;)
14:50:04 <ais523> revise-and-resubmit phase
14:50:14 <ais523> with instructions that were too banal for me to be interested in them
15:07:17 <alise> 22:12:52 <zzo38> Please help me. I tried to compile TeX but it just made a lot of errors, and it won't compile.
15:07:21 <alise> zzo38: you don't compile TeX itself...
15:07:24 <alise> you compile the C translation
15:07:33 <alise> it's written in a crazy 70s Pascal that Knuth used
15:07:49 <alise> 23:24:05 <zzo38> What could probably be done, although it would take a very long time, is looking at the source-codes of TeX and rewriting it in Enhanced CWEB, changing things as necessary as you are moving along.
15:07:51 <alise> again, the C translation
15:07:54 <alise> 23:28:30 <zzo38> Are you willing to collaborate on this project?
15:08:23 <ais523> at some point I need to figure out what Enhanced CWEB is
15:08:30 <ais523> I can't even tell if it's a programming language or not
15:09:00 <alise> ais523: you know Knuth's CWEB?
15:09:11 <alise> ais523: you know Knuth's WEB?
15:09:22 <ais523> again no, but I've vaguely heard of it
15:09:42 <alise> ais523: WEB = the first literate programming system; TeX + Pascal. TeX is written in it, for instance.
15:09:58 <alise> Assign blocks of code to a name-with-spaces, include it later, macros, documentation all around it, etc.
15:10:01 <alise> ais523: CWEB = that for C.
15:10:10 <alise> ais523: Enhanced CWEB = what happens when you apply zzo38 to CWEB and peel it off after a few days.
15:10:14 <ais523> and Enhanced CWEB is a zzo38 version
15:10:34 <ais523> hmm, what's the converse of Not Invented Here?
15:10:47 <ais523> when you assume that the entire world uses software that was written by you?
15:11:11 <ais523> that seems to be unique to zzo38 and Microsoft
15:11:40 <alise> ais523: zzo38 has NIH too, though!
15:11:54 <alise> except it tends to be Not Invented Here, But That's Okay, I'll Cannibalise It
15:11:57 <ais523> yes, the converse of something being true doesn't mean that the thing itself is false
15:12:12 <alise> ais523: Used Elsewhere
15:12:16 <alise> ais523: Used Everywhere
15:12:30 <alise> Microsoft: NIH, UE; zzo38: NIHBTOICI, UE
15:23:48 -!- sftp has joined.
15:31:04 <Vorpal> alise, btw perhaps you know I'm against eyecandy animations and such? Today I realised what exactly is the issue with a lot of them
15:32:40 <Vorpal> that they are slow enough that you end up having to wait before you can do whatever you planned to do next. For minimising windows on OS X, it is quite fast but still, I had to wait before I could start reading the text in the window behind due to the animation covering it, only a fraction of a second, but still enough to be annoying.
15:33:12 <Vorpal> similar issues seem to apply to many other such animations.
15:33:23 <alise> Vorpal: That's just bad design.
15:34:08 <alise> Vorpal: A *good* hiding animation -- one that smoothly indicates what's happening -- would last only a few milliseconds, and the window would go semi-transparent too.
15:34:18 <alise> Perhaps even simply fade out *while on the way* to the dock.
15:34:38 <alise> This would help indicate what has happened -- believe it or not, your brain *can* act more efficiently with these cues -- without being irritating.
15:34:38 <Vorpal> alise, indeed that would work, but that was not what that version of OS X did at least.
15:34:47 <alise> Vorpal: btw, you change and speed that animation up.
15:35:01 <alise> change: system preferences -> set it to the one that isn't genie, stops it being irritating as fuck
15:35:04 <alise> speed up: some terminal bullshit
15:35:06 <Vorpal> alise, was helping someone with a thing, not my computer
15:35:18 <Vorpal> anyway, it wasn't the genie one
15:35:49 -!- cpressey has joined.
15:37:20 <Vorpal> alise, but anyway, the os x minimising is just a rather extreme example of it (especially the genie style), there are lots of other such animations on both windows (vista or later only) and OS X that takes a fraction of a second too long. Sure there are some that don't annoy, but I guess you don't really remember those. Selective reporting and such.
15:37:44 <alise> Modern UIs aren't designed with HCI principles in mind; news at 11.
15:38:04 <alise> Jef Raskin died, now even less people care; news at 11.
15:38:55 <Vorpal> alise, the "zoom icon in dock while mouse is above it" is instantaneous as far as I remember, and is actually slightly helpful, at least if set to sane parameters for zoom level and such
15:39:27 <alise> Vorpal: It's not instantaneous -- it's... continuous.
15:39:35 <alise> As in, literally, the zoom changes focus even half-way between two icons.
15:39:44 <alise> The first hover over has a short animation to do the first zoom-and-expand of the Dock.
15:39:47 <Vorpal> yes, but I meant like:
15:40:06 <Vorpal> it doesn't take half a second if you quickly move your mouse down from middle of screen to the dock
15:40:06 <alise> I never turned it on; I could never predict where my mouse should go to reach an icon, due to the zooming.
15:40:11 <Vorpal> as soon as it is above it is zoomed
15:40:29 <alise> It probably helps a lot if you have tons of stuff in your Dock and thus have it at a small size.
15:41:52 <Vorpal> alise, or menus/submenus that takes a fraction of a second before they open. iirc windows xp had something like that. Was annoying.
15:42:04 <Vorpal> not an animation, but utterly pointless and not helpful at all
15:42:07 <alise> Vorpal: Uhh, as for submenus, GTK has that too.
15:42:16 <alise> As does *every interface you use*.
15:42:20 <alise> It's a shorter delay, but it's there.
15:42:32 <alise> Otherwise moving your mouse across tons of elements quickly would have menus flash everywhere.
15:42:33 <Vorpal> alise, yes, short enough to not notice, that's the key
15:42:35 <alise> Disorienting, to say the least.
15:42:47 <alise> Vorpal: Try it now; I bet you will notice.
15:42:48 <Vorpal> windows xp had "long enough to be annoying"
15:42:59 <alise> There's certainly a noticeable delay here, just not an annoying one or one you think about.
15:43:00 <alise> Vorpal: Well, yeah.
15:43:03 <alise> You could tweak that with TweakUI.
15:43:09 <Vorpal> alise, indeed there is a tiny delay
15:43:23 <Vorpal> alise, yes indeed, but why not make it right from the start
15:43:33 <alise> Vorpal: "It seemed like a good idea at the time."
15:43:46 <alise> TweakUI was released after XP. I don't think they changed the delay since at least 2000.
15:43:49 <alise> Maybe even before that.
15:44:00 <alise> And TweakUI's policy is basically "EXPOSE EVERYTHING", not "LET PEOPLE FIX EVERYTHING"
15:44:12 <Vorpal> alise, as for menus all over the place, not really annoying, classic mac OS used to do it that way. As do some other windowing toolkits
15:44:18 * alise decides to run OS X in VirtualBox.
15:44:22 <alise> "Hey, it can do EFI."
15:44:31 <alise> Although I'll have to emulate a 32-bit machine so this will be *fun fun fun*.
15:44:42 <alise> Oh god, it's distributed on a DVD, isn't it.
15:45:14 <alise> (Yes, I have a copy of Leopard; no, I don't have an optical drive.)
15:45:28 <Vorpal> when the base install of an OS doesn't fit on a single CD, something is wrong. :P
15:45:45 <alise> Vorpal: Yeah! Fuck you Slackware and Fedora!
15:45:51 <alise> Slackware is well-known for its bloat.
15:46:14 <Vorpal> alise, hm.... ubuntu: single cd, arch; single cd. And I don't mean netinstall.
15:46:22 <Vorpal> Gentoo: single cd last I looked
15:46:33 <Vorpal> though iirc they had a livedvd as well
15:46:34 <alise> Vorpal: It's simply a different distribution philosophy -- with the Slackware CDs, you can install anything you want.
15:46:51 <alise> Vorpal: It *does* suck, though.
15:47:05 <alise> But it isn't indicative of bloat.
15:47:14 <alise> (Well, in OS X's case, it basically is.)
15:47:23 <alise> (To be fair, there is also a whole lot of stuff on there.)
15:47:36 <Vorpal> alise, also note I said "base install". The definition of base install is rather fuzzy indeed. I would say it is completely different for arch and ubuntu for example
15:47:46 <pikhq> Still, you can fit your base install on a single CD just by defining a base install that's small.
15:47:48 <alise> Safari, all the i* programs, GarageBand, every single damn developer tool, misc. applications
15:47:56 <alise> pikhq: "Base install: The kernel!"
15:48:03 <pikhq> I think you can just about get Debian on a small stack of floppies still.
15:48:16 <cpressey> as long as i have butterflies flitting around on my desktop i'm happy
15:48:24 <pikhq> I know the *installer* still takes up 2 floppies.
15:48:55 <Vorpal> pikhq, has some minimum requirements. Bootable, useable package manager, some way to edit config files, a shell, can connect to network. Arch is pretty close to that minimum.
15:49:26 <pikhq> Vorpal: I... Can do that on a single floppy disk.
15:50:26 <Vorpal> pikhq, indeed, it might be possible. However arch comes with a usable environment on the livecd too. Like you get a shell there. Useful to start any non-standard stuff like software raid or dmcrypt or such.
15:51:13 <Vorpal> pikhq, and fitting that on a floppy as well would be.... tricky
15:51:31 <alise> pikhq: Namespaces in C: solve.
15:51:43 <Vorpal> pikhq, besides the kernel I need to boot my system, all the drivers I need compiled in and almost everything else turned off is about 2.5 MB iirc.
15:51:53 <alise> Vorpal: No, you can have VESA.
15:51:57 <alise> You just don't want VESA.
15:52:17 <Vorpal> alise, I do have VESA fb :P, and I have a system with just VGA, no fb support at all
15:52:23 <Vorpal> not sure of kernel size
15:52:36 <alise> pikhq: Actually, howsabout figuring out a way to have functions on structures look nicer to call, and we can go from there and use that for namespaces.
15:52:37 <Vorpal> and not going to boot it atm
15:52:58 <ais523> hmm, that was a fun conversation
15:53:17 <ais523> I was talking to my supervisor and another researcher about implementing the fixed-point combinator in hardware
15:53:19 -!- yorick has joined.
15:53:58 <ais523> rather unambitiously, they agreed that even third-order fixedpoint combinators ((a->a)->(a->a))->a were practically useful
15:54:07 <Vorpal> <alise> pikhq: Namespaces in C: solve. <-- hm, not impossible, what about having some syntax sugar for appending a prefix to all identifiers, I guess that would approach C++ name mangling quickly, though with just namespaces you could do far cleaner than that
15:54:14 <ais523> although I think you can get up to arbitrary orders (with the caveat that such circuits always have a risk of running out of memory)
15:54:24 <alise> Vorpal: I mean without modifying the language.
15:54:47 -!- sshc_ has quit (Read error: Connection reset by peer).
15:54:48 <ais523> is preprocessing allowed?
15:54:50 <cpressey> alise: #define, #define, #define. done.
15:55:00 <Vorpal> alise, <namespace>_<identifier> for all identifiers?
15:55:04 <alise> Vorpal: You can put everything as function pointers in a structure, but (1) that's ugly to construct and (2) you'd start wanting to be able to do foo->x(...) for a structure foo to mean foo->x(foo, ...) which is harder :-)
15:55:09 <ais523> Vorpal: but normally you want to be able to import namespaces
15:55:12 <alise> Vorpal: that's ugly and makes everything unconcise
15:55:15 <alise> ais523: not necessarily
15:55:26 <alise> even being able to just rename a namespace temporarily would be fine
15:55:49 <Vorpal> alise, is non-cpp preprocessing allowed?
15:56:26 <Vorpal> alise, I mean, with a custom preprocessor this would be simple. You could make one small enough to compile as the first step in the makefile or such
15:57:17 -!- Phantom_Hoover has joined.
15:58:29 <yorick> alise: what about preprocessing?
15:58:57 <cpressey> if only #define did pattern-matching. you could hack something together quite nicely i think
15:59:05 <alise> if you have to, I guess a simple textual substitution too
15:59:09 <alise> but nothing that involves tons of parsing :P
15:59:13 <yorick> alise: but what about?
15:59:29 <yorick> what do you want to accomplish?
16:00:15 <alise> <alise> pikhq: Namespaces in C: solve.
16:00:17 <ais523> hmm, I wonder if upp (which I haven't actually written yet, properly) would work for this?
16:00:23 <alise> ais523: upp? *scared*
16:00:30 <ais523> the Underlambda Preprocessor
16:00:44 <yorick> alise: what about namespaces in C
16:00:44 <ais523> it basically just does literal textual substitutions, but with a couple of interesting twists
16:01:03 <ais523> and is powerful enough to compile Underlambda into subsets of itself
16:01:23 <alise> yorick: see later on.
16:02:21 <ais523> let's see, it has only two commands
16:02:30 <ais523> one is / on a line by itself, which does nothing but which is referenced by the other command
16:02:37 <ais523> and the other is a/b for any character strings a and b
16:02:38 <cpressey> ais523: is "itself" bound to Underlambda or upp?
16:02:44 <ais523> cpressey: to Underlambda
16:02:58 <alise> cpressey: if it wasn't it'd be a trivial statement :)
16:03:18 <alise> since it means a subset of it could do underlambda code
16:03:18 <ais523> what that does, is it substitutes all instances of a with b in a) the thing you're preprocessing; b) the preprocessor program itself, but only beyond the next lone /
16:03:21 <cpressey> alise: i dunno, a preprocessor that can compile a language into subsets of a preprocessor...
16:03:22 <alise> meaning the full thing is TC
16:03:25 <alise> meaning it could compile it
16:03:35 <yorick> alise: for example, where are you responding to
16:03:40 <ais523> additionally, a/b will not substitute in text that itself was produced by a substitution, unless a lone / has executed in the meantime
16:04:01 <ais523> there, that's pretty simple
16:04:16 <ais523> but it lets you do block-replacements of fundamental commands, defining them in terms of each other
16:04:28 <yorick> <alise> pikhq: Namespaces in C: solve. <-- that has a "pikhq" in front of it
16:04:45 <cpressey> alise: even if it were TC, that wouldn't imply the statement was trivial
16:05:17 <ais523> Phantom_Hoover: implementing fixed-point combinators and typed lambda calculus in hardware; namespaces in C; preprocessing Underlambda
16:05:30 <ais523> also a metaconversation about what conversations are running
16:06:09 <yorick> Phantom_Hoover: personally, no clue
16:06:23 <alise> cpressey: i suppose not
16:06:26 <alise> yorick: i was addressing pikhq.
16:06:57 <alise> yorick: about "Namespaces in C: solve."
16:09:19 <Ilari> Hmm... Skin overheating by having laptops on lap for too long... Article included talk about cancer... I wonder if PUFA oxidation due to heat is involved...
16:10:20 <alise> Ilari: Only if you use a stupid laptop with an overly-hot processor :)
16:10:27 <alise> And the fan too low.
16:11:52 <Ilari> Of course, enough heat / radiation will damage skin no matter what, but stuff like PUFA concentrations could determine how sensitive or resistant one is (PUFAs are chemically unstable)...
16:12:30 <alise> pikhq: Do you have any opinions on Go?
16:13:44 <alise> ais523: heh, remember that mergesort I said I couldn't figure out what was wrong with?
16:13:52 <alise> ais523: changing "< end" to "<= end-1" fixed it.
16:14:09 <ais523> was it using floats as loop counters?
16:14:25 <ais523> (and non-integer floats, at that?)
16:14:37 <ais523> or was the code/ really/ isane?
16:14:54 <alise> ais523: well, in this case, floating point was involved, but there wasn't a loop counter
16:15:00 <alise> it was comparing elements in an array
16:15:07 <ais523> ah, and those were floats?
16:15:10 <alise> ais523: and tl;dr using floating-point infinity to denote end of array makes weird shit happen
16:15:53 -!- augur has quit (Remote host closed the connection).
16:16:02 <ais523> screwing with infinities tends to do that
16:17:04 <Ilari> FP also has other weird shit like +0 and -0 are seperate numbers (that's actually useful in calculations).
16:18:04 -!- lament has joined.
16:18:28 <Phantom_Hoover> Never did I think I'd see the day when there was a lambda calculus reference in a Star Wars webcomic.
16:24:26 <alise> pikhq: So, BF compilation.
16:27:00 -!- atrapado has joined.
16:27:42 -!- sshc has joined.
16:30:03 <Phantom_Hoover> It doesn't look like it from the first few Google results...
16:31:30 <ais523> hmm, I was just looking at the Linux manpage reboot(2)
16:31:37 <ais523> and I get the feeling I'm missing something
16:31:47 -!- Quadrescence has quit (Ping timeout: 245 seconds).
16:31:55 <ais523> some of the hex numbers were obfuscated by being translated to decimal, for no really obvious reason
16:32:01 <ais523> and I don't see any significance in the hex either
16:32:53 <alise> on the Commodore 64 TCP/IP stack: "It doesn't do state tracking. It puts the state in the TCP sequence numbers. Save on RAM by passing state back and forth through the network."
16:33:13 <alise> ais523: one is linus' birthday
16:33:20 <alise> they all have significance as numbers
16:33:33 <alise> ais523: I know this because I wrote a program that calls it relatively recently :P
16:33:34 <ais523> I was assuming it was something NSFW, based on how oblique they were being
16:34:04 <alise> ais523: LINUX_REBOOT_* are constants for them in the kernel, RB_* in glibc
16:34:09 <alise> i suggest using the hex directly to avoid the headache
16:34:09 <ais523> I understand the purpose (to make sure that reboot isn't called by mistake by a program in undefined behavior)
16:35:01 <alise> so, I am pretty sure my mind faked Wake-Induced Lucid Dreams to me today
16:36:06 <alise> as in, I was dreaming, and my mind made me think I was falling asleep consciously, and provided fake hypnagogic imagery and sounds to fool me into this
16:36:09 <lament> is that when you think you dream of waking up and are currently in a lucid dream, but actually you woke up for real?
16:36:14 <alise> leading to a dream-in-a-dream
16:36:22 <alise> lament: WILD is just a method of achieving lucid dreams
16:36:36 <alise> you stay conscious but become very relaxed and let your body go to sleep
16:36:44 <alise> sort of related to meditation, I guess
16:37:00 <alise> lament: but I think my dream decided to start with me falling asleep using WILD
16:37:13 <ais523> alise: it sometimes end up dreaming I've worken up
16:37:14 <alise> and manufactured some hypnagogic imagery and (really irritating clanging) sounds to go with it
16:37:18 <yorick> I should practice lucid dreaming some more
16:37:21 <alise> leading to a fumbled attempt at a lucid dream inside a non-lucid dream
16:37:24 <alise> where everything was hideously unrealistic
16:37:33 <alise> sometimes before i fully got "into" a dream induced that way
16:37:33 <ais523> and gone through my normal morning routine
16:37:40 <ais523> and then actually woken up and had to do it all over again
16:37:40 <alise> and i saw a flicker of the "real world" and me moving slightly in it
16:37:43 <alise> so I thought, I'm not asleep enough for that
16:37:45 <alise> but that can't be right
16:37:48 <alise> because you're paralysed by that point
16:37:52 <alise> and completely asleep
16:38:01 <alise> so I'm fairly sure that if i had let that happen, I'd have "woken up" into an actual dream
16:38:10 <alise> and presumably have performed a reality check due to the circumstances...
