00:00:19 <alise> It's a purposefully ridiculous theory
00:01:15 <alise> Is there something that emulates rule 110 in Life?
00:01:30 <alise> i.e., produces still-life pixels and spaces, going downwards, of rule 110's evolution
00:01:40 <alise> according to some initial pattern, preferably settable
00:05:35 -!- Gracenotes has quit (Ping timeout: 264 seconds).
00:06:49 -!- augur has joined.
00:08:02 <alise> Quadrescence: by hand?
00:09:05 <Quadrescence> If I sew them yes, if I glue them I do it mostly by hand with the help of a roller for the glue
00:09:51 <oerjan> bound with only the highest quality gold from free-ranging goldfinches
00:13:50 <alise> Quadrescence: Any pictures?
00:13:58 <alise> Also, what method of printing?
00:21:59 -!- maedhros777 has joined.
00:22:01 -!- maedhros777 has left (?).
00:25:56 <uorygl> He prints them by hand, too.
00:26:29 -!- myndzi\ has joined.
00:26:33 -!- myndzi has quit (Ping timeout: 268 seconds).
00:26:37 <uorygl> He has this bottle of ink, and he uses his body heat to heat it up so hot that a little bubble forms, displacing the ink and spewing it onto the paper.
00:26:40 <alise> I hope these don't doth require sound.
00:26:50 <alise> WHAT HAS YOUR CAT GOT TO DO WITH BOOKS
00:27:11 -!- BeholdMyGlory has quit (Remote host closed the connection).
00:27:17 <alise> Nice, though. Very nice.
00:27:55 <alise> In fact, I think I hereby grant thee a Certificate of Awesome.
00:27:55 -!- myndzi\ has changed nick to myndzi.
00:28:08 <alise> Now... give me a damn book
00:28:11 <uorygl> So, this WireWorld AND NOT gate. It's impossible to destroy it by means of incoming signals, but nevertheless, it does not work as expected in all cases.
00:28:28 <Quadrescence> alise: I sell the paperback ones for relatively cheap
00:28:56 <alise> give me an infinite number of paperbacks, then I'll consider it
00:29:02 <alise> (I could sell them and use the profits to buy a hardback)
00:29:26 <Quadrescence> Paperbacks are nice, I like them especially for smaller books or books I don't intend to read 1000 times or use as a reference constantly
00:30:23 <alise> i would pay good money for a nice hardback Hitchhiker's Guide to the Galaxy
00:30:29 <alise> I haven't found a nicely printed one
00:30:40 <alise> money that is good
00:30:45 <alise> Might be best split up into one volume per book
00:30:49 <alise> the omnibus edition I have is rather unwieldy
00:30:57 <uorygl> Let's see, the delay between inputs must be at least 5. Otherwise, if it gets an incoming signal on the negated side, it will destroy the incoming signal on the non-negated side.
00:31:20 <alise> Quadrescence: how nice is the typography, I can't tell from the jewtubes
00:31:43 <alise> that isn't typography.
00:31:58 <Quadrescence> Oh, well I have no control over the typography unless I type it
00:32:18 <alise> what sources do you print from then? I'm sort of guessing this whole enterprise is not entirely legal :-)
00:32:20 <Quadrescence> And if I do have control over the typography it's perfect
00:32:45 <Quadrescence> alise: I "assume" that the books I print for people they have the "rights" to
00:33:06 <alise> Right... but what sources would you, hypothetically, use if printing entirely separately from this operation, and laws didn't exist?
00:33:32 <Sgeo_2> "Your answer to the Queue question on the BCS370 Test was terrific! "
00:33:59 <Quadrescence> Hypothetically if I were to print books in an illegal fashion I would obtain ebooks from whatever site or server I could and in the last case I'd scan it myself and in the last last case I'd type it myself.
00:34:58 <alise> You should totally type out the entirety of the H2G2 N-ogy.
00:35:03 <alise> I assume you have invincible hands.