16:38:15 <alise> but i was too stupid to realise this at the time :)
16:38:17 <alise> then i woke up for real
16:38:23 <alise> OH GOD I HAVE 7 FINGERS
16:38:26 * alise has quit (Connection reset by peer)
16:38:28 <ais523> once you know you have to perform a reality check, everything becomes pretty easy
16:38:29 -!- Quadrescence has joined.
16:38:41 <ais523> but somehow, you rarely think of doing that in a dream
16:38:50 <alise> ais523: yeah I just look at my hands and if I have fingers sprouting out of my other fingers I just jump out the nearest window
16:38:58 <alise> that's the easiest way to get places in a dream!
16:39:04 <alise> (note: if I ever end up with deformed hands, I am so dead)
16:39:27 <ais523> alise: oh, I tend to invent methods of transportation in my dreams
16:39:33 <yorick> I tried the reality checks...most of them give false negatives inside dreams :/
16:39:41 -!- alise has left (?).
16:39:42 -!- alise has joined.
16:39:45 <Vorpal> <alise> as in, I was dreaming, and my mind made me think I was falling asleep consciously, and provided fake hypnagogic imagery and sounds to fool me into this <-- hm, soon there is no way we can trust anything :P
16:39:46 <alise> yorick: hands always works for me
16:39:54 <alise> I either have more than five fingers, or fingers are placed in ways that geometry doesn't quite allow
16:40:01 <alise> Vorpal: I N C E P T I O N
16:40:01 <ais523> you know about the science-fiction idea of going somewhere distant via going there over the course of years (at near-lightspeed), but in such a way that the people aboard don't perceive most of it?
16:40:11 <Vorpal> alise, rings a bell but can't place it
16:40:18 <alise> ais523: I haven't heard of it, but go on.
16:40:23 <yorick> alise: my fingers are always fine while I'm dreaming...and I can even touch them
16:40:27 <Vorpal> I never remember to perform reality checks when dreaming...
16:40:29 <ais523> well, when I move around in my dreams, it's normally using a means of transport that does that over really short distances
16:40:31 <alise> Vorpal: a recent film about nested dreams-within-dreams
16:40:43 <Vorpal> and I very rarely remember my dreams
16:40:44 <alise> Vorpal: and people from higher levels of dreams being able to change them and stuff
16:40:47 <yorick> alise: I haven't tried breathing with my nose closed yet :/
16:40:49 <alise> but it sounds worth watching
16:40:54 <ais523> some sort of capsule thing that takes days to go just a few miles, but you're unconcious for most of it so don't care
16:40:58 <alise> yorick: the fun thing with more than five fingers
16:40:59 <alise> is that you can feel them!
16:41:04 <alise> you can use a finger to touch a fake one
16:41:06 <quintopia> alise: :o it was the best movie of the summer what's taking you so long?
16:41:09 <alise> and it feels like a strange buzzy feeling
16:41:13 <alise> quintopia: i'm lazy
16:41:23 <Phantom_Hoover> alise, did you go into a lucid dream just to recreate Inception?
16:41:30 <alise> Phantom_Hoover: :D
16:41:34 <ais523> the thing that really annoys me about Inception, is that it seems that everyone who watches it goes to everyone they know and says "I watched Inception"
16:41:45 <ais523> to the extent that I worry that it's specifically designed to brainwash people into doing that
16:41:46 <alise> ais523: you know, you *can* just invent a magic mirror
16:41:47 * yorick was invisible last time :)
16:41:53 <alise> you don't have to have sci-fi dreams
16:42:11 <ais523> alise: I suppose it's to do with the sort of transport I use in RL
16:42:23 <ais523> I rarely move around via the fastest method
16:42:30 <quintopia> ais523: well, i did that, but i also saw it on the night it came out, so it wasn't a trend yet. now people do it because everyone keeps pestering everyone to see it so they have to tell them to stop the pestering
16:42:37 <ais523> many of my esolang ideas are developed waiting at bus shelters
16:42:44 <alise> ais523: Somehow, even when I'm totally lucid though, I can't convince myself that I can control reality. As in: it requires absolute belief in that what you're about to do will work to change the gameworld. I am apparently too rational to summon up such faith.
16:42:53 <alise> ais523: This saddens me.
16:42:55 <Vorpal> ais523, and a really strange transport method
16:43:16 <ais523> alise: it works better if you invent a pseudoscience explanation, even if you don't believe in pseudoscience
16:43:19 <alise> ais523: Maybe I should become religious for the practice.
16:43:27 <alise> ais523: Unfortunately my periods of lucidity also tend to be short. :(
16:43:39 <yorick> yeah...same...but you can prolong them :)
16:43:40 <ais523> shift into the fourth dimension, do stuff there, with the knowledge that you can manipulate that based on the fact that you own the device that lets you go there in the first place
16:43:42 <alise> Also, I tend to sleep only when I'm really tired on weekends; this rarely seems to lead to dreams I remember for me.
16:43:46 -!- lament has quit (Read error: Operation timed out).
16:43:51 <ais523> alise: you only sleep once a week?
16:43:52 <yorick> (spinning around will mostly work)
16:43:56 <Vorpal> ais523, start traveling by train until you finish of the current unfinished ones then!
16:43:56 * Phantom_Hoover decides to try defining the braid group in Coq for the halibut.
16:43:57 <alise> yorick: I've done the tactics, like feeling a brick wall a lot and staring at the details.
16:44:04 <alise> yorick: It never works.
16:44:09 <ais523> oh, I misparsed your sentence
16:44:17 <quintopia> reality check: can anyone here pinpoint the last time they had a stereotypical dream?
16:44:18 <alise> yorick: As soon as I move away and stop looking like someone with Down's syndrome, everything becomes fuzzy again.
16:44:19 <ais523> I didn't even notice it was ambiguous until you assumed I was being sarcastic
16:44:30 <ais523> quintopia: even my stereotypical dreams are rather non-stereotypical
16:44:42 <yorick> alise: maybe...the brick thing works better?
16:44:49 <alise> ais523: with normal days, I don't sleep long enough for a nice lucid dream really
16:44:52 <Phantom_Hoover> alise, there are some very common dreams and nightmares.
16:44:54 <alise> yorick: So just stare at a brick the whole time??
16:45:04 <alise> Phantom_Hoover: Yeah, I don't get them though. Or at least I never remember them
16:45:04 <yorick> alise: how am I supposed to have bricks?
16:45:09 <quintopia> aka, a dream where your teeth are falling out and/or a dream where you are having trouble controlling a car/driving it from the back seat
16:45:15 <alise> <alise> yorick: I've done the tactics, like feeling a brick wall a lot and staring at the details. <yorick> alise: works for me
16:45:20 <alise> <alise> yorick: As soon as I move away and stop looking like someone with Down's syndrome, everything becomes fuzzy again. <yorick> alise: maybe...the brick thing works better?
16:45:30 <yorick> (17:43) < alise> yorick: It never works.
16:45:33 <Phantom_Hoover> I think I've had some serious up-messery of my scale perception before
16:45:34 <yorick> (17:43) < yorick> alise: works for me
16:45:34 <alise> quintopia: I have never had any of those dreams.
16:45:40 <quintopia> i've had both of the former in the last month but I can't remember exactly when. I haven't flown since I was a child.
16:45:43 <alise> yorick: "how am I supposed to have bricks?" what does this mean
16:45:48 <yorick> alise: how am I supposed to find bricks to stare at
16:45:49 <Phantom_Hoover> But that was while trying to get to sleep, not actually once sleeping.
16:45:54 <alise> yorick: Uhh, leave the house you're in.
16:45:56 <alise> Look at the bricks.
16:45:57 <Vorpal> quintopia, I guess stereotypical dreams differ between persons
16:46:00 <alise> yorick: You said works for me; presumably you already do this.
16:46:02 <alise> So why are you asking me?
16:46:13 <yorick> alise: because I recognize the feeling if it becoming fuzzy again
16:46:17 <quintopia> Vorpal: well, had any dreams that sound similar to those ideas?
16:46:23 <yorick> alise: and yes, mostly the places I dream of are not made of bricks
16:46:32 <Vorpal> quintopia, I never seem to have recurring themes in my dreams either
16:46:38 <ais523> hmm, both my house and my workplace are made of bricks
16:46:56 <yorick> usually either polished plastic or metal
16:46:56 <ais523> Vorpal: mine rarely have recurring themes, but they tend to have consistent geography with each other
16:47:01 <alise> Phantom_Hoover: My most common recurring nightmare when I was young was -- imagine a zoomed-out satellite picture of a huge city on a regular British day -- not sunny, not raining. A ladder: I am on this ladder. It stretches down to the ground; my vision is that satellite picture. At the top: A hot air balloon. Someone in it -- that I recognise, encourages me to climb up further.
16:47:09 <ais523> which is unusual mostly in that the geography in the dreams is /not/ the same as the real world
16:47:12 <Vorpal> ais523, that is a theme by the definition I used
16:47:15 <alise> But before I get close to the balloon, the ladder tips forwards and falls away.
16:47:16 <quintopia> i have had a few recurring themes. but my best dreams are one time only
16:47:18 <alise> I start falling and wake up.
16:47:40 <yorick> my most recurring nightmares currently is being stabbed/shot by my friends :/
16:47:42 <ais523> alise: oh, you mean a solid ladder, rather than a rope ladder, etc.
16:47:42 <alise> ais523: interestingly, I universally start lucid dreams in a warped version of my old bedroom
16:47:46 <alise> ais523: yes, metal
16:47:48 <Phantom_Hoover> My dreams are either really boring or all very slippery in memory.
16:47:49 <ais523> that's balanced on the balloon?
16:47:55 <alise> ais523: although it's only barely resting on the hot air balloon IIRC
16:47:57 <Vorpal> ais523, I remember having two dreams tonight, one when the alarm clock woke me up, and one when it woke me up again after 5 minutes of snooze
16:48:01 <alise> so there's no real reason it should stay up for that long
16:48:02 <ais523> that could be an interesting situation to set up in RL
16:48:04 <Vorpal> ais523, I don't remember what they were about
16:48:06 <alise> but this *is* a dream
16:48:12 <alise> ais523: interesting, ha, you'll never see me on it :D
16:48:13 <ais523> dream physics is different
16:48:15 <alise> ais523: anyway it was ridiculously high
16:48:37 <alise> I looked down and big landmarks were like my thumb and ... whatever the finger next to the thumb is curled together
16:48:41 <ais523> alise: "ridiculously high" is about as accurate as you can get for dream physics
16:49:02 <alise> The Millennium Dome wouldn't be very big, even.
16:49:15 <alise> I was *seriously* high up.
16:49:16 <Vorpal> as for nightmares, haven't had one that woke up and/or that I remembered for months. Probably years
16:50:08 <alise> ais523: Anyway, it's weird -- once the ladder tipped forwards, it basically disappeared
16:50:09 <ais523> strangely, the detail that normally causes me to reality-check and wake up is something really minor
16:50:13 <Vorpal> <ais523> alise: "ridiculously high" is about as accurate as you can get for dream physics <-- hm
16:50:14 <yorick> last dream I remembered was having a conversation with the news-guy on the alarm clock when it was about to wake me up yesterday
16:50:24 <ais523> alise: of course it did, a ladder that tall can't stand on end without something to support it
16:50:24 <Vorpal> ais523, I have a vague memory of seeing an altimeter in a dream
16:50:26 <ais523> so it clearly doesn't exist
16:50:35 <alise> I rarely wake up after a reality check, but lucidity never lasts long.
16:50:40 <alise> ais523: I never said I became lucid
16:50:42 <ais523> (note: this logic /actually works/ in dream physics)
16:50:53 <alise> ais523: I've always wanted to end a lucid dream by setting pi to 3.
16:51:05 <alise> I imagine everything would explode in a whirl of circles and I'd wake up as my dream physics engine crashes.
16:51:05 * yorick has never done a reality check that came out negative :/
16:51:09 <Vorpal> ais523, it was reading out something like "8238aj and half a donut" or something equally ridiculous though, though it seemed normal in the dream
16:51:16 <alise> yorick: try looking at a clock, look away, check it again
16:51:20 <alise> reading is impossible in dreams
16:51:27 <ais523> yorick: well, it mostly happens for me when not lucid; I'm not sure if I've ever been properly lucid
16:51:33 <yorick> either I just go "wth...I'm dreaming!", or "I must be dreaming, lets try a reality check...no it never works"
16:51:35 <ais523> but sometimes I randomly decide to reality-check and it comes out negative
16:51:35 <alise> you can only focus four words at a time and you can never seem to read them, and if you look even to the right a bit then back they'll have changed
16:51:44 <ais523> and that always causes me to instantly wake up
16:51:45 <alise> yorick: you need to reality check in *usual* situations
16:51:49 <alise> the whole point is that dreams never seem unusual
16:51:50 <yorick> alise: I tried that once, it pointed to 4:01 twice
16:51:51 <cpressey> I had a dream last night with a terrible book about Python in it. (The text was English prose, but it was syntax-highlighted similarly to Python...)
16:51:52 <alise> so you can't rely on that
16:52:04 <ais523> cpressey: you've been using pastebins too much
16:52:12 <alise> Phantom_Hoover: i had an awesome dream recently
16:52:18 <Vorpal> alise, no reading is possible, as long as you don't try again or actually try to read. I dreamed reading street signs, as well as numbers a few times.
16:52:22 <alise> there was some sort of thing similar to a zombie apocalypse but not quite
16:52:30 <alise> it was like being in an action film, you know you can't die, it's just awesome
16:52:35 <cpressey> ais523: I've been looking at too much Python, that's for sure.
16:52:41 <Vorpal> alise, I didn't actually read though, just kind of dreamed that I had
16:52:42 <alise> also I did LSD twice... not sure why
16:52:53 <alise> it seemed like a good idea at the time
16:52:56 <yorick> alise: clocks work in my dreams, so does reading
16:53:04 <alise> ais523: (what was "bad idea" to?)
16:53:24 <ais523> instead of getting lucid dreams, you get non-lucid real life
16:53:29 <yorick> it feels strange to see it being 4:01 pm on sunday, then waking up and, 8 hours later, see it being 4:01 pm on sunday again
16:53:43 <quintopia> also, there's an experiment someone did once where they'd present some text on a screen to read, and they'd do pupil tracking, and every time they caught a saccade, they'd replace what the viewer was just looking at with different text.
16:53:45 <ais523> which is kind-of the worst of both worlds
16:53:45 <alise> ais523: I'd say ego death is a bit more than simple non-lucidity...
16:53:50 <Vorpal> alise, trying LSD in a dream or?
16:54:04 <yorick> quintopia: that's just horribly evil
16:54:06 <quintopia> and because the eyes can't really detect such subtle changes when saccading, it was really disturbing trying to read it...
16:54:07 <ais523> quintopia: that seems evil
16:54:12 <alise> ais523: besides, I'm pretty sure you can't think "hey, this isn't realistic" when you're on LSD
16:54:32 <quintopia> ais523: i don't even know what they were trying to test for. i should look it up again.
16:54:36 <ais523> I have almost a religious level of horror/abhorrance at things that affect my ability to think straight
16:54:46 <alise> ais523: but that doesn't matter because you'd be idiotic to do it without someone who's done it before around.
16:54:47 <ais523> quintopia: who cares, that's a great experiment anyway
16:54:50 <alise> ehh, it's on my list of things to try some day
16:54:53 <Phantom_Hoover> "it's the .NET framework, most people already have it on their computer" how does someone this stupid write anything that comes near working?
16:54:55 <alise> ego death sounds fun
16:54:57 <quintopia> i should try LSD sometime. i need a good babysitter tho
16:55:13 <ais523> Phantom_Hoover: hmm, most people with a computer probably /do/ have the .NET framework installed
16:55:19 <quintopia> you've seen the LSD sketching experiment haven't you?
16:55:24 <ais523> I'm assuming that most Windows XP users have needed it for something by now
16:55:32 <quintopia> ...but not with quite as high a dose
16:55:38 <ais523> meh, I have Mono installed even though I'm on Linux
16:55:53 <ais523> due to an occasional need to run .NET programs
16:56:04 <alise> quintopia: i'm not sure that was lsd
16:56:17 <alise> quintopia: if it was, it must have been a mild dose, surely
16:56:23 <ais523> Phantom_Hoover: it's a .NET bytecode interpreter, plus libraries
16:56:26 <alise> quintopia: i doubt anyone could draw on a regular dose of LSD.
16:56:38 <quintopia> alise: you've clearly not seen the experiment
16:56:52 <alise> quintopia: but i'm having a hard idea of perceiving someone pick up a pencil and put it on paper
16:57:10 <quintopia> well, they basically had to force him to, and at a certain point even that didn't work
16:58:02 <alise> high doses of LSD are interesting, since the active dose is so incredibly, ridiculously small, but the fatal dose is ridiculously high
16:58:33 <quintopia> i think they gave the guy like 100mg or something like that
16:58:42 <alise> I can't imagine how they could possibly do anything more than a regular highish dose, though...
16:58:53 <alise> i mean, i'd say that's fairly close to the most anything can do to you :P
16:59:19 <alise> anyway in the dream LSD was pretty boring really, everything was just a certain colour and like a day passed in a few minutes and i was back where i started
16:59:28 <alise> the non-zombie 'pocalypse was much more fun
17:00:14 <quintopia> i was close. apparently it was 100 g
17:00:14 <ais523> alise: I assume that if you actually fell asleep while on LSD, the dreams you got (if any) wouldn't be that different from normal dreams anyway
17:00:31 <ais523> quintopia: only out by a factor of 1000!
17:00:35 <ais523> that's close in one sense, but not in another
17:00:41 <alise> ais523: well, in the first LSD trip, Albert Hofmann fell asleep (after a bad trip)
17:00:51 <quintopia> ais523: nah, it's only 3 orders of magnitude
17:00:54 <alise> ais523: and then woke up feeling tingly and joyful
17:00:57 <alise> after an uneventful night
17:00:59 <quintopia> 1000 sounds like such a big number. 3 is better
17:01:13 -!- tombom has joined.
17:01:21 <alise> 100 mg of LSD is well below the fatal dose, I think
17:01:24 <ais523> yes, but orders of magnitude can be so much larger than individual units
17:01:28 <alise> (which is comparable to other drugs with active doses in the mgs)
17:01:33 <alise> (rather than the ... mugs)
17:02:02 -!- sshc has quit (Read error: Connection reset by peer).
17:02:03 <alise> http://upload.wikimedia.org/wikipedia/commons/3/3e/Drug_danger_and_dependence.png
17:02:06 <alise> active vs lethal doses
17:02:10 <ais523> it's worrying enough that you can be affected by a microgram of anything, if you think about it; humans have a chemical balance so complex it can be upset by even that small an amount of a chemical
17:02:14 <alise> doesn't give absolute values, just the ratio
17:02:21 <alise> but LSD has the lowest
17:02:30 <alise> surprise, surprise, heroin has the highest
17:03:10 <alise> [[Typical doses in the 1960s ranged from 200 to 1000 µg while street samples of the 1970s contained 30 to 300 µg. By the 1980s, the amount had reduced to between 100 to 125 µg, lowering more in the 1990s to the 20–80 µg range,[14] and even more in the 2000s.[15] [16]]]
17:03:18 <alise> this makes it hard to figure out the fatal dose since it depends on the figures the graph is using :D
17:03:35 <alise> "Estimates for the lethal dosage (LD50) of LSD range from between 200 µg/kg to more than 1 mg/kg of human body mass, though most sources report that there are no known human cases of such an overdose. Other sources note one report of a suspected fatal overdose of LSD occurring in November 1975 in Kentucky in which there were indications that ~1/3 of a gram (320 mg or 320,000 µg) had been injected intravenously. (This is a very extraordinary amount, partic
17:03:35 <alise> ularly when compared to the average LSD dosage of ~100 µg)."
17:04:31 <ais523> by comparison, the lethal dose of water is around 8 kg
17:04:43 <ais523> and even then, the antidotes are relatively simple and readily available
17:04:58 <alise> ais523: are you trying to say
17:05:00 <alise> "don't take LSD, take water"?
17:05:02 <ais523> (pretty much anything that dissolves in water works, as long as it isn't dangerous to eat itself)
17:05:09 <ais523> alise: I'm just trying to draw a comparison
17:05:25 <tombom> what if, like, we're all taking hundreds of ml of a drug every day??
17:05:36 <ais523> meh, I can get drunk on water pretty effectively anyway
17:05:43 <tombom> and we're all hallucinating and lsd reveals the real stuff...
17:05:50 <alise> tombom: I N C E P T I O N
17:05:51 <tombom> what if the drug is.... water
17:06:09 -!- sshc has joined.
17:06:16 <cpressey> So what is this Interpol language, and howcum it's not on the esowiki?
17:06:27 <ais523> there's an esolang called Interpol?
17:06:38 <quintopia> there's one with a simular name...
17:07:04 <cpressey> ais523: there's a *language* called Interpol, and it *looks* kinda esoteric: http://www.nyx.net/~gthompso/self_ipol.txt
17:07:11 <alise> it would be interesting to see someone try 1 mg of LSD (10x the active dose; absolute minimum estimated lethal dose is 200 µg/kg, so you'd have to be something like 5kg for this to kill you)
17:07:33 <alise> that's not a similar name, that's just oklopol
17:07:42 <alise> cpressey: well, you know Brian Raiter
17:07:51 <alise> and if he codes in it...
17:07:54 <quintopia> oklopol, interpol. . .same difference
17:08:01 <alise> quintopia: oklopol is a person
17:08:03 <alise> cpressey: i linked that yesterday btw
17:08:11 <alise> the last quine is my favourite
17:08:16 <cpressey> alise: yes. that's why it's in my browser. mocking me.
17:08:19 <alise> harfharfharfharfharfharfharfharfharfharf!
17:08:49 <alise> cpressey: gthompso@nyx.net
17:08:56 <alise> or ask brian himself
17:09:00 <ais523> alise: Brian Raiter works at Google, I doubt all his programming is in esolangs
17:09:15 <alise> breadbox [whirlpool] muppetlabs [spot] com
17:09:19 <alise> ais523: shaddap :)
17:09:33 <ais523> alise: I didn't get a reply last time I sent a query there
17:09:37 <ais523> although I can't remember what it was about
17:10:05 <ais523> I never got a reply from the author of that WP7 INTERCAL interp either
17:11:11 <ais523> hmm, according to a Reddit comment, ads on domain-parked pages have a click-through rate of 50-80%, because there's nothing else to click on
17:11:21 -!- pineal_aenimal has joined.
17:11:35 <ais523> I can believe that the average Internet user doesn't realise that not every page needs to have a link followed from it
17:12:44 <ais523> quite a few people commenting in the thread were actual former domain parkers
17:13:19 <quintopia> you'd have to pick a good domain to get good results with that tho
17:13:50 <ais523> yep, a high conversion rate is pointless if people never visit the site in the first place
17:16:00 <alise> ais523: Underload optimisation; discuss.
17:16:08 -!- oerjan has joined.
17:16:29 <ais523> alise: apart from the S command, you can do quite a bit due to the side-effect-free nature of the language
17:16:41 <alise> ais523: the S command is sort of the point :P
17:16:44 <ais523> given that going below the end of the stack crashes the program, any subprogram can only access finite stack
17:16:56 <ais523> thus, it can be seen as a function
17:17:27 <ais523> so Underload optimization works much like Haskell optimization, you just need to deal with the Ses somehow
17:18:23 <oerjan> S'es are simple, just treat the whole program as producing a list of output, surely?
17:18:49 <ais523> oerjan: that's effectively making them into a monad, which is one possible solution but one I've never seen used
17:19:16 <cpressey> i was actually wondering last night how to make oerjan's kolakoski sequence generator manage to output more than a dozen digits on my C impl of underload
17:19:43 <alise> cpressey: tail recursion
17:19:51 <cpressey> alise: it does do tail recursion
17:20:22 <ais523> tail concatenation helps, too
17:20:26 <alise> ais523: Unlambda compilation; disgust.
17:20:29 <oerjan> well i don't recall the bots having a problem with it...
17:21:03 <cpressey> alise: or do you mean, write the underload program tail-recursively? i don't even know how that would be possible
17:21:08 <ais523> have I actually posted the source to derlo anywhere, yet?
17:21:20 <ais523> cpressey: tail-recursion in Underload is a ^ just before a )
17:21:42 <ais523> e.g. the standard infinite while loop is (:^):^, and any sane interp should be able to run that forever without overflowing stack, etc
17:22:00 <ais523> well, I should just say "infinite loop", there's nothing /that/ while-loopy about it
17:22:08 <oerjan> it still shouldn't crash on a dozen digits, i doubt my program is _that_ space leaky :D
17:23:24 <alise> ais523: (^), the most pointless program ever; discuss
17:23:32 <alise> although (^)* might be useful for something
17:23:35 <alise> although what i know not
17:23:39 <alise> a(^)* is definitely silly, though
17:23:48 <ais523> it's just a fancy identity
17:24:08 <ais523> (^) is plausible for use as a data element in some encoding scheme
17:24:20 <ais523> IIRC one of oerjan's programs used it, although I can't remember the context
17:24:37 <ais523> arguably, (^) and (!) could be used for true and false
17:24:48 <oerjan> the rule 110 one, probably
17:25:14 <alise> true = (~!^); false = (!^), imo
17:25:17 <alise> at least useful true and false :P
17:25:32 <ais523> alise: I normally use either those, or () and (!())
17:25:42 <alise> those are confusing
17:25:43 <oerjan> : and ^ were the simplest way to encode data using just 1 char per bit
17:25:43 <cpressey> wow. i totally cannot think today.
17:25:44 <ais523> if you append ^ to the second set, you get (^) and (!)
17:25:47 <alise> hmm, what's zero and one?
17:25:54 <ais523> (!()) and () respectively
17:25:59 <quintopia> (^) almost reminds me of the standard Imp. Someone should make a Underload Battle Arena for fighting Underload programs.
17:25:59 <alise> there you go then :P
17:26:08 <oerjan> : and a were also possible iirc but more messy
17:26:11 <ais523> they make nice booleans because "loop n times" is easy in Underload
17:26:18 <ais523> so an if statement is "loop 0 times" vs. "loop 1 time"
17:26:23 <alise> ais523: i don't like how zero's different :<
17:26:30 <alise> it's unlike, e.g. the lambda calculus
17:27:53 <ais523> alise: it isn't really: you can construct the numbers as 0 = (()'!_), 1 = (()':_*'!_), 2 = (()':_*':_*'!_), etc
17:27:56 <alise> does anyone here like Logo?
17:28:03 <ais523> they just happen to optimise into much neater forms
17:28:06 <alise> ais523: hey, no using underlambda syntax
17:28:30 <ais523> (that's underlambda; 'x = (x); _ = ~a*^)
17:28:59 <ais523> hmm, I'm missing some ~s there
17:29:03 <ais523> cpressey: it's far from finished
17:29:07 <alise> cpressey: vapourware
17:29:07 <ais523> and I tend not to put partial langs up there
17:29:15 <ais523> alise: less vapourware than Feather
17:29:21 <ais523> because at least I have an idea where it's going
17:29:21 <alise> except with ais523 you can get close to the vapour and inhale
17:29:30 <alise> and for a few lovely minutes, your brain doesn't work at all
17:29:39 <quintopia> ais523: did you choose the underload command names so that source would naturally be littered with emoticons, or was that just happenstance?
17:29:41 <alise> as you try to understand retroactive non-synchronicitic variable term rewriting of the past
17:29:44 <ais523> besides, aren't all my esolangs vaporware at some point
17:29:52 <alise> quintopia: it was clearly designed for (:aSS):aSS
17:29:57 <ais523> quintopia: I tend to gravitate towards punctuation marks
17:30:02 <alise> you can even make an argument that that can be read as
17:30:06 <ais523> the aSS thing is actually entirely concidental
17:30:07 <alise> "push double ass, double ass"
17:30:21 <ais523> if it were deliberate, it would have been properly capitalisd
17:30:40 <alise> (because : is dup)
17:30:57 <alise> ais523: any objections to me removing [[Underload#Self-interpreter]]?
17:31:00 <alise> it's ridiculous and blatantly false
17:31:18 <ais523> it's "correct" in a joke-esolang sort of way
17:31:30 <alise> ais523: right, but keymaker was serious at the time :)
17:31:30 <ais523> but pretty much every lang has a self-interp on that basis
17:31:43 <alise> Ah, my bad. I honestly didn't think about it, this idea of passing control is obvious now, but quite new to me. I was thinking along the lines that if the data gets run, it's interpreted. :) Anyways, how would one convert the Underload program to Church numerals? (I have no idea about those.) And would some other encoding be ok (probably would)? And what would it be if my Underload-interpreter-in-brainfuck was modified to have the program we want to execute
17:31:43 <alise> directly in the memory without the interpreter reading it from user, and that brainfuck program was then converted to Underload with ais523's brainfuck-to-Underload program. Would that suffice? --Keymaker 14:37, 8 January 2008 (UTC)
17:31:59 <ais523> the problem is that there are several different conceputal ways to think of Underload
17:32:07 <ais523> and that method is perfectly natual in some, and abhorrent in others
17:32:20 <alise> "HTML source code" --you
17:32:24 <ais523> (S is a wart on the lang, in that it screws some of them up; Underlambda is going to redefine output to fix that, I just haven't decided how yet)
17:32:34 <ais523> alise: well, it's hardly a binary
17:32:37 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
17:32:45 <alise> ais523: "HTML source" i would have accepted
17:32:53 <quintopia> so what would the correct way of writing a program that recursively interprets itself in underload (the same way that one that what's his name wrote for C does)?
17:33:02 <alise> ais523: it's markup!
17:33:07 <alise> okay so i'm not normally this anal
17:33:11 <alise> "HTML source code" just threw me off
17:33:20 <Vorpal> is golfscript usable for anything but golfing?
17:33:21 <ais523> quintopia: you'd have to write an interp for something, then quine into the lang that interp used as input
17:33:29 <alise> i vaguely recall that
17:33:34 <ais523> Vorpal: obviously, it permits embedded Ruby
17:33:51 <Vorpal> ais523, okay but apart from that, which is kind of cheating.
17:34:00 <ais523> which reminds me, I want to make a usable application in pure CSS (plus a DOM element to start it off) someday
17:34:10 <ais523> even if it's sub-TC, you can still create something useful
17:34:20 <Vorpal> ais523, what would it do?
17:34:31 <alise> ais523: well, you have content: and attr() (I think it's spelled like that)
17:34:33 <ais523> some sort of state machine, I think
17:34:36 <alise> as well as hover and focus
17:34:42 <quintopia> i was thinking of madore's scheme self-interp
17:34:42 <ais523> I was planning to use mouseovers for input
17:34:43 <alise> so you can do some things with it
17:34:54 <alise> ais523: oh, and the custom-defined list numbering in CSS(is it 3? I think so)
17:34:57 <alise> lets you do ridiculous things
17:34:58 <cpressey> quintopia: underload interpreter in underload?
17:35:01 <quintopia> is there a really short way to do that in underload?
17:35:14 <alise> quintopia: an underload self-interp is non-trivial
17:35:21 <alise> although not that non-trivial
17:35:22 <ais523> alise: hmm, not really
17:35:29 <alise> ais523: mm, maybe not
17:35:44 <ais523> you just need a lookup table and a bunch of concatenations
17:36:21 <cpressey> that reminds me, i want to do the meta-circular thing with a rewriting lang someday
17:36:33 <cpressey> but underload is somewhat interesting too
17:37:02 <quintopia> remind me what meta-circular means? is that the name for the "evaluate string" as flow control idea?
17:37:06 <oerjan> choosing the input encoding looks like the hard bit ;D
17:37:18 <alise> <ais523> you just need a lookup table and a bunch of concatenations
17:37:23 <alise> oh i thought it was about css counter stuff
17:37:26 <alise> but it's non-trivial, certainly
17:37:27 <cpressey> quintopia: an interpreter for language X, written in language X, basically
17:37:28 <alise> trivial is trivial
17:37:30 <ais523> ^ul (!())((zero)(!^)(one)(!^)(two)(!^)(three)(!^)(four)(!^)(five)(!^)(six)(!^)(seven)(!^)(eight)(!^)(nine)(!^)(ten)())~(!!)~^^S^
17:37:30 <fungot> (zero)(!^)(one)(!^)(two)(!^)(three)(!^)(four)(!^)(five)(!^)(six)(!^)(seven)(!^)(eight)(!^)(nine)(!^)(ten)() ...out of stack!
17:37:44 <alise> http://en.wikipedia.org/wiki/Meta-circular_evaluator
17:37:50 <alise> that's just a self-interpreter
17:37:57 <ais523> also, wrote the table backwards
17:38:09 <alise> I hate it when useful distinctions are lost due to people misusing terms...
17:38:25 <oerjan> yeah that's ironic *ducks*
17:38:47 <alise> oerjan: It's like rain / on your wedding day
17:40:04 <ais523> ^ul (!())(:*:*)(::**)(:*:*:*)(~(()(ten)(!^)(nine)(!^)(eight)(!^)(seven)(!^)(six)(!^)(five)(!^)(four)(!^)(three)(!^)(two)(!^)(one)(!^)(zero))~a*^(!!)~^^( )*S^:^):^
17:40:04 <fungot> eight three four zero ...out of stack!
17:40:41 <alise> ais523: proposal: remove () from underlambda
17:40:42 <ais523> ^ul (::**)(:*)(::*:**)(~(()(ten)(!^)(nine)(!^)(eight)(!^)(seven)(!^)(six)(!^)(five)(!^)(four)(!^)(three)(!^)(two)(!^)(one)(!^)(zero))~a*^(!!)~^^( )*S^:^):^
17:40:43 <fungot> five two three ...out of stack!
17:40:44 <alise> ' and * make it redundant
17:40:53 <ais523> alise: () is still useful, though
17:41:05 <ais523> (also, you need a as well to make it redundant)
17:41:18 <ais523> Underlambda is somewhat golfed
17:41:24 <ais523> and parens are useful for that
17:41:37 <alise> ais523: is there a non-()-using program that pushes () on the stack?
17:41:39 <alise> I don't think so, but...
17:41:46 <quintopia> cpressey: so the distinction is that when an interpreter provides direct access to it evaluation methods, one can write an interpreter for that interpreter that really just asks the parent interpreter to do interpretation?
17:41:50 <alise> (abcd) -> ':'!*'a*'b*'c*'d* :-D
17:41:54 <alise> quintopia: sort of
17:41:59 <ais523> no, but only for the trivial reason that the first command in any Underload program has to be ()
17:42:09 <ais523> or something containing ()
17:42:12 <alise> quintopia: it's like writing a scheme interpreter in scheme that just handles lambda and similar control structures, making them into real lambdas, and passes off the actual evaluation to APPLY
17:42:27 <ais523> in Underload, there's a command to push () onto the stack, it's called 1
17:42:30 <alise> i.e. it does name lookup, basic control structures, and things like macros if it does them; but actual (f x y z) is handled by (apply f (list x y z))
17:42:43 <alise> it's a self-interpreter that uses the language's actual interpreter to do most of the work for it, basically
17:42:54 <ais523> alise: that's a correct sort of interp, though, and often is used to interp one lang in another
17:42:56 <quintopia> yeah that's what i was trying to say
17:43:00 <alise> <ais523> no, but only for the trivial reason that the first command in any Underload program has to be ()
17:43:30 <ais523> in Underlambda, you can do ((x)!) or something like that
17:43:42 <ais523> because it's only the semantics of a codeblock that matter, not its literal representation
17:43:49 <ais523> cpressey: what don't you get?
17:44:17 <alise> <ais523> because it's only the semantics of a codeblock that matter, not its literal representation
17:44:36 <cpressey> show me a self-interpreter that isn't meta-circular. to me, it's just a matter of degree (how direct is your circularity?)
17:44:53 <alise> cpressey: for instance, a Python-in-Python implementation that implements its own object system
17:44:57 <ais523> Underlambda's S command outputs functions, and its D command inputs them
17:44:58 <alise> and does tree-based AST interpretation
17:45:08 <ais523> it's interp-defined what format's used to output the functions
17:45:55 <ais523> anyway, there are any number of subsets of Underlambda that are TC
17:45:59 <quintopia> or any interpreter for B written in A, where A and B are unrelated, which is then interpreted by a A interpreter written in B.
17:46:04 <ais523> for a while, I even thought about embedding BF-minus-comments in it
17:46:17 <quintopia> that pretty much forces you not to ask the A interpreter for help
17:46:21 <ais523> but < and > don't fit well with the execution model without requiring a really complex definition of + and -
17:46:48 <ais523> quintopia: I was thinking about that just now; for Scheme, the sticking point seems to be closures in the intermediate language
17:47:10 <ais523> as in, a self-interp's considered to be a "true" self-interp if it goes via an intermediate representation that doesn't include closures
17:47:31 <cpressey> alise: in your example, the object system isn't directly meta-circular... but since it's still defined, indirectly, in terms of itself, i can't bring myself to call it "not" meta-circular
17:47:41 <quintopia> ais523: so Clojure is out of the question? that seems silly, since Clojure can be compiled to java bytecodes...
17:47:57 <alise> cpressey: You know, there is a use for terms that are not precise slicings of the world in two.
17:48:09 <alise> In this case, "metacircular" is obviously not precise, but even Lojban has vague adjectives.