00:35:10 <Quadrescence> ebooks meaning scanned books or sometimes raw text docs, in which case I'd format it myself in latex
00:35:30 <alise> XeTeX or plain LaTeX?
00:35:38 <alise> The former has more typography and font support.
00:35:45 <Sgeo_2> alise, they're online >.>
00:35:50 <alise> Sgeo_2: I own them.
00:35:58 <Sgeo_2> I'm only going to download the ones that I own physically though
00:36:00 <Quadrescence> it supports dumb fonts like TRUE TYPE and whatever
00:36:08 <alise> Quadrescence: Well, there aren't many good text fonts available for LaTeX.
00:36:15 <alise> And XeTeX /does/ undeniably have more microtypography support.
00:36:30 <Quadrescence> No, Latex has wonderful microtypography support
00:36:50 <alise> But XeTeX has better.
00:37:14 <alise> Anyway, http://nitens.org/taraborelli/latex shut up and print a book in Garamond, now.
00:37:22 <alise> Quadrescence: Considering it was one of its design aims...
00:37:44 -!- Gracenotes has joined.
00:38:03 <alise> I expect the first hardback, LaTeX-set volume of H2G2 by tomorrow.
00:38:20 <Sgeo_2> Hm, who makes money from DNA books?
00:38:21 <alise> Or, if possible, yesterday.
00:38:29 <alise> Sgeo_2: the publishers and presumably whoever inherited the rights
00:38:31 <alise> but mainly the publishers
00:38:36 <alise> probably his widow
00:38:39 * Sgeo_2 doesn't care too much about the publishers
00:38:56 <Quadrescence> alise: http://www.tex.ac.uk/tex-archive/macros/latex/contrib/microtype/microtype.pdf
00:39:15 * alise tries to figure out if he /has/ a widow
00:39:45 <alise> Quadrescence: I agree that LaTeXZ has superb microtypography support.
00:39:51 <alise> It's just that XeTeX extended such.
00:40:16 <Quadrescence> The only advantage I see of XeTeX is its ability to support arbitrary fonts essentially
00:40:27 <alise> Well, and more convenient Unicode.
00:41:21 <alise> You should bind some books in olde-style cases.
00:42:05 <alise> I hate dust jackets
00:42:28 <Quadrescence> I could do leather covers, book cases, headbands, bookmark ribbons, bla bloo bla bla
00:42:43 <alise> All for the low low price of $5,000,000
00:43:36 <Quadrescence> The regular old paperbacks at around 8x5" at <400 pages, I charge $8 + 4 for shipping :O
00:44:00 <alise> how much do you charge for hardbacks?
00:44:59 <Quadrescence> Well first let me clarify that hardback also implies a sewn binding. Anyway I charge more so around $25 though I contemplate charging more only because it takes so much time and I haven't been charging for that
00:45:34 <alise> What about a hardback that you re-typeset :P
00:45:38 -!- oerjan has quit (Quit: Good night).
00:45:47 <Quadrescence> I use expensive book cloth and acid free glues and 100% linen threads and real beeswax and stuff
00:46:00 <Quadrescence> That I retypeset, depends really, I'd probably talk about it in PM and whatever
00:46:30 <alise> "Now! What about FIVE of those!"
00:46:35 <alise> I'll get the piggy-bank...
00:47:42 <alise> Hmm... 180 pages, 208 pages, 160 pages, 192-224 pages (UK-US), 229-240 pages (UK-US). But that's paperback, and so useless.
00:48:19 <alise> So clearly I need to be using my endless bank account. I'll go get Hilbert
00:49:00 <Quadrescence> My original goal was to provide books cheaply for people, so cheap that only materials costs are covered, and nothing else (so I don't get any profit)
00:49:43 <alise> I think if you re-typeset the entirety of a book series then printed it in hardback by hand you deserve some bloody profit.
00:49:55 <alise> But I'm crazy like that :P
00:50:10 -!- FireFly has quit (Quit: null).
00:51:16 <Quadrescence> Well ideally I'd want profit (though the profit would go to better supplies since shit is so expensive)
00:51:41 <alise> You're like the Folio Society... except you're only one person and don't require buying like 50 books just to get one
00:51:51 <alise> So, not really much like that at all
00:52:26 <alise> I know! You should... print interactive fiction. In hardback.
00:52:31 <alise> Clearly this is an excellent, feasible idea.
00:53:59 <alise> http://www.abebooks.co.uk/servlet/BookDetailsPL?bi=1415009849&searchurl=an%3Ddouglas%2Badams%26bi%3D0%26bx%3Doff%26ds%3D30%26sortby%3D1%26x%3D58%26y%3D14
00:54:12 <alise> Quadrescence: LOOKS LIKE YOU'VE BEEN OUTBID BY 1252
00:55:49 <alise> actually I kind of have an urge to try and typeset h2g2 now :/
00:58:13 <alise> I'm not so good at typesetting prose with LaTeX though.
00:58:20 <alise> I haven't really done it before.
00:59:09 <alise> Heck no; Adams (the second Macintosh owner in the UK) would roll in his grave... as would typography.
00:59:17 <alise> Plus, you know, I find Word much less usable than LaTeX.
01:00:44 <pikhq> alise: Fortunately for you, TeX is very, very capable of typesetting prose. :)
01:00:54 <alise> The author of groff is awesome, just look at http://jclark.com/bio.htm
01:01:04 <alise> "I was educated at Charterhouse. I read Mathematics and Philosophy at Merton College, Oxford, where I obtained Class I Honours." + that picture
01:01:08 <alise> Could he be more upper-class
01:01:19 <alise> Answer: Only if he was the Queen
01:01:22 <pikhq> alise: Yes, but not by much.
01:01:29 <alise> (Kings are lower-class than queens.)
01:01:57 <alise> Having said that I have a completely un-classy friend who goes to Oxford... but not Merton College, so there.
01:02:00 <Quadrescence> I love them more if they typeset mathematics with it
01:02:07 <alise> <pikhq> alise: Fortunately for you, TeX is very, very capable of typesetting prose. :)
01:02:10 <alise> yeah but http://www.tex.ac.uk/tex-archive/macros/latex/contrib/microtype/microtype.pdf
01:02:15 <alise> WHAT AM I SUPPOSED TO SET THESE TO
01:02:20 <alise> Quadrescence: Eqn, dude! Eqn!
01:02:22 <alise> http://en.wikipedia.org/wiki/Eqn
01:02:28 <alise> FUCK YEAH EQN http://www.kohala.com/start/troff/v7man/eqn/eqn2e.ps
01:02:39 <pikhq> alise: SOMETHING ABOUT YOUR MOTHER
01:02:41 <alise> LEAVES SUMS AT A SMALL SIZE NO MATTER WHAT FUCK YEAH
01:02:50 <alise> LITTLE SIGMA SITTING AROUND THERE AIN'T DOIN' SHIT
01:03:11 <Quadrescence> alise: hahaha. I just like how old-style it looks, sort of.
01:03:14 <alise> I love how eqn fails horribly at tall square roots
01:03:17 <alise> look at http://www.kohala.com/start/troff/v7man/eqn/eqn2e.ps
01:03:20 <Quadrescence> Like, it looks like a typesetter actually typeset it
01:03:48 <alise> eqn looks like the kind of stuff you see in old books, yeah
01:03:51 <alise> Quadrescence: i like it apart from the raggedness
01:04:10 <alise> I think TeX-style typesetting became needed when mathematical notation got really huge and crufty and complex
01:04:19 <alise> because we really needed something that could size any damn equation
01:05:20 <alise> I know nothing of Lout
01:06:32 <pikhq> Seems like ATM i've got about modem-level speed. But hell, at this point I'LL TAKE IT.
01:06:44 <pikhq> That at least means I'm getting packets!
01:07:29 <Quadrescence> alise: ftp://ftp.cs.usyd.edu.au/jeff/lout/lout-3.38-user.ps.gz
01:07:53 <alise> is it groff-made-latex or latex-made-groff? I know it's based on a functional language.