17:48:27 <ais523> quintopia: it's a self-interp either way, but there's sort-of a distinction between "compile then execute", and "interpret without compiling"
17:48:40 <ais523> the first is often considered cheating, especially if the compile step is very simple
17:48:52 <cpressey> alise: you seem to feel it's precise enough to yell *wrong* at me when I do a first pass explanation
17:48:53 <ais523> on the other hand, it's not a straight distinction in that there's no cutoff, it's a continuum
17:49:14 <alise> cpressey: If I was yelling, I would have used uppercase. And it was more so that quintopia doesn't get confused.
17:49:17 <ais523> what does "meta-circular" mean anyway?
17:49:24 <ais523> I haven't been yelling about the definition because I have no idea what it is
17:49:26 <alise> ais523: http://en.wikipedia.org/wiki/Meta-circular_evaluator
17:49:41 <alise> i *did* link that definition before, but ofc you didn't see it :)
17:49:47 <ais523> ugh, clog's being slow
17:50:03 <alise> ais523: en.wikipedia.org/wiki/Meta-circular_evaluator
17:50:09 <alise> that filter causes you way more problems than it solves
17:50:11 <ais523> got it before you posted that
17:50:17 <ais523> alise: it's solved loads of problems for me
17:50:27 <Vorpal> argh, I'm away for three minutes and I come back to several screenfuls
17:50:29 <quintopia> ais523: do you filter them to empty string or to [missing link] ?
17:50:31 <Vorpal> anyway, bbl for quite a bit
17:50:32 <ais523> the problems it causes are problematic, but not that bad
17:50:40 <ais523> empty string would be rather confusing
17:50:40 -!- oerjan has quit (Quit: I never meta-circ).
17:51:02 <alise> ais523: also relevant: http://goatse.cx/
17:51:10 <alise> quintopia: he doesn't like links
17:51:14 -!- alise has left (?).
17:51:16 -!- alise has joined.
17:51:16 <ais523> people pestering me, mostly
17:51:20 <cpressey> alise: i apologize for spreading confusion.
17:51:27 <alise> cpressey: Sacrifice goats!
17:51:37 <ais523> the vast majority of IRCers assume that if they post a link, everyone will read it
17:51:55 <ais523> making myself unable to follow them at least gives me a plausible excuse
17:52:05 <ais523> to say "no I haven't read it because I filter links, and I don't particularly care anyway"
17:53:06 <ais523> alise: hmm, the wikipedia article seems to consider meta-circular evaluators as not being self-interps
17:53:12 <ais523> whereas I see them more as a special case
17:53:18 <alise> ais523: err, the wikipedia article doesn't say that
17:53:21 <alise> and you're obviously right
17:53:27 <quintopia> if i were to do the same i'd have my client go ahead and fetch the link up until it reaches a <title> tag, then filter the link to the title
17:53:33 <ais523> "The difference between self-interpreters and meta-circular interpreters is that the latter restate language features in terms of the features themselves, instead of actually implementing them. (Circular definitions, in other words; hence the name). They depend on their host environment to give the features meaning."
17:53:35 <ais523> oh, it's a direct quote
17:53:49 <quintopia> that way i can know what the link was to, while still being able to deny having received a link
17:53:50 <ais523> sorry, Wikipedia, it's actually Reginald Braithwaite who I disagree with
17:54:12 <ais523> but there are some sites I don't even want to send TCP requests too
17:54:23 <ais523> the firewall here is somewhat insane
17:54:33 <alise> someone's been lying to the pavement again
17:54:51 <alise> ais523: ah, what raganwald meant there is
17:54:57 <alise> "difference between them and REGULAR self-interps"
17:55:05 <ais523> could just be lack of context
17:55:24 <alise> he's a smart guy, so
17:55:46 <ais523> now I want to write an Underload quine where every string is treated either entirely as data, or entirely as code
17:56:05 <ais523> as in, either S is never run on it, or ^ is never run on it
17:56:47 <ais523> it probably wouldn't fit into one line of IRC, though
17:57:45 <quintopia> ais523: do you have someone that goes through your web access logs seeing which domains you've accessed?
17:58:00 <ais523> quintopia: yes in theory
18:00:39 -!- aloril has joined.
18:01:25 <ais523> at least the firewall's stopped portscanning me
18:03:27 <ais523> (yes, all my ports are closed to anyone but 127.0.0.1. Why do you care? You're a NAT, it's not like anything would happen even if the ports were open, as I don't have a public IP...)
18:05:20 <alise> apparently y is not defined
18:05:22 <alise> stupid scoping rules
18:05:45 <alise> ais523: INTRA-UNIVERSITY ILLEGAL FILESHARING OVER PORT 453
18:06:16 <ais523> alise: I hadn't even thought of communicating with other people on the same subnet
18:06:18 <ais523> if indeed there are any
18:06:43 <ais523> surely they could just firewall it at the router if they cared that much?
18:06:54 <ais523> (as in, insist all traffic went to a different subnet?)
18:07:21 <alise> w[Y,X]=46;X-=1;w[Y,X]=64;D(0,0)
18:07:21 <alise> w[Y,X]=46;X+=1;w[Y,X]=64;D(0,0)
18:07:28 <alise> guess what it does!
18:07:45 <ais523> moving the character in a roguelike
18:07:58 <ais523> 64 is @, the characters are typical roguelike movement keys
18:08:11 <ais523> I don't know 46 off by heart, but suspect it's . based on context
18:08:25 <alise> i like how C is ais523's go-to calculator
18:08:34 <alise> ais523: it is, yes
18:08:36 <alise> guess what D does :P
18:08:40 <ais523> most langs don't convert between integers and characters transparently
18:08:55 <ais523> hmm, D is less obvious, especially as the params are always 0
18:08:59 <ais523> also, EgoBot isnt here
18:09:21 <alise> cpressey: yes; what are the arguments?
18:09:26 -!- Phantom_Hoover has joined.
18:09:39 <cpressey> offset into screen to start updating at, maybe?
18:10:42 <ais523> Phantom_Hoover: it's a language project by me
18:10:43 <alise> cpressey: here's D's definition:
18:10:45 <alise> for B in range(b,b+23):
18:10:45 <alise> for A in range(a,a+80):
18:10:45 <alise> s.addch(B-b,A-a,w.get((B-12,A-40),46))
18:10:51 <ais523> based around an esolang, also with the same name
18:10:51 <alise> note: b/B is y, a/A is x
18:10:55 <alise> just renamed them to avoid clashes
18:11:22 <ais523> the idea is to build an esolang that's easy to compile into other langs, and easy to compile other langs into
18:13:03 <Phantom_Hoover> So taking Brainfuck's niche as the standard language for proof by isomorphism?
18:13:16 <ais523> Phantom_Hoover: nah, most langs are hard to compile /into/ BF
18:13:22 <ais523> the idea is to make it work both ways
18:13:24 <quintopia> goddamit they set off the fire alarm again
18:13:39 <ais523> to get, eventually, an automatic converter between any two esolangs
18:13:46 <ais523> hopelessly inefficient, ofc, but who cares
18:14:09 <alise> yay, my roguelike now moves around properly
18:14:13 <alise> now to do scrolling
18:14:17 <alise> (it's on an infinite plane)
18:15:05 <Phantom_Hoover> I would capitalise the 't', but is there such a thing as non-Euclidean topology?
18:15:13 <alise> interestingly, in this game, pressing a direction key for long enough will cause you to run out of memory
18:15:16 <ais523> hmm, does Python accept thin-spaces for indentation?
18:15:16 <alise> as the sparse array is filled
18:15:35 <cpressey> Phantom_Hoover: i think topology might be non-Euclidean by default.
18:15:41 <ais523> it would be great if you could mix all the different space-widths in Unicode
18:15:42 <cpressey> in that, Euclid never touched the stuff
18:15:46 <alise> ais523: do you want to see what i currently have?
18:15:57 <ais523> I might take a look when I'm finished
18:16:06 <alise> ais523: it's 23 lines, FWIW
18:16:11 <ais523> hmm, I have a strange aversion to unreleased projects
18:16:24 <alise> it's finished as far as moving around goes ;)
18:16:25 <ais523> I tend to like to get things done first before an official release
18:16:28 <alise> well apart from scrolling
18:16:38 <ais523> yet, provide copies of the work-in-progress to anyone who asks
18:16:55 <ais523> (the only programs where people frequently have asked are jettyplay, and occasionally AceHack)
18:17:11 <ais523> I tend to incorrectly assume that the rest of the world operates like that, for some reason
18:17:24 <alise> if we're more then, let's say, 15 characters out of the centre
18:17:28 <alise> then scroll one place
18:17:33 <ais523> btw, we were discussing mono earlier? I seem to remember that the mono program I ran, I downloaded the source from codeplex.com
18:17:34 <alise> so X,Y are the centre
18:17:47 <ais523> and both their web-links and svn links weren't working properly, so in the end I used a recursive wget
18:18:59 <alise> note to self: make that faster
18:19:20 <alise> well that isn't working
18:19:40 <ais523> hmm, Slashdot are debating commercial breaks on television
18:19:55 <ais523> when I was in Canada (mostly receiving US TV channels), whenever a commercial break came on I changed channel
18:20:06 <ais523> and checked back a few minutes later to see if it was still there
18:20:30 <ais523> I mean, I do that in the UK too, except I rarely watch television there
18:20:33 <ais523> and tend to get stuck on the BBC
18:20:40 <ais523> Phantom_Hoover: it's long, frequent breaks
18:21:31 <alise> ais523: hmm, more frequent than ours though I think
18:21:48 <alise> e.g. Star Trek, over here, tends to do the abrupt-fade-out-then-in-again that usually signals an advert break, even when there's no adverts
18:21:51 <ais523> yep, typical in the UK is one or two minutes every 10-30 minutes, randomized
18:23:17 <alise> it's random? really?
18:23:19 -!- augur has joined.
18:23:31 <alise> ais523: also, not on Sky it isn't!
18:23:31 <ais523> alise: when I actually cared, which wasn't for very long, I didn't notice an obvious pattern
18:23:42 <alise> more like 3 minutes every 15 minutes
18:23:43 <ais523> I could only get the terrestrial channels then
18:24:04 <ais523> some of the adbreaks were extremely short, like 20 seconds altogether, but those were rather rare
18:24:17 <alise> most likely just sponsors
18:24:58 <alise> Phantom_Hoover: uh, occasionally.
18:25:09 <alise> why doesn't this work...
18:25:20 <alise> somehow decreasing x increases Y!
18:25:57 <alise> hmm, doesn't quite work with diagonals
18:26:10 * alise puts some random dust into the playfield to make it easier to see
18:26:27 <yorick> alise: it's overflowing!
18:27:13 <alise> oops, now it randomises every time :-D
18:27:21 <alise> and i leave a snail trail
18:29:01 -!- atrapado has quit (Quit: Abandonando).
18:29:33 <ais523> Phantom_Hoover: have you not seen Forte?
18:29:44 <alise> ais523: I have done more than the Crawl developers could.
18:29:45 <ais523> every command in that which isn't a no-op or simple I/O permanently gets rid of a number from that program
18:29:53 <alise> (Made a larger-than-screen playfield that scrolls non-annoying.)
18:29:56 <ais523> http://esolangs.org/wiki/Forte
18:30:02 <alise> That was not difficult!
18:30:16 <ais523> alise: NetHack scrolls like that too, but only if you're using an unusually small terminal
18:30:20 <ais523> (normally there's no need)
18:30:33 <alise> ais523: hmm, what values does it use instead of 15 and 5? I guess you're unlikely to know :P
18:30:56 <alise> i think 15 and 5 may be a bit too low
18:31:08 <alise> although 5 has the nice property that it scales well from 15
18:31:21 <alise> (the last line is reserved for my babble)
18:31:27 <alise> the terminal is 80x24
18:31:29 <alise> so it should be *23
18:31:49 -!- wareya_ has changed nick to wareya.
18:31:51 <alise> hmm, so it should be 4
18:33:05 <ais523> ugh, the code uses some sort of complex dead-reckoning for efficiency, which makes it hard to read
18:33:10 <ais523> seriously, NetHack, you micro-optimised /that/?
18:35:05 <alise> ais523: do you know the simplest way to get curses to just bloomin' restore the terminal at the end?
18:35:47 <ais523> it moves the screen left and right by 20 at a time if it gets within 5 of the left or right screen edge; vertically, by half the screen height (not counting topl, botl) if it gets within 2 of the screen edge
18:36:00 <ais523> alise: endwin does that automatically, or should
18:36:09 <ais523> as long as your termcap's set up correctly
18:36:18 <alise> you don't even need echo
18:36:46 <ais523> hmm, your code isn't identical to NetHack's code
18:36:59 <alise> ais523: do you know the values it uses?
18:37:05 <alise> i've done s/15/17/
18:37:06 <ais523> <ais523> it moves the screen left and right by 20 at a time if it gets within 5 of the left or right screen edge; vertically, by half the screen height (not counting topl, botl) if it gets within 2 of the screen edge
18:37:33 <alise> 1 at a time is the only way to avoid disorientation
18:37:40 <ais523> also, edge cases are handled in the actual code, but not in that description
18:37:53 <ais523> alise: well, Crawl scrolls 1 at a time, if it gets more than 0 from the centre
18:38:52 <ais523> so you're doing it in a mix between the NetHack and Crawl styles, which are close to being opposites
18:39:18 <alise> ais523: mine's the style that doesn't give you a headache ;)
18:39:39 <ais523> strangely, I found that in Enigma, which has level-configurable scrolling, the least confusing tends to be scrolling an entire screen instantly whenever you move within a few pixels of the edge (keeping one line)
18:40:09 <alise> ais523: nethack's is especially awful if you end up holding down one of the keys at the edge
18:40:16 <alise> since your character bounces around
18:40:43 <ais523> alise: well, bear in mind that NetHack levels are only 80 characters wide
18:40:54 <alise> whereas mine is infinite
18:40:55 <ais523> you'd bounce at most three times before you reached the other end of the level, no matter how small your terminal
18:40:58 <alise> and procedurally generated, hopefully
18:41:18 <alise> (the function to *draw the screen* will randomly assign stuff to unassigned cells in view, I think)
18:41:20 <ais523> they're all procedurally generated, pretty much
18:41:23 <alise> ais523: it's also a golfed roguelike :P
18:41:44 <ais523> hmm, I wonder if you could make the Mandelbrot set into a roguelike?
18:41:58 <ais523> place stairs on bits more detailed than the current zoom level
18:42:06 <ais523> going downstairs zooms in, upstairs zooms out
18:42:07 <alise> <ais523> place stairs on bits more detailed than the current zoom level ;; you mean every bit?
18:42:10 <alise> you can zoom in anywhere
18:42:17 <ais523> alise: I mean, where there's anything interesting
18:42:20 <ais523> rather than solid black or white
18:42:25 <alise> and i think if you use naturals instead of silly colours, you have detail everywhere but the centre
18:42:26 <ais523> that's rather more limited, to the "edge" of the set
18:42:28 <alise> very boring detail, but still
18:42:29 <Gregor> Just s/stairs/some magic/
18:42:52 <alise> ais523: Unfortunately, the Mandelbrot set is a bit slow to compute.
18:43:15 <ais523> I remember writing my own Mandelbrot program
18:43:22 <alise> heh, my program crashes with no message if you resize the terminal and do anything
18:43:29 <ais523> where you could zoom right in until you started hitting distortions due to floating point rounding
18:43:32 <alise> it just refuses to run with no message if your terminal is the wrong size
18:43:59 <ais523> the interesting thing is, the rounding errors created little sets of their own which looked like distorted Mandelbrot sets
18:44:09 <alise> were they fractal?
18:44:20 <ais523> yes, although not as detailed as the set itself
18:44:25 <ais523> eventually you ended up dividing by zero
18:45:55 <Phantom_Hoover> < pikhq> Microsoft Word: the worst program to design web pages in, and this *includes* Malbolge. ← you know how much I hated my school's computing course?
18:48:07 <ais523> I still maintain that MS Publisher is an /even worse/ program for designing web pages
18:49:04 <ais523> the markup is even less semantic than Word's (tables everywhere, even for simple text), the page is normally forced to a width different from that of your actual screen (Word doesn't do /that/), and it has a habit of randomly replacing text with images because it can't figure out how to render it as HTML
18:50:07 <alise> woot, cursor positioning works
18:52:00 <Vorpal> ais523, what is "MS Publisher"?
18:52:14 <Vorpal> ais523, I heard the name once or twice, but I have never seen it
18:52:15 <alise> Phantom_Hoover: Python :/
18:52:17 <ais523> Vorpal: be glad that you don't know the answer
18:52:20 <alise> Vorpal: "horrible"
18:52:36 <Vorpal> alise, okay but what role is it intended to fill?
18:52:39 <Vorpal> that is all I'm asking
18:52:42 <alise> `addquote <Vorpal> ais523, what is "MS Publisher"? <Phantom_Hoover> Vorpal, you don't want to know. <ais523> Vorpal: be glad that you don't know the answer <alise> Vorpal: "horrible"
18:52:42 <ais523> basically, you know when you open up a PDF in the GIMP or OpenOffice Draw or something like that?
18:52:59 <ais523> you get a bunch of maybe editable text, some images, all at exact locations on the page
18:53:06 <Vorpal> ais523, I opened pdf in inkscape, to edit some vector graphics
18:53:13 <ais523> inkscape will do fine as well
18:53:34 <ais523> now, you can open up printed documents like that, but it isn't too useful for actually understanding the document, agreed?
18:53:36 <alise> Vorpal: MS Publisher is a bunch of little Word documents arranged in absolutely-positioned boxes
18:53:39 <alise> with absolute sizes
18:53:44 <ais523> as in, you can copy-paste bits of text, but not much else
18:53:50 <ais523> MS Publisher is that in reverse
18:53:52 <HackEgo> 236|<Vorpal> ais523, what is "MS Publisher"? <Phantom_Hoover> Vorpal, you don't want to know. <ais523> Vorpal: be glad that you don't know the answer <alise> Vorpal: "horrible"
18:53:58 <ais523> as in, actually constructing printed documents via this method
18:55:02 <ais523> (fun fact: I actually had a useful printed document I made this way in Publisher ages ago, but it was too hard to transfer from one laptop to another, and almost impossible to edit in Linux; nothing reads .pub files, and the .ps files it outputs are absolute abominations)
18:55:07 <Vorpal> ais523, um what? Opening up pdfs can be useful I found it useful to open up that thing in there, the vector graphics had too thin lines, so when included in a latex document (as "Figure 1: Diagram from simulation showing current over time" or something like that) the lines were invisible.
18:55:11 <ais523> (so I recreated it in Excel, which was actually easier despite it being a text document)
18:55:24 <ais523> Vorpal: I mean, that's an insane way to /create/ a PDF
18:55:47 <Vorpal> ais523, the simulation program could only print the result iirc, not save it as an image
18:55:48 <alise> ais523: i *think* your method of explaining has a few too many steps of brainpower required to interpret.
18:55:53 <alise> try using words of two syllables or less
18:56:09 <ais523> alise: were you forced to use Publisher at school?
18:56:11 <Vorpal> ais523, yes indeed insane way to create pdfs though
18:56:19 <ais523> I think I was, but can't remember, I suspect my brain has erased memories of it
18:56:24 <alise> ais523: yeah. i'm talking about for Vorpal though
18:57:07 <ais523> Vorpal: Adobe Reader always shows lines at least a pixel thick (maybe even at least 2 pixels), even if they're thinner
18:57:15 <Vorpal> ais523, wait, can't you edit the text directly in publisher?