01:08:24 <Quadrescence> When you look at it, you'll think groff made it
01:10:17 <alise> I wish there was a digital typesetting system that produced results that look like the old, pre-computer typesettings.
01:10:21 <alise> Nothing quite reaches that level.
01:10:51 <pikhq> Sadly, TeX is the closest there is.
01:11:18 <Quadrescence> It's not sad really. I mean, I am kind of happy it can't be replicated
01:11:43 <alise> It is more convenient to use computers; if we can have convenience and quality, that is all the better.
01:11:45 <Quadrescence> Mostly for faggy reasons like that typesetters (human ones) have their mark in history and even a computer can't do their job :)
01:11:50 <pikhq> And LilyPad tries very hard to manage it for music.
01:11:53 <alise> Nostalgia for older, more difficult ways isn't productive.
01:12:02 <alise> Typesetting is a dead profession, so we can only try to improve the state of play.
01:12:10 <Quadrescence> alise: I don't think it's necessarily better quality
01:12:21 <Quadrescence> I just think so many people suck at typing these days. By typing I mean typesetting
01:12:36 <Quadrescence> Most people just whip something up and publish it :(
01:12:39 <alise> Quadrescence: Well, I don't know. Traditional typesetting has some sort of precise quality that its digital counterpart lacks.
01:13:01 <alise> It just seems to be, somehow, perfect by less than a "pixel".
01:13:06 <Sgeo_2> Will you be sad if programming becomes obsolete due to the Singularity?
01:13:10 <alise> I'm speaking metaphorically -- bullshitting, that is -- here, but you know what I mean.
01:13:32 <alise> Sgeo_2: I kind of think a singularity would adjust my value system so I don't value things like that, because valuing things that aren't there any more would make me less happy.
01:13:34 <Quadrescence> alise: I sort of get what you mean. I still think its the authors' fault
01:13:36 <pikhq> Quadrescence: Well, you see, traditional typesetting was done by people who had labored long to become experts in the craft.
01:13:44 <alise> Quadrescence: *it's; sorry, I arguably have OCD.
01:13:52 <alise> MUST. CORRECT. GRAMMAR
01:14:06 <Sgeo_2> alise, but you just used incorrect grammar!
01:14:07 <pikhq> Modern typesetting is done by retards who think Word is a replacement for a good typesetter.
01:14:20 <pikhq> (even though it, at *best*, is a replacement for a freaking typewriter)
01:14:36 <alise> Fine; MUST. CORRECT. ORTHOGRAPHY
01:14:41 <Sgeo_2> What would LyX be a replacement for?
01:15:15 <alise> LyX seems to me perennially useless.
01:15:31 <alise> If you want precise typesetting, you need to tweak, analyse, even specify, so you need plain LaTeX.
01:15:43 <alise> If you want to take notes, or make a quick document, it is too heavy.
01:15:45 <Quadrescence> LyX is just easier for people, but unfortunately it churns out meh results
01:15:54 <Quadrescence> People think that if something uses latex, it's automagically pretty
01:16:22 <pikhq> This is true for simple typesetting purposes.
01:16:24 * Sgeo_2 likes automagically pretty
01:16:41 <alise> Quadrescence: Have you ever used ConTeXt?
01:16:51 <Sgeo_2> Why shouldn't my homework look somewhat nice
01:16:54 <alise> Apparently it's more focused on publishing than LaTeX.
01:17:02 <alise> "It is especially suited for structured documents, automated document production, very fine typography, and multi-lingual typesetting."
01:17:02 <Sgeo_2> And use styles defined by someone else, because I'm clueless
01:17:17 <alise> "ConTeXt from the ground up is a typography and typesetting system meant to provide users easy and consistent access to advanced typographical controlimportant for general-purpose typesetting tasks. The original vision of LaTeX is to insulate the user from typographical decisionsa useful approach for submitting, say, articles for a scientific journal."
01:17:23 <Quadrescence> Sgeo_2: There's no problem with using other styles and stuff
01:17:39 <Sgeo_2> Quadrescence, the fact that I'm clueless about what makes something look goo
01:17:46 <Quadrescence> Sgeo_2: The problem is when people use another person's style then they deviate from it
01:18:46 * Sgeo_2 would be too clueless to attempt something like tha
01:18:59 <Sgeo_2> But how can minor adjustments "ruin" something?
01:19:21 <Quadrescence> Sgeo_2: That minor dent in that Ferrari ruins it
01:19:47 <Quadrescence> Yes. Of course from a functional standpoint it doesn't.
01:20:09 * Sgeo_2 remembers arguing with his dad about some scratch or something in the new car his dad was going to buy. I didn't really care, my dad did.
01:20:20 <Quadrescence> But I care about the content and aesthetics of a document.
01:20:40 <Quadrescence> Some minor changes are okay, but most often the minor changes are made by someone clueless
01:20:41 <alise> Sgeo_2: even from a functional standpoint,
01:20:49 <alise> the brain/visual system is very sensitive to minor typographical changes
01:20:54 <alise> this is why we have the concept of "rhythm"
01:21:04 <alise> because unless there is a very simple, monotonous rhythm to a text, it's harder to read
01:21:32 <Sgeo_2> But if a document uses the same style througout, then wouldn't that be preserved, even if the style is different?
01:21:50 <pikhq> Man. Reading stuff typeset by TeX makes me really wish that all computer text display was as well-done.
01:22:02 <alise> Sgeo_2: If you change a well-set margin, you ruin the flow.
01:22:21 <Quadrescence> Sgeo_2: It depends on the changes. Not all changes ruin everything. Some changes can be beneficial
01:22:39 <Quadrescence> But changing all headings to bold-italic-underline in Papyrus is ugly
01:23:06 <Sgeo_2> Hm, which is better/worse: Snapdragon processor, or 866MHz Pentium III
01:23:34 <alise> Snapdragon is better, almost certainly.
01:23:41 <alise> It being a modern, 1GHz processor.
01:23:47 <alise> Quadrescence: Ew, Papyrus.
01:24:01 * alise is trying to decide on a typeface to have a go at typesetting H2G2 in.
01:24:07 <alise> It needs to be really, really British.
01:24:28 <alise> Garamond doesn't seem to sit right. It's too flowery, too subtly-serifed-at-the-edges.
01:24:43 <alise> Minion seems too... wimpy?
01:24:51 <uorygl> Don't you want a serif font?
01:25:06 <alise> But, I mean, look: http://upload.wikimedia.org/wikipedia/commons/e/e0/AdobeGaramondSp.svg
01:25:07 -!- augur has quit (Remote host closed the connection).
01:25:12 * uorygl frowns at not being able to remember the meaning of this sentence.
01:25:13 <alise> Little tweaks and curves at the edges.
01:25:16 <uorygl> `translate Kesyyntymisen alkuaikoina koirasta on ollut hyötyä varoittavana vahtina sekä jätteensyöjänä.
01:25:18 <HackEgo> The early days of domestication, dogs have been useful as well as a warning vahtina jätteensyöjänä.
01:25:25 <alise> It's just not something that seems to fit the text.
01:25:29 <pikhq> uorygl: Sans-serif not on computer screens should make everyone cry.
01:25:36 * uorygl waits for alise to explain that learning a language is not about learning the meaning of each individual sentence.
01:25:51 <alise> Quadrescence: Yes, I'm thinking Baskerville.
01:26:04 <Quadrescence> Baskerville is one of my favoritest fonts ever
01:26:07 <uorygl> And, as someone explained to me, that's actually ". . . have been useful as a warning as well as as trash-eaters".
01:26:13 <alise> It's certainly British.
01:26:15 <uorygl> Wow, that's a lot of ass.
01:26:19 <Quadrescence> It is classy, old, pretty, hardcore, yes british
01:26:22 <Sgeo_2> Link to something to show me what Baskerville looks like?