18:57:17 <ais523> I know, because I made a PDF with zero-width lines by mistake
18:57:23 <ais523> Vorpal: yes, just as you can in an opened PDF
18:57:37 <ais523> it looked fine in Reader, but broken in Sumatra
18:57:38 <alise> "My old school used Publisher incestuously!"
18:57:42 <Vorpal> ais523, well, I'm not sure what caused it. I was using evince to view it, and it had to be scaled to fit into the latex figure
18:57:54 <Vorpal> ais523, anyway I couldn't see the lines, and that trick worked perfectly
18:58:02 <alise> i like how Vorpal's just gone on to ignoring ais523's lines and relaying his anecdote instead
18:58:17 <Vorpal> alise, I did both at once
18:58:25 <Vorpal> alise, maybe you failed to keep up?
18:58:33 <ais523> alise: I'm semi-convinced that Vorpal permanently has his scrollbar a few lines from the bottom of the screen
18:58:39 <ais523> he seems to only say things said around 15-20 lines ago
18:58:42 <alise> ais523: or his brain just works *that* slowly
18:59:27 <Vorpal> ais523, anyway I did get a bit lagged when writing that long line about inkscape, and I didn't bother to read the rest until I finished that line
19:00:21 <quintopia> what annoys me is people that fill up the screen with related lines they could have combined in a single message
19:00:34 <ais523> Phantom_Hoover: I think my theory's correct
19:00:36 <quintopia> i'd rather read a few long lines than a lot of short ones
19:00:38 <pikhq> alise: Y'know, I've got half a mind to extend Quod Libet to handle video.
19:00:41 <ais523> assuming that there wasn't a response yet, but will be soon
19:01:01 <Vorpal> ais523, response to what?
19:01:06 <alise> <Vorpal> quintopia, hm indeed <-- *you* do it all the time
19:01:08 <ais523> Phantom_Hoover's message
19:01:14 <ais523> it probably hasn't scrolled onto your screen yet
19:01:22 <Vorpal> ais523, oh that, I ignored it
19:01:22 <quintopia> Phantom_Hoover: well, i don't complain about it do I? I just sit silently and be annoyed.
19:01:37 <quintopia> I'm making this one exception to make my peeve known and shall not mention it again
19:01:42 <alise> ais523: hmm, my roguelike actually uses the characters to determine what objects are
19:01:47 <ais523> alise: I can't sensibly claim victory in this argument because my own argument was self-contradictory
19:01:49 <ais523> but it's still hilarious
19:01:49 <pikhq> Or at the very least write a video player that actually uses MKV metadata.
19:01:51 <alise> quintopia: people don't think of things all at once
19:01:55 <alise> also, it speeds up conversation
19:01:56 <Vorpal> alise, ooh, are you coding one?
19:02:04 <alise> Vorpal: yes, it's golfed
19:02:14 <Phantom_Hoover> Vorpal, MY GOD, WHY AREN'T YOU EXACTLY FOLLOWING THE CONVERSATION
19:02:16 <alise> and played on an infinite plane filled with silly enemies and silly gold
19:02:18 <Vorpal> alise, cool, how extensive?
19:02:22 <alise> Vorpal: Umm... infinite
19:02:30 <quintopia> alise: speeding up conversation is exactly the part about it that annoys me. People should just be smart enough to have large gestalts
19:02:37 <Vorpal> alise, can you use the money for anything? Like shops?
19:02:41 <alise> quintopia: dude, this is IRC
19:02:46 <Vorpal> or would that take too much space?
19:02:57 <alise> it would be pretty boring if not
19:03:05 <alise> i'm not aiming for any absolute limit, just trimming down code wherever possible
19:03:07 <Vorpal> alise, and hm, is there any specific goal or does it just go on until you die?
19:03:08 <ais523> quintopia: what does "gestalt" mean in that context?
19:03:28 <Vorpal> alise, if the latter, the term "arcade rougelike" comes to my mind
19:03:33 <Vorpal> which seems rather silly
19:03:33 <alise> Vorpal: probably the latter
19:03:37 <alise> a victory condition would be too hard
19:03:39 <quintopia> ais523: the series of connective ideas from one thought to another
19:03:43 <alise> and require non-randomness of some sort
19:03:45 <Vorpal> alise, an arcadelike rougelike perhaps!
19:03:53 <alise> Vorpal: maybe after you get enough gold, you can go to a special boss level
19:04:43 <Vorpal> alise, not just teleporting straight away IMO. Would be better to spawn a teleporter or some stairs or something next to the player
19:04:51 * alise makes a ? tile which actually *is* undetermined right up until you hit it
19:05:00 <alise> Vorpal: yeah, there'd be like booths every now and then
19:05:03 <alise> with a quite low probability
19:05:05 <alise> that you'd have to go to
19:05:17 <alise> Vorpal: you know payphone booths?
19:05:28 <Vorpal> rather rare these days
19:05:33 <alise> Vorpal: it's a pay-TARDIS
19:05:37 <alise> they're common over here
19:05:45 <alise> Vorpal: yes: the portal to the boss is a pay-TARDIS
19:05:48 <alise> I can think of no better solution
19:05:49 <Vorpal> I haven't seen one for... 5 years or such?
19:06:03 <alise> open booth, bigger on the inside, insert coins :P
19:06:13 <alise> also has phone and internet
19:06:20 <Vorpal> alise, any leveling system?
19:06:42 <alise> How to weight probability: in your random selection, have more copies of one tile!
19:06:48 <Vorpal> alise, I feel it wouldn't be much of a rougelike without leveling and equipments and such
19:07:18 <Vorpal> alise, that is a reasonable way to do it yes
19:07:22 <alise> endwin() unfortunately erases the error message :P
19:07:34 <Vorpal> as long as you don't have something with like 1/1000th of the probability of the another thing
19:07:48 <alise> Vorpal: thankfully, i am far too lazy to have that
19:07:58 <alise> also, nobody would find it, the gameworld is too boring to explore _that_ long
19:08:16 <Vorpal> you still need those booth to be uncommon
19:08:41 <alise> Vorpal: hmm, in my test version i have some code i didn't change when the floor tile became space, not ., and it leaves a trail of .s behind you when you walk
19:08:43 <Vorpal> alise, as for crashing, hm... it managed to run endwin() after it crashed?
19:08:44 <alise> i think i might make that an item
19:09:00 <Vorpal> alise, oh like "red yarn" or something?
19:09:01 <alise> World's Largest Ball of Twine
19:09:36 <Vorpal> alise, a bit annoying if you don't move in a mostly straight or curving line
19:09:46 <alise> Vorpal: not really, it just helps you find where you went
19:09:54 <alise> which is useful if you're looking for that booth you saw seven screens ago
19:10:10 <Vorpal> alise, not if you criss crossed your path a lot before maybe hm
19:10:25 <alise> well, can't have everything.
19:10:35 <Vorpal> alise, you plan to use fixed screens? Not centering on the player all the time?
19:11:13 <alise> you can move around your centre area, but then it scrolls outside it
19:11:23 <alise> so it scrolls around just fine, but doesn't give you a headache (I'm looking at you, Crawl)
19:11:45 <Vorpal> alise, I find "always-centered-on-player" isn't too bad
19:11:59 <alise> remove the +n and -n if you want that :P
19:12:09 <Vorpal> alise, as long as you can see what you are moving to wards
19:12:34 <alise> the whole field is visible at all times
19:12:40 <Vorpal> alise, which language?
19:12:48 <alise> because, unlike NetHack, there is light, and you are not hideously short-sighted
19:12:50 <alise> Vorpal: python. meh.
19:13:11 <Vorpal> alise, that explains those "not very golf-y newlines" at least
19:13:32 <alise> len(';') == len('\n')
19:13:44 <alise> the indentation has an effect, but i can't avoid that
19:13:52 <alise> because it's an if
19:13:54 <alise> so a block structure
19:13:58 <alise> so i can't just do ; and more of them
19:14:05 <alise> first line of that function:
19:14:07 <alise> global x,y,X,Y;w[y,x]=46;x,y=a,b;w[y,x]=64
19:14:12 <Vorpal> alise, um you can, but not sure it helps that much. Remember that irc bot in python with just lambda?
19:15:25 <alise> to move away from there
19:15:40 <alise> T=' $$$$$%%%!' ;; I can't actually use this, it has to be charcodes, so:
19:15:41 <alise> T=[32]*10+[36]*5+[37]*3+[33]
19:16:01 <alise> ehird@dinky:~/Code/vagrant$ wc -c vagrant.py
19:16:11 <alise> Vorpal: [x]*n = [x,x,x,x...] n of them
19:16:18 <alise> [...]+[...] = [...,...]
19:17:06 <Vorpal> alise, also I assume the player is @ in the best of traditions? What about giving it a completely useless pet. Wait I'm detecting feature creep.
19:17:29 <alise> Vorpal: hmm, those weightings are a little off.
19:17:45 <alise> Vorpal: http://imgur.com/JTySy.png
19:17:47 <Vorpal> alise, I mean, nethack pets... just die a lot. When playing val I find it just gets in the way.
19:17:53 <Vorpal> for wiz it makes sense but...
19:17:57 <alise> they are very useful in sokoban.
19:18:08 <alise> (note: i have done that exactly once)
19:18:10 <alise> i had like four pets
19:18:13 <alise> all gained in sokoban
19:18:16 <Vorpal> alise, for curse testing?
19:18:20 <alise> the room at the end
19:18:25 <alise> i just let my four pets fight them all
19:18:27 <alise> didn't get a scratch
19:18:30 <alise> Vorpal: no, the monsters inside
19:18:32 <Vorpal> aren't the monsters all asleep
19:18:41 <Vorpal> or didn't you have stealth?
19:18:53 <alise> it's the harder version of the level, btw
19:19:21 <alise> Vorpal: but a monster got spawned on the Elbereth
19:19:23 <Vorpal> alise, pets near the end of game can be useful. I mean, tame archeon? Or tame ki-rin (probably only reasonable for knights)
19:19:27 <alise> and a werecreature stole it, I think
19:19:31 <alise> it certainly wasn't there
19:19:57 <alise> lol even changing the space weighting to 100 doesn't work
19:20:23 <Vorpal> alise, lucky you didn't do the expanded array thing then
19:20:36 <alise> T=[32]*1000+[36]*5+[37]*3+[33]
19:20:58 <Phantom_Hoover> Is there a way in Coq of bundling theorems together given some axioms?
19:21:07 <Vorpal> what did you say that the ! was?
19:21:07 <alise> Phantom_Hoover: yes
19:21:14 <alise> Phantom_Hoover: modules, "Parameter"
19:21:22 <alise> or was it sections? I forget
19:21:24 <alise> Vorpal: I didn't -- potion.
19:21:34 <Phantom_Hoover> i.e. you have some object Group which gives lots of goodies when passed the group, composition and axioms.
19:21:39 <Vorpal> alise, and $ is gold I presume... % is food?
19:21:58 <alise> the hp potion will most likely just incr an internal hp potion counter by something random
19:22:05 <Vorpal> alise, what weapons will be available?
19:22:13 <alise> and then quaffing a potion will heal min(sum, 15) or whatever
19:22:15 <Phantom_Hoover> Sections would seem to do it, but I don't know how they actually work.
19:22:20 <alise> Vorpal: your fists
19:22:25 <alise> Phantom_Hoover: see manual ;)
19:22:46 <alise> Vorpal: you could also think of it as a sword
19:22:51 <Vorpal> not really a rougelike without the equipment system
19:23:22 <alise> Vorpal: dude, it's close enough!
19:25:24 <alise> Vorpal: i don't even have walls
19:25:28 <alise> although i might add them.
19:26:40 <Vorpal> alise, need them for boss and booth at least
19:26:55 <alise> the booth will actually just be a single character.
19:27:04 <Vorpal> alise, yes but larger on the inside you said?
19:27:09 <alise> that's a TARDIS joke.
19:27:17 <alise> the effect of it being a pay-teleport will be conveyed entirely through one line of message :P
19:27:20 <Vorpal> but I thought that you would show it on screen
19:27:35 <Vorpal> alise, randomly placing walls would be silly, because sooner or later you would then run into a barrier you couldn't pass.
19:27:47 <alise> "You insert your gold into the slot. ... The door opens! --More--"
19:27:56 <alise> "Wow -- it's bigger on the inside! You see a big, shiny button. --More--"
19:27:57 <Vorpal> given perfect randomness and so on
19:28:02 <alise> "You press the button... --More--"
19:28:12 <alise> "Suddenly, you find yourself in a barren desert, with this evil guy."
19:28:24 <alise> Vorpal: so go in another direction :P
19:28:45 <alise> Vorpal: also, "technically", you could end up surrounded entirely by monsters
19:28:47 <Vorpal> alise, well sooner or later you will run into a wall that surrounds you
19:28:51 <alise> or there could never be a booth generated, ever
19:28:54 <alise> Vorpal: no, you won't
19:28:58 <alise> that's not even *close* to remotely probable
19:29:05 * Phantom_Hoover wonders if you can embed the hyperbolic plain into a terminal sensibly-ish.
19:29:05 <Vorpal> not probably at all indeed
19:29:17 <Vorpal> but sooner or later it will happen, given an infinite plane
19:29:23 <Vorpal> unless I'm completely wrong
19:29:45 <Vorpal> and perfect randomness of course
19:29:46 <alise> the probability approaches 1.
19:29:48 <Vorpal> Phantom_Hoover, yes it would
19:29:56 <alise> however, you do not have time to play that long
19:30:02 <alise> given, say, the predicted age of the universe.
19:30:11 <Vorpal> I never claimed it was likely
19:30:23 <Vorpal> alise, getting surrounded by walls at the start is more probable
19:30:35 <Vorpal> I mean, that is just 8 tiles
19:30:42 <alise> that's still hideously improbable :P
19:30:45 <alise> considering the low probability of walls
19:30:49 <Vorpal> than running into it at some distance
19:31:04 <Vorpal> alise, indeed, getting two wall segments next to each other would be low
19:31:10 <Vorpal> unless you try to deal with that
19:31:21 <alise> Phantom_Hoover: "small"
19:31:27 <Vorpal> alise, oh also: with walls you need LOS calculations
19:31:33 <alise> i will likely generate them separately or not at all
19:31:59 <alise> T=[32]*1000+[36]*5+[37]*3+[35]*3+[33]
19:32:03 <alise> except more likely than that, probably
19:32:14 <alise> Vorpal: or you have x-ray vision
19:33:03 <Vorpal> alise, I prefer the dungeon to be filled with strange cubes made of breakproof glass :P
19:35:00 <cpressey> "Note that if the attribute is found through the normal mechanism, __getattr__() is not called." <-- I love how this leaves what "the normal mechanism" is, up to the imagination.
19:35:26 * cpressey pulls a normal mechanism out of his pocket
19:35:32 * alise decides to have a dividing line before the status line
19:35:34 <alise> otherwise it's too confusing
19:36:23 <Vorpal> cpressey, I detect python naming there
19:36:38 <Vorpal> __slots__, __init__ and so on
19:37:12 -!- coppro has joined.
19:37:12 -!- coppro has quit (Changing host).
19:37:12 -!- coppro has joined.
19:38:12 -!- cheater99 has quit (Ping timeout: 240 seconds).
19:41:59 <cpressey> Vorpal: what else would I be carping about?
19:42:13 <alise> Vorpal: does banging into a wall in nethack affect the turns?
19:42:15 <alise> it doesn't, does it
19:43:08 <Vorpal> alise, you mean, if you advance a turn if you try to walk into a wall?
19:43:12 <ais523> alise: it erodes engravings, but has no other effect
19:43:16 <ais523> the erosion on engravings is probably a bug
19:43:20 <Vorpal> alise, I don't think I ever tried XD
19:43:29 <ais523> but is exploited to great effect by people reverse-engineering the RNG
19:43:40 <ais523> as it advances the RNG seed
19:44:55 <Vorpal> listened to an interview on radio with a professor in discrete math. Quite unusual.
19:45:50 <Vorpal> ais523, err, how could you know the seed
19:46:02 <ais523> Vorpal: well, it's seeded with the current date and time, right?
19:46:09 <alise> Vorpal: I have no message line, woo
19:46:17 <ais523> besides, seeds follow a pattern, set off enough random events and you can figure out where in the pattern you are
19:46:33 <ais523> someone made a bunch of rainbow tables, they had to change the RNG to reseed from /dev/random in order to block that on NAO
19:46:41 <Vorpal> ais523, is it time() or gettimeofday() ?
19:46:49 <Vorpal> I mean, the former you couldn't probably manage to figure out
19:47:02 <Vorpal> the latter has way too high res for you to figure out when it ran without a debugger
19:47:14 <ais523> Vorpal: the former, and you can get it within a few seconds pretty easily
19:47:31 <ais523> and within a second allowing for network lag and clock skew, which you can manage by seeing what happened with your failed attempts
19:48:05 <Vorpal> ais523, anyway how easy would it be to figure out the seed if it started off from /dev/random? I mean, you could probably still figure it out with a table
19:48:12 <Vorpal> looking up start sequences
19:48:14 <ais523> Vorpal: as I said, it was rainbow-tabled
19:48:32 <ais523> so on NAO, it reseeds from /dev/random every now and then, paxed's keeping the exact interval secret
19:48:49 <ais523> on /dev/null, it uses a cryptographically secure RNG seeded from /dev/random
19:49:14 <ais523> there's even a bunch of tables for doing AES quickly
19:49:44 <Vorpal> ais523, /dev/random is a bit iffy though, compared to /dev/urandom. It would get stuck quickly quite easily if many people start games at the same time
19:49:55 <cpressey> rainbow tables! man, i love that game
19:49:59 <ais523> agreed, probably /dev/urandom
19:50:09 <ais523> apparently using /dev/urandom directly was too slow
19:50:13 <cpressey> Vorpal: you know. rainbow tables!
19:50:22 <Vorpal> cpressey, yes, I know what they are. But "game"?
19:50:43 <cpressey> better than musical chairs even!
19:50:46 <Vorpal> I thought cpressey meant a game called "rainbow tables"
19:50:57 <alise> Okayyy, my M function is fucked up.
19:51:06 <Vorpal> alise, what does M do?
19:51:13 <alise> and updates everything :P
19:58:07 <alise> def Q(x):s.move(23,0);s.insertln();s.addstr(23,0,x);s.getkey()
19:58:12 <pikhq> alise: Why does everything suck?
19:59:09 <cpressey> class object: def __suck__(self): return True
20:04:02 -!- augur has quit (Ping timeout: 264 seconds).
20:04:05 <alise> ais523: L+=int(not randint(0,3)and randint(5,10))
20:04:19 <ais523> is that a bitwise and?