01:26:31 <alise> Sgeo_2: http://en.wikipedia.org/wiki/Baskerville
01:26:43 <alise> I think the H2G2 omnibus I have may be typeset in Baskerville.
01:27:10 <pikhq> uorygl: In order to learn a language, you should use said language.
01:27:15 <Sgeo_2> Except for that uppercase Q, it looks ordinary
01:27:27 <Sgeo_2> I think I like "flowery"
01:27:42 <alise> Baskerville's uppercase Q is quite unique.
01:27:52 <alise> Nothing wrong with flowery. But you have to choose the face that fits the text.
01:28:00 <alise> If there is one thing H2G2 is not, it is flowery.
01:28:03 <Quadrescence> alise: He's saying the rest of the text looks ordinary
01:28:15 <alise> Well, most typefaces look ordinary to most people.
01:28:21 <Sgeo_2> The numbers are quite nice
01:28:24 <alise> They probably shouldn't be noticed; the whole point is to support flow.
01:28:33 <alise> So I can't deride him for his (uninformed) opinion.
01:28:42 <pikhq> Most people think MS Comic Sans is a perfectly workable font.
01:28:42 <Quadrescence> Because they're old-style, which is probably what you are noticing
01:29:04 <alise> Sgeo_2: Why interesting? Because they are text figures?
01:29:10 <Quadrescence> pikhq: It's not that really, just some people can't tell the difference between two fonts
01:29:26 <alise> Sgeo_2: Numbers are /meant/ to hang below the baseline.
01:29:45 <Sgeo_2> I was more noticing the way that some numbers were above others, I think
01:29:48 <uorygl> pikhq: I *guess* I could learn Finnish by talking to people who speak only Finnish.
01:29:53 <Sgeo_2> But without that, they still look interesting
01:30:00 <alise> Quadrescence: do you like Hoefler & Frere-Jones?
01:30:03 <uorygl> I would speak entirely using the nominative, accusative, and genitive, though. :P
01:30:13 <Sgeo_2> The italics look nice. But the regular letters.. are just letters
01:30:31 <Sgeo_2> I guess that's a good thing
01:31:06 <Quadrescence> alise: Yes of course. They make a lot of classy fonts
01:31:19 <alise> Quadrescence: do you like Didot?
01:31:19 <pikhq> Also, Baskerville looks absolutely gorgeous.
01:31:42 <Sgeo_2> The font itself costs money?
01:31:49 <alise> Sgeo_2: so naive :)
01:31:55 <Sgeo_2> Or is there a free Baskerville font somewhere?
01:31:59 <pikhq> Sgeo_2: Yes, welcome to professional fonts.
01:32:10 <Sgeo_2> I was hoping I could use it for homework or something
01:32:16 <Quadrescence> I am pretty sure there's a free version out there.
01:32:20 <alise> Typeface designers make a nice living from their works, as do digitisers.
01:32:21 <pikhq> Be amazed that there *exist* free fonts that don't gouge your eyes out.
01:32:36 <alise> Perhaps it shouldn't be so, perhaps it's yet another instance of that abuse known as copyright.
01:32:40 -!- MizardX has quit (Ping timeout: 276 seconds).
01:32:40 <alise> But that is how it is.
01:32:57 <Quadrescence> Sgeo_2: I feel *embarrassed* to typeset something in baskerville that isn't something I'm very proud of
01:33:14 <pikhq> Though, as a typeface itself cannot be copyrighted, you would be perfectly free to create your own .ttf file of Baskerville.
01:33:33 <alise> Quadrescence: *typeface; come on, let's be pretentious.
01:34:08 <alise> I dunno; I found some series of typefaces describing themselves on Flickr and set in Didot was "I am a whore." I couldn't disagree.
01:34:20 <alise> Quadrescence: Didot is a typeface not a font :(
01:34:26 <alise> or rather... a group of typefaces, but shut up
01:34:48 <alise> Quadrescence: Good <3
01:34:54 <Quadrescence> I do not like Didot, I think a lot of it is very fugly
01:35:03 <alise> http://en.wikipedia.org/wiki/Didot_(typeface)
01:35:19 <alise> Think of "that pretentious French font that looks like wine and cheese".
01:35:22 <alise> You're thinking of Didot.
01:35:35 <alise> Quadrescence: Yeah, I don't find Didot readable; it's like an artistic manifesto gone wrong.
01:35:43 -!- augur has joined.
01:35:58 <alise> And it seems that the bits that are thick are not so important and the bits that are thin that are important, so you feel like you're reading a reverse image of sorts...
01:36:44 * alise decides that H2G2 isn't Hoefler Text
01:36:54 <alise> (I'm just iterating through my favourite typefaces...)
01:36:55 <uorygl> I don't have much to say about typography, but I really like Adobe Caslon Pro.
01:36:57 <Quadrescence> alise: it's like half the letters were just sheared at some angle
01:37:02 <alise> It probably isn't Mercury.
01:37:14 <alise> Caslon is quite nice.
01:37:45 <alise> The italic J is problematic though; I always read it as an F.
01:38:03 <uorygl> This italic J? http://en.wikipedia.org/wiki/File:CaslonSp.svg
01:38:07 <alise> Quadrescence: Hopefully not for text.
01:38:37 <alise> pikhq: Reply to Courier for disambiguity: O0
01:39:05 <alise> Quadrescence: I dunno, I don't see any reason code should be monospaced. IMO we should embrace what the mathematicians found; that rich notation is useful.
01:39:24 <alise> It's like we're still at the analogy of the "writing out equations in words with minimal symbols" stage.
01:39:31 <uorygl> Hmm, the word "awkward" would be much less awkward if it were spelled "acquard". Except that's not really a valid way to spell that.
01:39:35 <Gregor> OK alise, you go have fun with Fortress, we'll keep using real languages.
01:39:38 <alise> That's a failing of computers.
01:39:41 <alise> Gregor: I don't support Fortress.
01:39:41 <uorygl> "Auquard", on the other hand... I like that spelling.
01:39:53 <alise> Gregor: And are you really dismissing all programming language research?
01:40:06 <Gregor> As a programming language researcher ... yes.
01:40:07 <alise> "Gee, nice idea, kid... too bad it's not in a REAL language, like C."
01:40:09 <alise> "Gee, nice idea, kid... too bad it's not in a REAL language, like C."
01:40:22 <Quadrescence> alise: It's easy to print, it's easy to write, it's easy to read, no syntax has to be learned...or minimal at least, and there are enough things to tell things apart
01:40:33 <uorygl> alise: wow, you managed to get all that out of "you go have fun with Fortress, we'll keep using real languages"?
01:40:45 <uorygl> I'll never be that skilled at reading between the lines.
01:41:00 <Quadrescence> We aren't going to use weirdo identifiers to denote NumberOfCustomers
01:41:01 <alise> uorygl: The pattern was obvious, and I've observed facets of it in Gregor before: "[Thing we should do]" "I reject that because it hasn't already been done"
01:41:04 <Gregor> uorygl: I nose, it's amazo.
01:41:14 <alise> Quadrescence: I didn't say we should use mathematical notation.
01:41:24 <Gregor> alise: No, that's just how I express that it's a terrible idea.
01:41:27 <alise> I said we should embrace the /lesson/ of mathematical notation: that something more than text is useful.
01:41:49 <alise> Wow, I should really learn not to express ideas too radical in the... esoteric... programming... languages... channel...
01:41:58 <uorygl> Eh, mathematical notation isn't that hard to type.
01:42:02 <Quadrescence> I mean for jebus christ's sake, inputting math isn't even all that easy
01:42:03 <alise> Quadrescence: Of course; that's a failing of current input designs. We can't go all-out like mathematical notation. I didn't say I knew the happy medium.
01:42:06 <alise> I just said we should find it.