20:04:52 <ais523> as in, it returns the right argument only if the left is 0
20:05:03 <alise> and int(True)=1, int(False)=0
20:05:05 <ais523> in that case I don't get the not
20:05:13 <alise> ais523: i'll elaborate on the logic:
20:05:21 <alise> "1/3 chance: increase HP by random in range 5 to 10"
20:06:11 <ais523> oh, and most of the time nothing happens
20:07:13 <alise> but it should be 13
20:09:21 <alise> that's some hunger
20:11:10 <alise> ais523: hunger increases properly unless i hold down for a while
20:11:19 <alise> in which case it stops increasing, then goes to something random when i move in a different direction
20:12:47 <alise> Vorpal: want to see the WIP version?
20:12:58 <alise> Phantom_Hoover: everyone: you too
20:13:36 <alise> ais523: HOW could hunger possibly go down like that?!
20:14:00 <ais523> I take it you're not mixing longjmp and autos, either
20:14:14 <ais523> in which case, the random number is probably significant, you should figure out what it's referring to
20:14:16 <alise> ais523: in Python?
20:14:21 <alise> also, it's not random
20:14:23 <alise> it actually decreases
20:14:36 <ais523> have you used the wrong variable name somewhere?
20:14:39 <alise> now it's mysteriously gained another digit
20:14:47 <alise> s.addstr(22,0,'_'*80);C('$:%-17s T:%-17s H:%-17s HP:%-3s (%s)'%(G,N,H,L,P));s.move(y-Y+11,x-X+40)
20:14:53 -!- coppro has quit (Quit: Lost terminal).
20:15:47 <alise> and no, there is nowhere else I change these values...
20:15:57 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
20:16:08 <alise> that 8xx thing behaves like 1xx
20:16:14 <alise> because 900 made me die
20:16:47 <pikhq> That's really annoying. Really really annoying. All the rips of Monty Python's Flying Circus out there are from the NTSC DVDs.
20:18:00 * alise decides to make hunger actually be satiation
20:18:30 <pikhq> WHY WOULD I WANT NTSC VERSIONS OF A PAL BROADCAST?
20:20:10 <alise> ais523: I swear, this is *utterly* inscrutable to me.
20:20:15 <alise> pikhq: Wanna debug my golfed Roguelike?!?!?!?!?!
20:22:24 <pikhq> But, I can find from-PAL rips of the *movies*.
20:22:34 <pikhq> Y'know, the ones that are 24 fps.
20:22:56 <pikhq> (and that I could de-telecine from the NTSC source)
20:24:00 <pikhq> Why do I have to be pickier than everyone who does encodes for torrents?
20:24:02 <alise> I swear, it's like key repeat does nothing to this.
20:25:07 <pikhq> I may have to purchase the series from amazon.co.uk just to not be irritated.
20:26:07 <ais523> wait... you torrent TV programs, but buy them if the torrents are in the wrong format?
20:26:29 <ais523> I'm trying to figure out a code of laziness/ethics/piracy that would cause that to be your typicla behaviour
20:27:05 <pikhq> ais523: Actually, I torrent them, but then get irritated at the low quality of the torrents, and I am now being irritated.
20:27:06 <alise> ais523: "Piracy is not wrong, and I am a perfectionist."
20:27:15 <pikhq> Also, what alise said.
20:27:15 <alise> This is ... pretty much also my position.
20:27:47 <pikhq> Well, in *this* case, I made the DVD rips from roommate's box set, and am now being irritated that it wasn't in PAL.
20:28:58 <pikhq> Also: seriously, if I had the hard drive space to make it practical, I'd just be storing remuxes of the DVD.
20:30:05 <pikhq> As it is, 1.2 Mbps h264 & source audio works.
20:32:01 <alise> pikhq: Please figure out why Python is ignoring physics.
20:32:14 <cpressey> pikhq: btw, how do you pronounce your nick? because i tried last night and what came out sounded really awful.
20:32:54 <pikhq> cpressey: Peek aitch kyuu
20:32:54 <ais523> I mentally pronounce it as in pik HQ
20:32:55 <cpressey> in case you care, i found your hs bf compiler i had saved to my flash drive, and i said to myself, "oh yeah that's pikhq's"
20:33:03 <olsner> I usually just stop reading after "pik"
20:33:40 <olsner> alise: probably because guido doesn't understand it?
20:33:43 <cpressey> pikhq: oh! is it... supposed to sound similar to "Pikachu"? i'm surprised i never noticed
20:34:01 <pikhq> cpressey: I was 8 and fond of Pokémon.
20:34:17 <alise> I pronounce it "pikhq"
20:34:22 <Vorpal> <alise> Vorpal: want to see the WIP version? <-- sure
20:34:24 <alise> Let me tell you, pronouncing "khq" is a BITCH.
20:34:35 <pikhq> alise: I demand some IPA.
20:35:02 <alise> "pi" as the start of pikachu; ktch-kyu but don't pronounce the u
20:35:18 <olsner> Your GStreamer installation is missing a plug-in. | Your GStreamer installation is missing a plug-in. | Internal data flow error. | Your GStreamer installation is missing a plug-in. | Your GStreamer installation is missing a plug-in. | Internal data flow error.
20:35:23 <alise> Vorpal: As soon as I get quaffing working :P
20:35:26 <alise> Vorpal: No max HP hooray
20:36:29 <olsner> grr, music players on linux worked much better before gstreamer
20:36:36 <pikhq> I also need to go through my anime collection and get rid of all the hardsub'd stuff.
20:36:47 <pikhq> Hardsubs anger me.
20:40:11 -!- Phantom_Hoover has joined.
20:42:33 <alise> WHY DOES THIS NOT FURK
20:43:25 <ais523> alise: pastebin it somewhere, I'm interested now
20:43:43 -!- coppro has joined.
20:43:43 -!- coppro has quit (Changing host).
20:43:43 -!- coppro has joined.
20:44:00 <alise> def C(x):s.insstr(23,0,' '*80);s.addstr(23,0,x);s.redrawwin();s.refresh()
20:44:03 <alise> this inexplicably fixes everything
20:44:05 <alise> (the redraw lines)
20:44:24 <alise> ais523: http://pastie.org/1201439.txt?key=fo9d7wsmz1xh8b6gwnkbvg
20:44:30 <alise> remove ";s.redrawwin();s.refresh()"
20:44:42 <alise> things that fail: get a ! (potion), q(uaff) it, doesn't show until next turn
20:44:43 <ais523> oh, I wasn't planning to run it
20:44:57 <alise> hold down j (only j works, I have no idea why), watch turn and satiation counters not change after a while
20:45:00 <alise> move in another direction
20:45:07 <alise> ais523: it's perfectly innocuous...
20:45:19 <Vorpal> <alise> Vorpal: No max HP hooray <-- heh
20:46:47 <Vorpal> it isn't much of a rougelike without that :(
20:46:56 <ais523> alise: does anything else on your botl update?
20:47:00 <alise> Vorpal: oh eff of, it's going to be like 150 lines *with* the boss
20:47:03 <ais523> I'm wondering if it's an issue with the cursor position
20:47:07 <alise> alise: $ when you get $
20:47:12 <Vorpal> alise, what do you call it?
20:47:42 <Vorpal> alise, ah I need to make a Vagrant'ELM then (Extended Levels and Magic) ;)
20:48:15 <Vorpal> probably won't do it though, not enough motivation
20:49:15 <Vorpal> it seems I more and more prefer thinking about programming than actually programming. Not just the theoretical parts, but also sometimes the implementation details
20:49:23 <alise> ais523: if you actually figure it out, do enlighten me :P
20:49:26 <Vorpal> if only someone invented a serialization interface for the brain
20:49:56 <ais523> so N and S don't change, but G does?
20:50:19 <Vorpal> I mean, I thought about brainfuck optimisation quite a bit recently, and thought of some interesting things, but meh, can't be bothered to code all the analysis needed for it.
20:50:30 <Vorpal> mostly ways to optimise unbalanced loops
20:50:53 <Vorpal> require quite a lot of graph operations to figure out invariants and such
20:52:51 <alise> ais523: N and S change, yes.
20:52:53 <alise> all of them change
20:52:57 <alise> at different times
20:53:02 <alise> N and S change in lockstep except when you eat
20:53:02 <ais523> I mean, when holding down k
20:53:07 <ais523> rather than in general
20:53:12 <alise> ais523: N and S change but not G unless you run into anything.
20:53:18 <alise> any $s, in particular
20:53:24 <alise> P would change if you run into a potion
20:53:26 <alise> but this happens on open space
20:53:28 <ais523> when holding down j and N and S become bugged, does G also change?
20:53:34 <ais523> when you hit a $ at random?
20:53:37 <alise> and somehow, *not redrawing* causes the variables to change state permanently(?!?!?!)
20:53:48 <alise> ais523: i'm not sure, it's never happened to me
20:53:53 <Vorpal> alise, argh not vimkeys
20:54:21 <ais523> Vorpal: I assumed you'd use vikeys for roguelikes...
20:54:27 <cpressey> numpad controls + laptop = argh
20:54:33 <alise> no, he is allergic to vikeys
20:54:35 <ais523> cpressey: that's why I learnt vikeys initially
20:54:38 <alise> and religiously insists on numpad
20:57:28 <cpressey> Python would be just that much less obnoxious if only it had 'isa' and 'has' and 'can'
21:00:34 <pikhq> alise: Imagine the storage space that could be had if we brought back 5¼" hard drives.
21:01:21 <cpressey> pikhq: I touched one once! It was already dead, alas.
21:02:27 -!- augur has joined.
21:04:58 <cpressey> http://en.wikipedia.org/wiki/Draco_%28programming_language%29
21:08:05 <cpressey> I need a language where tokenization happens on case-change boundaries
21:09:08 -!- pineal_aenimal has quit (Ping timeout: 241 seconds).
21:09:12 <fizzie> I think there's a 40 MB 5.25" IDE HD in my closet. Imagine the storage space.
21:09:42 <pikhq> A modern 5¼" drive. Imagine what could be.
21:10:18 <fizzie> If I recall correctly, it's split to a 32 MB and 8 MB FAT partitions, because DOS ~3.2 didn't support such hugeness.
21:13:19 <ais523> alise: when you get back, I can't see what's causing the error, but am confused about scopes; why is there a "global" in M but not D?
21:14:32 <cpressey> wait, that should be: swapDIVprint
21:17:57 <cpressey> i should totally write a utility that just fills my terminal with randomly coloured solid squares.
21:18:02 <cpressey> i would actually find this useful
21:18:36 <cpressey> my poorman's version is ls -la with dir colourization active
21:18:58 -!- cpressey has quit (Read error: Connection reset by peer).
21:19:14 -!- cpressey has joined.
21:19:53 <cpressey> irssi threw "status access violation" or something
21:21:52 -!- GreaseMonkey has joined.
21:21:52 -!- GreaseMonkey has quit (Changing host).
21:21:52 -!- GreaseMonkey has joined.
21:25:13 <ais523> cpressey: why do you want a utility to do that?
21:25:14 <Phantom_Hoover> To beat a dead horse some more, aren't groups and monads the same kind of thing?
21:25:16 <ais523> it shouldn't be too hard...
21:25:57 <cpressey> ais523: To easily see when I've reached the top of the output of the last command I issued when I browse the scrollback.
21:26:23 <cpressey> Running 'rainbow vomit' or such would be much cooler, though.
21:27:33 <cpressey> Phantom_Hoover: Same kind of thing? Sure.
21:28:02 <cpressey> We have this thing, and this other thing, and they do stuff.
21:29:42 <Phantom_Hoover> i.e. a group is a set G, a function . : G×G→G and the conditions of identity, inversion and associativity.
21:29:42 <Vorpal> <ais523> Vorpal: I assumed you'd use vikeys for roguelikes... <-- this is not the first time you told me that
21:29:46 <Vorpal> and I told you I do not
21:30:01 <Vorpal> ais523, it is like the 7th time over the past few years
21:30:48 <Vorpal> <cpressey> Python would be just that much less obnoxious if only it had 'isa' and 'has' and 'can' <--- like... "import foo" → "can has foo"?
21:31:02 <Phantom_Hoover> A monad is a functor m, functions unit and join and the monad laws.
21:31:24 <cpressey> Phantom_Hoover: A group is a tool for studying symmetry mathematically.
21:33:25 <alise> ais523: you only need global to assign
21:33:27 <Phantom_Hoover> Well, I think the category-theoretical definition of monads involves wrapping around data.
21:33:37 <alise> Phantom_Hoover: not really.
21:33:56 <ais523> alise: are you sure? does it not form a closure without?
21:34:15 <alise> ais523: not if you do +=
21:34:15 <cpressey> Vorpal: not exactly what I had in mind
21:34:18 <alise> works if you do =, just forms a closure
21:34:21 <alise> but if you do += it fails
21:34:27 <alise> because it expands to
21:34:29 <Vorpal> cpressey, I suspected as much
21:34:30 <alise> and x isn't definef
21:34:33 <alise> because you're defining it
21:34:38 <alise> so it must be part of the new scope, not globals
21:34:47 <cpressey> all hail the simplicity of unscopedness
21:34:51 <ais523> alise: I mean, if you use a variable inside a definition, don't you read the value it had when the definition was defined?
21:34:52 <Vorpal> cpressey, hovering I just couldnt resist joking about that horrible lolcode
21:35:05 <alise> ais523: let's put it this way
21:35:14 <alise> -------------------
21:35:30 <ais523> what sort of scoping is that?
21:35:43 <alise> because it sees that x is defined somewhere, yet you use it before it's defined!
21:35:47 <alise> ais523: no, that scoping is okay
21:35:51 <alise> the second one doesn't modify the global
21:35:58 <alise> it creates a local
21:36:01 <ais523> the second one looks like some sort of scope-by-reference
21:36:09 <alise> it doesn't mutate global x
21:36:13 <Vorpal> the third one is the real issue
21:36:14 <alise> i already said that
21:36:33 <ais523> I'm asking about x=3; def f(): return x; x=9; print f()
21:36:38 <ais523> which isn't a case you've suggested so far
21:36:49 <ais523> that's what I was bletching at
21:37:28 <ais523> hmm, I suppose it's using lexical scoping there
21:37:43 <alise> you need to show f's grouping
21:37:57 <alise> print f() => prints 9
21:37:59 * Vorpal throws a hail storm at cpressey
21:38:02 <alise> that's what everything does
21:38:24 <Phantom_Hoover> Doesn't Haskell allow you to define monads that don't even obey the monad laws.
21:38:25 <Vorpal> English isn't even consistent about *which* words it writes as one
21:38:25 <ais523> so here, x in f() means "the current value of the variable x that exists in the scope where f was defined"
21:38:41 <ais523> I think that's lexical scoping
21:38:54 <alise> ais523: there, yes
21:38:58 <alise> Phantom_Hoover: yes
21:39:12 <Vorpal> ais523, can you explain this in English: "hailstorm" but "car engine", why not "hail storm" or "carengine"
21:39:31 <ais523> Vorpal: historical accident
21:39:37 <ais523> although "hail storm" is also correct
21:39:50 <Vorpal> ais523, is there any pattern to when words are written together and when they aren't?
21:39:50 <cpressey> although "hail stone" might not be
21:40:05 <ais523> Vorpal: not that I know of
21:40:19 <alise> Vorpal: Swedish isn't totally consistent either so shaddup
21:40:32 <Vorpal> alise, a lot more though, also any specific examples?
21:41:27 <Vorpal> ais523, and you imported the word "gravad lax" from Swedish (much like you did with smörgåsbord), except you turned it into "gravadlax" iirc. Since usually it is English who writes as separate words it doesn't make much sense XD. (Also it is completely logically that it should be two words in Swedish)
21:41:28 <alise> Vorpal: i don't know swedish, but i know for a fact it isn't totally consistent.
21:41:34 <Vorpal> (if anyone cares I could explain why)
21:41:42 <alise> i also know that if anyone used it as much as english, it would be just as inconsistent
21:42:00 <Vorpal> alise, correct, we have other issues, you have easy rules for when to use "a" and when to use "an", we have en/ett and no easy rules for when to use which
21:42:36 <alise> ais523: I simply have no idea how that code could possibly cause the number of turns to decrease, ever.
21:42:54 <Vorpal> alise, I was just looking for a pattern in this specific issue, I did not claim Swedish was consistent in general of course. Such a claim would be absurd. Where did you get that from?
21:43:33 <alise> i'm just saying stop acting like it's strange that english is so abhorrently inconsistent
21:43:38 -!- Mathnerd314 has quit (Ping timeout: 276 seconds).
21:43:52 <Vorpal> I did not. I was just wondering about a specific issue
21:44:04 <cpressey> a language doesn't travel halfway across the globe without picking up a few venereal diseases
21:44:06 <Vorpal> alise,, maybe you should try to base your attacks on something more substantial than thin air next time
21:44:25 <alise> Vorpal: you often complain about english.
21:44:29 <alise> it is really irritating and boring.
21:44:36 <Vorpal> cpressey, my interest in this was simply finding a better way to figure out than checking if aspell accepts the written together form
21:44:45 <pikhq> Vorpal: The pattern is this: English does compound words with spaces between the components. Things that don't do this are exceptions.
21:44:55 <Vorpal> alise, don't generalise. I didn't do it in this case.
21:45:04 <Vorpal> alise, so yeah your attack was based on thin air
21:45:13 <pikhq> And also: always-adding-spaces is also correct.
21:45:25 <Vorpal> pikhq, like "hail stone"?
21:45:26 <alise> pikhq: Correct but inidiomatic.
21:45:31 <pikhq> Perfectly correct.
21:45:33 <alise> And with English, really, only idiomatic matters.
21:45:49 <alise> a hailstone isn't a type of stone
21:45:53 <alise> it's just a hailstone
21:45:57 <alise> whereas a hail storm is a storm of hail
21:46:01 <pikhq> alise: It doesn't get noticed at all as unidiomatic. And in fact I'd write it as "hail stone", likely.
21:46:03 <cpressey> well, i will certainly know what you mean if you say "a hail stone hit me in the forehead"
21:46:16 <Vorpal> pikhq, what about "arrow head" then?
21:46:27 -!- augur has quit (Read error: Connection reset by peer).
21:46:41 <Vorpal> alise, nice, what sort of monsters?
21:46:41 <cpressey> germanic vs latinate steel cage match
21:46:44 <pikhq> Vorpal: I'd write that as "arrowhead".
21:46:54 <Vorpal> cpressey, correction: "fore head" :P
21:46:56 <alise> Monsters that only start moving when you walk into their view: realistic!
21:47:01 -!- augur has joined.
21:47:03 <pikhq> Don't think "arrow head" is wrong, though.
21:47:06 <Vorpal> pikhq, same, but would "arrow head" be correct?
21:47:22 <pikhq> Vorpal: "fore head" screams "wrong" though.
21:47:29 <cpressey> "arrow head" has amusing connotations to me. like it's an unusual part of an arrow.
21:47:44 <pikhq> I'm parsing "fore" as a morpheme but not an individual word.
21:47:51 <Vorpal> pikhq, yeah, which shows that the "spaces is always correct" rule suddenly breaks down :P
21:48:04 <Vorpal> pikhq, you have fore and aft hm
21:48:18 <alise> Hmm, what should monsters look like.
21:48:25 <pikhq> Vorpal: English orthography is hard, mmkay?
21:48:33 <Vorpal> cpressey, that should be "stearing side", iirc that is the history of the term
21:48:42 <alise> Q. Q is a good monster colour.
21:48:44 <Vorpal> comes from scandinavian langauges iirc
21:49:08 <Vorpal> it's "styrbord" in Swedish, which is a lot closer to making sense in the modern form
21:49:24 <Vorpal> like starboard which changed so much that it doesn't make immediate sense any more
21:49:49 <cpressey> well, i hasten to point out that there is no "steering side" on a modern ship either
21:49:59 <alise> http://upload.wikimedia.org/wikipedia/en/8/86/Large_Hailstons_in_Leipzig_Jun06.jpg
21:50:05 <Vorpal> cpressey, "The origin of the term starboard comes from early boating practices. Before ships had rudders on their centerlines, they were steered by use of a specialized steering oar. This oar was held by an oarsman located in the stern (back) of the ship. However, like most of society, there were many more right-handed sailors than left-handed sailors. This meant that the steering oar (which had been broadened to provide better control) used to
21:50:05 <Vorpal> be affixed to the right side of the ship."
21:50:19 -!- Mathnerd314 has joined.
21:50:22 <alise> I don't want to store health for every single monster... Hmm.
21:50:44 <Vorpal> cpressey, not on modern ones, but surely you have seen viking ships on TV and museums and such?
21:50:47 <Vorpal> or maybe not over there
21:50:50 <ais523> alise: make hitting the monster kill it outright with small probability, do nothing otherwise?
21:51:00 <ais523> so the more you pound on any given monster, the more likely it is to die in that time?
21:51:05 <alise> ais523: Stupid interpretation of US law ahoy: "if corporations are people, and slavery is the trade/exchange/purchase/sale of people, and the stock market is the trade/exchange/purchase/sale of corporations. Then in effect, the Stock Market is a slave Market."
21:51:16 <alise> Because the law literally says "CORPORATIONS ARE PEOPLE"
21:51:19 <alise> And this is why corporations get jobs!
21:51:21 <alise> ais523: no, he's *serious*
21:51:26 <alise> ("The Citizens United ruling either needs to be overturned or the Stock Market needs to be eliminated for violating the Constitution...")
21:51:49 <ais523> the difference is that the word "owning" has a different meaning wrt corporations, and wrt natural persons, to some extent
21:51:55 -!- cheater99 has joined.
21:52:01 <ais523> still, if you can control a person's actions via voting at them, isn't that slavery in some respects?
21:52:04 <cpressey> he did put it in terms of 'if then'
21:52:23 <ais523> you know, I think that argument might technically be legally correct with a literal meaning
21:52:29 <Vorpal> <alise> ais523: no, he's *serious* <-- who?
21:52:43 <alise> ais523: yes, but no law says "corporations are people"
21:52:48 <alise> just "corporations have these rights, yada yada yada"
21:53:03 <ais523> I thought there was a law defining corporations as "persons"
21:53:05 <cpressey> the concept is "corporate personhood"
21:53:07 <ais523> (stupid legal plurals...)
21:53:10 <alise> if not randint(0,14):
21:53:10 <alise> Q('Euuch! That must have been poisoned...',1);L-=randint(15,20)
21:53:13 <alise> food is a bitch in this game
21:53:16 <cpressey> that corporations have all the same rights as people
21:53:26 <alise> cpressey: not all, IIRC
21:53:40 <cpressey> alise: the concept. not the reality
21:53:40 <alise> http://www.reddit.com/r/politics/comments/dn3b9/can_we_make_this_happen_redditor_suggests/c11fnbm
21:53:47 <alise> ^ from someone who actually knows it :P
21:53:56 <alise> "While I disagree violently with the ruling, and I think it sets a dangerous precedent in law and allows a destabilizing financial force to enter our political process, the OP's remarks above have no basis in law or reality and would certainly have no power to convince a jury to repeal this verdict."
21:54:34 <alise> ais523: a while back, I had a bit of crisis, in that I oppose corporate personhood but supported agoran partnerships
21:54:37 <alise> ais523: then I realised IT'S A GAME
21:54:54 <ais523> agoran partnerships are clearly a bad idea if you're trying to build a fair democracy
21:55:06 <ais523> good thing that that isn't the goal at Agora, or it would be a very boring game
21:56:49 <ais523> in fact, I think you can oppose corporate personhood and support agoran partnerships for the same reasons
21:56:55 <ais523> Phantom_Hoover: yes, and has been done before now
21:56:59 <ais523> but I doubt it would be all that interesting
21:57:26 <ais523> one of them changed precedences at random, IIRC
21:57:32 <ais523> as in, it was a rule fragment that made the rule defer to other rules
21:58:20 <Vorpal> ais523, how would it spread?
21:58:40 <ais523> I think it wasn't a truly independent virus, but rather was spread by a separate rule
21:59:01 <Vorpal> ais523, how would a truly independent one work?
21:59:12 <ais523> it'd contain the code for replicating itself
21:59:17 <alise> pikhq: what is it with crazy Americans and not wanting to pay income tax?
21:59:23 <ais523> Phantom_Hoover: no, as in any rule can in theory be changed via a 3:1 majority
21:59:51 <Vorpal> ais523, couldn't that be changed to require a 4:1 ?
22:00:10 <ais523> you could change the rules to make them truly immutable, although such a change would be unlikely to pass
22:00:11 <alise> ais523: hmm, even the fountain?
22:00:34 <alise> just use a power=3 rule to kill it
22:00:36 <ais523> alise: the fountain can legally be changed at AI 3; most people would think it very bad form to change it at an AI less than 4, though
22:00:50 <ais523> and many people think it should only be changed via scam
22:00:58 <Vorpal> Phantom_Hoover, Agoran Intelligence?
22:01:00 <ais523> Phantom_Hoover: the higher the AI, the harder it is for a proposal to pass, but the more it can if it does pass
22:01:05 -!- augur has quit (Read error: Connection reset by peer).
22:01:13 <ais523> Phantom_Hoover: and of course not, they change frequently in fact
22:01:19 <Vorpal> ais523, what does AI stand for?
22:01:38 -!- augur has joined.
22:01:40 <Phantom_Hoover> ais523, so can the 3:1 rule be disabled and then a permanent rule introduced?
22:01:42 <ais523> anyway, 3 is enough to change anything, but by convention some things need more (and people vote against otherwise)
22:01:46 <ais523> Phantom_Hoover: of course
22:01:52 <ais523> other rules need to be disabled too to make a permanent rule
22:01:59 <ais523> such as the one that says permanent rules are disallowed
22:02:05 <ais523> and no, because permanent rules are a stupid idea
22:02:17 <ais523> as the whole point of a nomic is to not have permanent rules
22:02:23 <alise> Phantom_Hoover: the 3:1 rule is just the proposal-passing rule.
22:02:24 <ais523> even that's a bad idea
22:02:35 <ais523> why would you even want a permanent viral rule?
22:02:38 <Phantom_Hoover> A single rule which says only "this rule is immutable".
22:02:51 <ais523> even that would be bad form IMO
22:02:58 <ais523> if you want a permanent trophy, make it too cool to repeal
22:03:15 <ais523> as in, people will never vote to repeal it
22:03:17 <Vorpal> ais523, such as the fountain?
22:03:20 -!- impomatic has joined.
22:03:30 <Vorpal> ais523, wasn't there a whale too?
22:03:34 <ais523> rule 104's never been changed, and as a result, most players will refuse to vote for changes to it
22:03:42 <ais523> especially as there's no reason to change or repeal it
22:04:08 <Vorpal> ais523, http://www.agoranomic.org/ <-- wait, did they redesign that page
22:04:10 <ais523> that one's was explicitly intended for people to scam their way into, eventually, but it lasted longer than expected
22:04:14 <Vorpal> the agora nomic website that is
22:04:20 <alise> Vorpal: ais523 rewrote it, now it's unreadable
22:04:28 <alise> (because of the silly two columns)
22:04:29 <ais523> and alise has gone all crazy about inability to read the page
22:04:37 <ais523> alise: if a page is in two columns and one is ads, is it unreadable?
22:04:52 <Vorpal> alise, why would two columns be unreadable
22:04:52 <alise> ais523: you'd fill a whole column with ads?
22:04:56 <Vorpal> wikipedia main page uses that
22:04:58 <ais523> alise: I've seen it done
22:05:01 -!- MigoMipo has quit (Read error: Connection reset by peer).
22:05:08 <alise> also, I try and keep a policy of not talking to anyone who takes disagreement as "going all crazy"
22:05:09 <Vorpal> w3c uses 3 columns iirc
22:05:24 <ais523> but if you can filter out ads from a second column, why not text?
22:05:29 <alise> Vorpal: w3c uses three columns, of which one is content and the other two is navigation
22:05:30 <Phantom_Hoover> Agora would seem an interesting thing to do, but I suspect it'd require a large investment of time and effort...
22:05:43 <Vorpal> alise, anyway why is two columns of text wrong?
22:05:54 <alise> Phantom_Hoover: no, it requires basically 0
22:06:01 <ais523> Phantom_Hoover: not that much, you only need to pay attention once a week or so, maybe even less
22:06:06 <alise> i'm a lazy arse who just mocks wooble every now and then on the mailing list
22:06:08 <alise> and i'm still a player
22:06:30 <ais523> if you want things to /happen/, you basically have to do them yourself, otherwise it just sits there doing nothing but the occasional report
22:06:42 <ais523> but if you're content to watch and chip in occasionally, hardly any effort's required
22:07:14 <Vorpal> ais523, I don't understand rule 104
22:07:18 <Vorpal> ais523, I just read it
22:07:28 <Vorpal> ais523, what is a speaker in agora? It takes too much to find the relevant rules
22:07:31 * impomatic offers ais523 as a sacrifice to summon Egobot.
22:07:32 <ais523> The Speaker for the first game shall be Michael Norrish
22:07:48 <Vorpal> ais523, yes, you said that "<ais523> rule 104's never been changed, and as a result, most players will refuse to vote for changes to it"
22:07:52 <ais523> the Speaker for the first game /was/ Michael Norrish, it doesn't matter what it means now
22:08:08 <ais523> thus, Michael Norrish is obligated to be MIchael Norrish, an obligation that he takes very seriously
22:08:30 <alise> once, Michael Norrish was not Michael Norrish
22:08:34 <alise> we exiled him from the game retroactively
22:08:47 <ais523> alise: is that just a blatant lie?
22:08:52 -!- augur has quit (Read error: Connection reset by peer).
22:08:52 <ais523> or do you know something you aren't telling me?
22:08:58 <alise> ais523: I know ... all things ...
22:09:09 <alise> (it's as much of a blatant lie as him being serious about his responsibility to be himself)
22:09:12 <alise> (which is to say, entirely)
22:09:23 -!- augur has joined.
22:09:29 <ais523> alise: well, it was found via CFJ that he legally had to be Michael Norrish
22:09:38 <ais523> and he's generally been pretty good about keeping to the rules
22:09:51 <alise> ais523: ooh, I have an excellent idea
22:10:02 <alise> ais523: Michael Norrish is breaking the rules; I'll elaborate in /msg
22:10:42 <Vorpal> hm does "suffusion of yellow" come from Dirk Gently or does it have some older source?
22:11:04 <cpressey> impomatic: a sacrifice to Gregor, I assume
22:11:22 <ais523> Vorpal: incidentally, I learnt WML from looking at existing examples
22:11:28 <alise> Vorpal: dirk gently
22:11:46 <Vorpal> alise, Dirk Gently would be older
22:11:59 <Vorpal> ais523, it shouldn't be too hard
22:12:03 <alise> didn't you see how much time travel was in that story?
22:12:17 <ais523> also, what a language!
22:12:29 <ais523> it's like someone decided to make an XML-based language, and was serious about it
22:12:35 <ais523> but mixed it with the C preprocessor
22:12:45 <Vorpal> ais523, but it is a DSL so learning it from examples wouldn't be impossible
22:12:58 <Vorpal> at least in my experience
22:13:02 <ais523> Vorpal: it isn't a DSL, really
22:13:06 <ais523> it's a pretty general language
22:13:33 <Vorpal> ais523, also I thought it was heavy on [ ] for syntax? was that the preprocessor stuff?
22:14:17 <ais523> Vorpal: no, that's the XML
22:14:22 <ais523> it uses square brackets rather than angle brackets
22:14:31 <ais523> the preprocessor uses #ifdef, etc, as in C
22:14:37 <ais523> except that # is /also/ a comment character
22:14:42 <ais523> and also, it uses braces
22:15:30 <ais523> cpressey: very pluggable
22:15:41 <ais523> you can do {path/to/directory}
22:15:48 <ais523> and it automatically #includes every file in that directory
22:16:01 <alise> ais523: is WML the WAP one?
22:16:02 <Vorpal> ais523, not #include or anything such?
22:16:10 <impomatic> cpressey: a sacrifice to anyone with the power to summon egobot. I want to test a BF Joust entry before I add it to the wiki :-)
22:16:12 <ais523> alise: no, Battle for Wesnoth
22:16:14 <Vorpal> alise, no? it is wesnoth
22:16:28 <ais523> Vorpal: it's not called #include, it's called {}, which is the syntax also used for something entirely different
22:16:41 <ais523> apparently, the game disambiguates by checking to see if the file in question exists or not
22:16:44 <Vorpal> ais523, I guess it is not LR(1) or?
22:16:56 <Vorpal> indeed crazy parsing at least
22:16:56 <ais523> (actually, hopefully it checks the list of definitions first)
22:17:03 <cpressey> ais523: the best kind of context dependency ever!
22:17:06 <ais523> Vorpal: no, it's basically XML, it parses incredibly regularly
22:17:20 <Vorpal> ais523, huh, but not preprocessing?
22:17:20 <ais523> in fact, the only way I found to make it error at all is to include a mismatched bracket or something like that
22:17:29 <ais523> anything else just fails silently
22:17:37 <Vorpal> ais523, that's rather nasty
22:17:58 <ais523> the lang isn't aware that it's trapped inside Wesnoth
22:17:58 <Vorpal> ais523, also why is the preprocessor used? Not for control flow I presume?
22:18:03 <ais523> Vorpal: for subroutines
22:18:26 <Vorpal> alise, not large enough toaster
22:18:26 <ais523> there's a way around it, but it's really complex, involving setting up events to call each other in the future
22:18:29 <Vorpal> alise, otherwise: sure
22:18:31 <ais523> so the subroutines are more efficient
22:18:41 <alise> Vorpal would eat toasted bear
22:18:49 <ais523> I suspect that the loading bar when you start playing is mostly inlining subroutines
22:18:56 <Vorpal> I didn't say I would eat the result
22:19:15 <Vorpal> alise, please read what it says, I would test it on you first
22:19:19 <Vorpal> to see if it was eatable
22:19:50 <ais523> besides, who eats a /toaster/
22:19:51 <Vorpal> ais523, edible as well
22:20:05 <alise> eatable and edible, what a requirement
22:20:14 -!- impomatic has left (?).
22:20:17 <Vorpal> ais523, eatable I define as "physically possible to eat, like you can get it into your mouth and so on"
22:20:30 <Vorpal> edible I define as the usual meaning
22:20:30 <alise> Vorpal: your definition does not agree with the english language
22:20:33 <alise> cheater99: because it isn't chrome
22:20:38 <alise> Anything edible; That can be eaten without harm; non-toxic to humans; suitable for consumption; That can be eaten without disgust
22:20:39 <alise> en.wiktionary.org/wiki/edible
22:20:41 <Vorpal> alise, of course not, since when did it ever do that?
22:20:41 <olsner> cheater: due to excessive suckage
22:20:49 <alise> Vorpal: "edible" does not mean "nice"
22:20:51 <ais523> chrome seems like an appropriate material to plate a toaster with...
22:20:56 <alise> "edible" means "literally able to be eaten"
22:21:07 <Vorpal> alise, http://en.wiktionary.org/wiki/eatable
22:21:08 <alise> so what you mean by "eatable" is actually "edible"
22:21:13 <cheater99> <cheater99> why is it so hard to find an extension for firefox
22:21:13 <cheater99> <cheater99> that makes the address bar search things, like in chrome?
22:21:13 <cheater99> * cheater99 says that with a french accent
22:21:15 <Vorpal> alis<alise> Vorpal: "edible" does not mean "nice" <-- correct
22:21:27 <Vorpal> alise, but see my link
22:21:35 <alise> edible (not comparable)
22:21:35 <alise> That can be eaten without harm; non-toxic to humans; suitable for consumption.
22:21:35 <alise> eatable (comparative more eatable, superlative most eatable)
22:21:35 <alise> Able to be eaten; edible
22:21:39 <alise> they mean the same thing.
22:21:39 <cheater99> alise: what if you say a girl is edible
22:21:46 <alise> except that eatable makes no sense.
22:21:55 <Vorpal> alise, it exists as a word
22:21:56 <alise> edible also means "That can be eaten without disgust." but less so
22:22:05 <alise> cheater99: "Yeah, she's totally non-toxic to humans".
22:22:28 <cheater99> alise: well "inedible" can still be "digestible"
22:22:37 <cheater99> like, there are some types of mushrooms
22:22:40 <alise> a guy digested a plane
22:22:43 <alise> doesn't mean the plane is edible :P
22:22:48 <cheater99> but if you force yourself to eat them, you won't die
22:23:26 <cheater99> but in a pinch, they provide protein
22:23:44 -!- sshc has quit (Read error: Connection reset by peer).
22:24:32 <cheater99> alise have you ever picked mushrooms
22:24:42 <cheater99> are you a person who enjoys trip to the woods
22:25:54 -!- sshc has joined.
22:26:10 -!- augur has quit (Ping timeout: 252 seconds).
22:26:44 <Vorpal> cheater99, here is what I dislike about doing that:
22:27:05 <Vorpal> Exhibit A: Mushroom tasting nice.
22:27:09 <cheater99> you suddenly start hallucinating and find yourself in devonshire?
22:27:16 <Vorpal> Exhibit B: Looks exactly like A but lethal
22:27:24 <cheater99> with clothes on you've never seen before?
22:27:29 <Vorpal> Exhibit C: Looks exactly like A but causes you to hallucinate
22:27:40 <Vorpal> Exhibit D: Looks exactly like A but just tastes boring
22:27:45 <cheater99> inedible mushrooms look completely different
22:27:52 <cheater99> i'm not sure what you're talking about
22:27:57 <alise> i see no problem with exhibit c :)
22:28:21 * cheater99 suddenly thinks alise is a miscreant
22:28:34 <Vorpal> alise, well, depends on if you serve it at the annual anti-drug society meeting :P
22:29:02 <Vorpal> alise, yes B is the issue
22:29:20 <Vorpal> cheater99, anyway what about "kantarell" whatever that is called in English I don't know
22:29:31 <Vorpal> iirc there are some sorts that look extremely similar to a lethal one
22:29:51 <cheater99> and then tell me what they're called in english.
22:30:38 <alise> takes two seconds with interwiki, maybe Vorpal could take the effort.
22:30:41 <cheater99> unless you mean podgrzybek szatański, whatever that is called in english i don't know.
22:31:01 <Vorpal> I found out it was not the one I was thinking of
22:31:13 <cheater99> vorpal, i think you need to use the wyszukiwarka tekstu.
22:31:26 <Vorpal> cheater99, I'm looking for the one I meant
22:31:46 <Vorpal> which was not kantarell (chanterelle)
22:32:10 <cheater99> should i go eat some french cheese?
22:32:48 <cheater99> Vorpal: i bet you're all for cheese
22:33:27 <Vorpal> yes, well not all sorts
22:33:32 <Vorpal> I can't stand goat cheese
22:33:55 <Vorpal> ah yes it was chantarell
22:34:04 <Vorpal> cheater99, this is the very similar one: http://en.wikipedia.org/wiki/Hygrophoropsis_aurantiaca
22:35:41 -!- Phantom_Hoover has quit (Remote host closed the connection).
22:35:47 <Vorpal> cheater99, anyway, there you are, night now →
22:36:05 <Ilari> There are pairs of very good and very toxic mushrooms that look very near the same...
22:36:51 <alise> Vorpal: AI is hard, maybe I'll just make them walk randomly.
22:37:37 <cpressey> alise: make them kill their own kind
22:38:31 <cpressey> alise: and add mushrooms of hallucination
22:38:47 <alise> hallu would be hard with my design
22:38:48 <alise> although maybe not
22:39:01 <cheater99> Ilari: they look nearly the same, except their biotopes don't intersect.
22:39:08 <cheater99> and if they do, every guide book has a big warning about it
22:39:30 <cheater99> and lists the give away characteristics
22:39:44 <Ilari> And then there are at least one pair of mushrooms that look nearly the same, but one is very good and one tastes extremely bad.
22:41:25 -!- tombom has quit (Quit: Leaving).
22:42:12 <alise> haha wow hallu is amazing
22:45:57 -!- Quadrescence has quit (Ping timeout: 245 seconds).
22:46:59 <alise> Even you can change character.
22:47:49 <cheater99> This mushroom is commonly confused with the Chanterelle; the distinguishing factors are color (true Chanterelle is uniform egg-yellow, while the false one is more orange in hue and graded, with darker center) and attachment of gills to the stem (true Chanterelle does not have true, blade-like gills--rather, has rib-like folds running down the stem).
22:49:00 <alise> cheater99: hallu in vagrant, my silly roguelike
22:49:27 <cheater99> alise: oh, i thought you meant a hallucination
22:49:35 <alise> 65 lines of python and no architecture that'll let enemies behave non-stupidly!
22:49:37 <cheater99> alise: i thought you have realized your plan of taking magical mushrooms
22:49:44 <alise> i can make them chase you no matter what
22:49:52 <alise> cheater99: lol, that'd be some speed
22:50:32 <Vorpal> cheater99, another one: http://en.wikipedia.org/wiki/Blusher
22:51:31 <alise> hmm, maybe killing a dude should give you like $1,000
22:52:26 <Sgeo> alise, your roguelike needs to be easily FooTV-like-system-able
22:52:34 <Sgeo> It's probably one of my favorite things about Crawl
22:52:40 <alise> Sgeo: there is only one enemy and it doesn't even move.
22:53:01 <Sgeo> Probably lots of people
22:53:12 <cheater99> how could you ever think it's edible?
22:53:24 <cheater99> edible mushrooms are yellow-brown.
22:53:26 <Vorpal> "Although edible, it can be confused with deadly poisonous species, and should definitely be avoided by novice mushroomers."
22:53:47 <cpressey> and novice bear toasters alike
22:54:02 -!- augur has joined.
22:54:04 <alise> ehird@dinky:~/Code/vagrant$ wc -c vagrant.py
22:54:06 <alise> pretty good charcount for this
22:54:10 <Vorpal> cheater99, still my point stands: they all look alike unless you are an expert
22:54:27 <cheater99> why would you ever want to do that
22:54:41 <Vorpal> eat mushrooms? good question
22:54:45 <cheater99> search for mushrooms which look like shitty ones
22:55:21 <Vorpal> cheater99, irrelevant for my original claim
22:55:34 <cheater99> Vorpal: your original claim was that it was a problem
22:55:52 <cheater99> to someone who does dumb things, yes
22:56:02 <Vorpal> the biotope thing is not enough to help me at least. I'm not a nature person
22:56:09 <Vorpal> I couldn't tell what biotope it was
22:56:19 <Vorpal> thus my original claim stands
22:57:05 <Vorpal> cheater99, also /<cheater99> u/s/u/you/;s/loze/lose/
22:57:56 <Vorpal> cheater99, yes it was completely correct sed
22:58:18 <Vorpal> cheater99, I had not turned off monitor yet
22:58:50 <Vorpal> cheater99, also please learn to type. people writing "u" instead of "you" and so on is *really* annoying
22:58:52 <cheater99> also, you don't say "correct sed", you say "correctly sed".
22:58:59 <Vorpal> it makes them look like idiots
22:59:04 <alise> Vorpal: well ull hav 2 get used 2 it
22:59:13 <Vorpal> alise, And indeed, so you said
22:59:34 <Vorpal> <cheater99> also, you don't say "correct sed", you say "correctly sed". <-- no, I used sed as a verb
22:59:52 <Vorpal> you can use it both ways
22:59:55 <alise> punk-tutuashion is unk00wl
23:00:11 -!- Quadrescence has joined.
23:00:15 <Vorpal> alise, you know, this is just filtered out :P
23:00:33 <Vorpal> and now I say good night, unlike last time I used /away as well
23:00:38 <Vorpal> that makes a difference →
23:00:51 <alise> cheater99: you should debug my code
23:01:19 <alise> cheater99: http://pastie.org/1201769.txt?key=xtvbpktbvv5gbzotg5bphw
23:01:42 <alise> the monster movement code (the nested for loop in M)
23:01:44 <alise> crashes it somehow
23:01:49 <alise> unfortunately the endwin means you never see the exception :D
23:01:56 <alise> something's wrong with it, anyway
23:02:04 -!- oerjan has joined.
23:02:09 <alise> good luck figuring out what all the variables do, i pretty much just pick a letter of the alphabet
23:02:15 <alise> w[y,x]=46;x,y=a,b;w[y,x]=64
23:02:20 <alise> w[y,x]=32;x,y=a,b;w[y,x]=64
23:02:26 <alise> (that doesn't fix it)
23:04:09 <alise> cheater99: "made to be as short as possible"
23:04:11 <alise> ugliness be damned
23:04:17 <alise> after code golf, the sport
23:04:24 <alise> http://golf.shinh.org/ is the prime hub for that malarkey
23:05:00 <alise> oh joy, it crashes no longer
23:05:02 <alise> but the guys don't MOVE!
23:06:17 <alise> cool i can walk into walls now, why.
23:06:42 <alise> cheater99: okay, remove that first for loop in M and write something that makes all the Qs on the visible board move closer to the guy.
23:06:57 <oerjan> <Ilari> There are pairs of very good and very toxic mushrooms that look very near the same...
23:07:00 <alise> (you may want to play it a few times to figure out what all the variables are)
23:07:15 <cheater99> that's great, my problem is that my pc is going nuts
23:07:23 <cheater99> i think the gfx card is overheating
23:07:47 <oerjan> it also depends on geography, i hear some asian immigrants to norway get poisoned because one of our poisonous mushrooms look like an edible east asian one
23:07:57 <oerjan> s/hear/read in the newspaper/
23:07:59 <alise> cheater99: thankfully, vagrant runs even without much of a graphics card!
23:08:06 <alise> (can it do text? Yes? YOU WIN!)
23:08:35 <alise> cheater99: http://pastie.org/1201784.txt?key=gvzfssnyfqjigfcfm92luq
23:08:36 <cpressey> alise: so your game is about eating
23:09:09 <alise> cheater99: x,y is your position, X,Y is position of centre cell (infinite plane so we scroll), G is gold, N is # of turns, P is health that can be restored by potions, U is whether we're hallucinating or not
23:09:14 <alise> S is satiation, L is HP
23:09:20 <alise> w is the gamefield indexed by pairs of y and x
23:09:34 <cpressey> square root of minus gamefield
23:09:57 <alise> cheater99: charcodes
23:10:09 <oerjan> (incidentally chanterelles are among the "safe" mushrooms in norway)
23:10:11 <alise> the cells at that position
23:10:20 -!- ais523 has quit (Remote host closed the connection).
23:10:30 <cheater99> i mean, what do they represent in game?
23:10:47 <alise> # is inexplicable cube of unbreakable glass that you can't walk through
23:11:00 <alise> to find the numbers
23:11:02 <alise> that's what i do :P
23:11:04 <alise> 81 is Q, I know that
23:11:09 <alise> oh, and space is just open space
23:11:46 <alise> cheater99: controls: vi keys to move, q to restore health from potions you've picked up (amount you can in parens after HP display), space waits
23:12:10 <alise> cheater99: and basically all the Qs have to do is move one step closer to the @
23:12:23 <alise> you could do it more fancy if you're that masochistic, but if you want to play with the monster code that's what i'd do
23:12:28 <alise> ofc they can only walk onto space
23:12:34 <alise> since there's no memory of what's underneath a monster
23:12:38 <alise> so they'd inexplicably suck up gold
23:12:40 <alise> (maybe that's a feature)
23:13:39 <cheater99> maybe numbers > 1000 should be amount of gold + 1000.
23:13:59 <cheater99> so you have a countable amount of monsters
23:14:10 <cheater99> and it's a parametrized type!!!!!!
23:14:20 <alise> cheater99: i lost you after "yes"
23:15:03 <cheater99> w[x,y] == A > 1000 -> there is a monster at x,y with A-1000 gold.
23:15:44 <quintopia> anyone know if there is a polynomial time or approximately polynomial time algorithm for finding an embedding of a graph in a plane that minimizes the number of edge intersections?
23:16:16 <alise> cheater99: oh i see
23:16:19 <alise> i was thinking rather
23:16:24 <alise> kill monsters before they suck up all the gold ;P
23:16:43 <alise> cheater99: pretty sure the condition to go to the boss will just be having a certain amount of gold and getting to a booth
23:16:58 <alise> cheater99: still, i think figuring out why the buggers won't move is a good first step
23:17:18 <cpressey> quintopia: sounds NP-complete-ish on first blush, but maybe not
23:18:28 <alise> cheater99: meanwhile I'll get rid of the M function, since we don't need it! hooray for obfuscation!
23:18:39 <quintopia> cpressey: it doesn't seem so obvious to me, but even if it is NP-complete in some parameters it may be polynomial in others. . .
23:19:05 <quintopia> alise: so this is what python looks like...
23:19:22 <oerjan> quintopia: i think it's polynomial if the no. of edge intersections is bounded, no idea otherwise
23:19:36 <cheater99> is w in relation to the center, or to the player?
23:19:59 <alise> cheater99: it's global
23:20:04 <oerjan> (because if there are < k intersections you can just try removing all k-edge sets in turn until you find one that makes the rest planar)
23:20:07 <alise> (0,0) is the centre of everything, regardless of where you are, it's where you started
23:20:15 <alise> cheater99: if you go far enough you'll get to (3945783489579435,3459873459843759435)
23:20:57 <oerjan> but obviously this algorithm is exponential in k
23:21:17 <quintopia> oerjan: so, polynomial in number of vertices and numer of edge intersection?
23:21:17 <quintopia> oerjan: can you point me to an algorithm?
23:21:30 <quintopia> that was about a 30 second lag in text entry
23:21:44 <quintopia> aka, i typed all that before you gave me an alg
23:22:34 <oerjan> oh hm wait there's a subtlety here
23:22:57 <oerjan> adding an edge it may have to intersect _several_ others
23:23:22 <quintopia> sure, but if you start at 1 and work your way up to k, you'll find that case first...
23:24:22 <oerjan> ok if you have < k intersections you of course must have < k edges forcing them, so yeah
23:25:59 <quintopia> which brings an interesting question: what's the maximum number of intersections an n node graph can contain?
23:26:21 <quintopia> (assuming that said graph is drawn in a minimal intersection way)
23:26:38 <quintopia> aka, what's the fewest number of intersections the complete graph can contain
23:27:41 -!- Mathnerd314 has quit (Disconnected by services).
23:28:02 -!- Mathnerd314_ has joined.
23:28:09 <oerjan> quintopia: oh wait there's another subtlety - removing an edge, then an arbitrary planar embedding of the rest may not be what gives minimal intersections when you add the edge back in
23:28:12 <quintopia> meh, seems like it'll end up being O(n) anyway
23:28:17 -!- GreaseMonkey has quit (Remote host closed the connection).
23:28:17 <oerjan> that may be more serious
23:28:34 -!- Mathnerd314_ has changed nick to Mathnerd314.
23:29:28 <quintopia> indeed, it may be that no planar embedding of the rest will yield a minimizer
23:29:48 <quintopia> and in fact one needs to find a planar embedding when two edges are removed or something
23:29:51 <cpressey> "if it is NP-complete in some parameters it may be polynomial in others"
23:30:04 <oerjan> quintopia: oh right even that
23:31:05 <oerjan> cpressey: basically that the restricted problem of looking _just_ at < k edge intersecting graphs may be polynomial to decide for a fixed k even if it is exponential in k
23:31:12 -!- zeotrope has joined.
23:31:30 <quintopia> what oerjan said faster than i could
23:31:53 <quintopia> you can bound some parameter and what remains is polynomially solvable
23:31:56 <cpressey> oh, parameter being # of intersecting edges. ok
23:32:50 <oerjan> cpressey: for example i recall that the problem of deciding whether one graph is isomorphic to a subgraph of another is NP-complete, but if the smaller graph is _fixed_ (or bounded in size) then it's a polynomial problem
23:34:03 <quintopia> once you know there is a polynomial time algorithm for unifying graphs on the same number of nodes/edges
23:34:03 <cheater99> alise: just so you know, you've been doing it in the wrong place.
23:34:20 <alise> cheater99: oh. goody.
23:34:25 <alise> cheater99: have you fixed it? :P
23:34:44 <quintopia> alise: the bedroom is the wrong place. next time try it in the library.
23:35:26 <quintopia> (look out for Col. Mustard's lead pipe. you never know where he'll stick it.)
23:35:31 <alise> cheater99: both :P
23:35:38 <oerjan> quintopia: um on a _fixed_ number of nodes/edges. unifying with the same but arbitrary is the graph isomorphism problem which is not known to be P (and is one of the most famous problems in NP to be neither known P or NP-complete)
23:35:38 <alise> cheater99: zeroes are uninitialised things
23:35:43 <alise> that will get initialised when they scroll into view
23:35:48 <alise> so no problem with assigning to them
23:35:51 <alise> spaces are open space
23:36:01 <alise> a zero could be anything, you just don't know until you move so that it's in view
23:36:05 <alise> obviously we can't calculate infinite cells
23:36:14 <quintopia> oerjan: yeah yeah yeah, i know what i meant
23:36:46 <Sgeo> alise, about convincing my dad: When I suggested that this was my normal weight, he said something along the lines of "Maybe. Some people are like that" or something to that effect
23:37:46 <quintopia> in particular if the smaller fixed graph has k nodes, there is a trivial n^k*k! alg. for subgraph isomorphism
23:39:02 <quintopia> so anyway, we still don't even know if there's a solution for the original problem with edge intersections bounded
23:41:35 <cheater99> alise: i made them go to the center. :p
23:41:37 <quintopia> okay apparently this is a really really difficult question
23:43:04 <alise> cheater99: yeah you might want to check that it ==32 or ==0
23:43:13 <alise> cheater99: also if you can do that, just s/0,0/x,y/ (or maybe y,x)
23:43:29 <alise> cheater99: or if you want it to eat items !=whatever '#' is :P
23:43:34 <oerjan> quintopia: i'd imagine graph drawing software would find such an algorithm useful if it existed
23:44:01 <oerjan> (maybe that was how you thought of it?)
23:45:24 <alise> if c==36:G+=randint(5,50)
23:45:40 <alise> interestingly this is not determined until you hit it. totally quantum maan
23:45:59 -!- augur has quit (Remote host closed the connection).
23:46:00 <cheater99> cpressey: the quantum economy!!!!!!
23:46:47 <quintopia> oerjan: i thought of it in the context of the wire-crossing problem. what if the strong hypothesis turns out to be true for some definition of "state"? then, the next obvious question would be "what's the minimum number of crossings required?"
23:47:48 <quintopia> and so i immediately asked "well, how do we even check the number of crossings required for simple machines we already have state diagrams for?"
23:48:52 <quintopia> and I thought it must be a very hard problem, since there is a game about it (gPlanarity)
23:49:18 <oerjan> my intuition on the wire-crossing problem is sort of based on the fact that the 3-coloring problem is NP-complete even for planar graphs.
23:49:54 <quintopia> you intuit that the strong hypothesis holds?
23:50:12 <oerjan> which i think means you very easily can get full computational power without wire-crossing
23:51:37 <alise> cheater99: I will give you $947595486749567945698456 in exchange for your code.
23:52:43 <alise> cheater99: I hope you are omitting spaces!
23:52:51 <oerjan> cheater99: hint, that's zimbabwean dollars
23:53:03 <pikhq> And another set of hardsubs bites the dust.
23:53:22 <alise> headache vomit bleargh
23:53:36 <quintopia> the other reason i thought of it is exactly what you think: In fizzie's grasp language, programs are graphs, and they might be nonplanar. ais523 wants underlambda to be able to compile to any language, and so it would benefit from the ability to have output grasp programs have as few edge crossings as possible.
23:54:26 <oerjan> quintopia: found something maybe relevant
23:54:32 <oerjan> "More generally, for any fixed constant k, it is possible to recognize in linear time the k-apex graphs, the graphs in which the removal of some carefully-chosen set of at most k vertices leads to a planar graph.[6] If k is variable, however, the problem is NP-complete."
23:54:40 <oerjan> http://en.wikipedia.org/wiki/Apex_graph
23:54:57 <quintopia> oerjan: i'd take that much in ZWD, if it still existed. that's actually an appreciable amount of money despite hyperinflation.
23:55:26 <oerjan> yeah but you'd better use it fast
23:56:14 <quintopia> i actually bid on a 20trillion ZWD on ebay once
23:56:33 <quintopia> i hear they aren't very high quality bills. just paper.
23:57:04 <oerjan> quintopia: i imagine the second sentence strongly suggests that your problem is NP-complete for variable k too
23:57:11 <quintopia> anyway, this apex graph thing implies that the general problem is probably NP-complete, but it still doesn't answer the fixed k question
23:57:21 * pikhq hates hardcoded subs.
23:57:55 <alise> can you gimme your code before-ai too just in case it develops sentience
23:58:01 <alise> i don't want sentience
23:58:06 <alise> cheater99: also there are no attacks yet...
23:58:31 <oerjan> yes but is it _sapient_
23:59:13 <quintopia> but: is sapience possible without sentience?
23:59:25 <oerjan> i mean it's all good and well with an AI complaining of pain in its diodes, but that's not the same as intelligence
23:59:33 <alise> quintopia: restate that without using the words sapience and sentience