01:42:12 <uorygl> Well, it depends on your editor.
01:42:14 <alise> Of course, it should always be easy to input on computers.
01:42:19 <Sgeo_2> Radical idea: Time is a dodecahedron!
01:42:32 <uorygl> "Time is a dodecahedron" is a type error.
01:42:39 <alise> You're a type error
01:42:44 <pikhq> alise: We should definitely input programs using NOTHING BUT KANJI
01:42:52 <uorygl> Except hmm, maybe that actually makes sense.
01:43:05 <alise> Quadrescence: Ooooooooooh print Alice's Adventures in Wonderland and Through the Looking Glass
01:43:24 <uorygl> I mean, suppose that the domain of physics' time variable is a dodecahedron, with calculus and everything defined in the natural way...
01:43:31 <Quadrescence> alise: you're supposed to ask me for math books and stuff
01:43:36 <alise> Quadrescence: I know.
01:43:45 <alise> But you said "any books", so I am happily misinterpreting to infinity.
01:43:47 <uorygl> Nah, I have no idea how that would work.
01:44:03 <uorygl> Quadrescence: wait, are you saying we can get any book?
01:44:21 * Sgeo_2 would love to have changed scientific thought in a correct way entirely by accident
01:44:34 <uorygl> Sgeo_2: Feynman did that, on a very small scale.
01:44:51 -!- Gracenotes has quit (Ping timeout: 252 seconds).
01:44:56 <alise> Quadrescence: Print "Making Resin for Dummies".
01:45:03 <uorygl> Someone handed him a huge, indecipherable diagram and started spieling about it.
01:45:15 <pikhq> Feynman just plain did a lot.
01:45:28 <uorygl> So he found a random symbol, guessed that it denoted a valve, and said, "Uh... what happens if this valve gets stuck?"
01:45:45 <alise> Feynman was an algorithm that turned women into algorithms that turn bongos into theorems.
01:45:49 <alise> He was the first metamathematician.
01:45:59 <uorygl> So the guys looked at that spot for a moment, and said, "Gosh, you're absolutely right."
01:46:06 <uorygl> And so they thought he was a genius.
01:46:11 <uorygl> Which, you know, was kind of true.
01:46:23 <alise> A first-order mathematician, indeed!
01:46:39 <Sgeo_2> What was the diagram about, and what were the consequences?
01:46:49 <Sgeo_2> Or are you making it up, because Google's not giving me any love
01:46:51 <pikhq> I need to grab my copy of "Surely You're Joking, Mr. Feynman!" from parent's house...
01:46:54 <Quadrescence> Everyone listen to my fave song http://www.youtube.com/watch?v=XGK84Poeynk
01:47:08 <uorygl> The diagram was of some piece of machinery that processed nuclear material.
01:47:20 <uorygl> If that mistake hadn't been caught... something really bad might have happened. :P
01:48:51 <uorygl> Who's it by and what's it about?
01:50:45 <pikhq> Quadrescence: 習 英語 or else 私 will keep on 使ing 漢字 at 君. Oh, and yes, this 是, 中 真, 勿事 but 英語. :P
01:50:55 <uorygl> Heh heh, "Sinfonia Domestica".
01:51:10 <uorygl> A symphony about stuff that happens around the house.
01:51:40 <Quadrescence> pikhq: fucking fuck or else fuckers will keep on fucking fucking at fuck. Oh, and yes, this fucker, fucking fucker, fucked but fucking fucks
01:52:14 <uorygl> `translate 習 英語 or else 私 will keep on 使ing 漢字 at 君. Oh, and yes, this 是, 中 真, 勿事 but 英語
01:52:16 <HackEgo> Learn English or else private will keep on making ing character at jun. Oh, and yes, this is, in truth, do not matter but in English
01:52:56 <pikhq> I do believe that went through as Chinese.
01:53:20 <pikhq> Worked surprisingly well, though.
01:53:33 <uorygl> `translatefromto jp en 習 英語 or else 私 will keep on 使ing 漢字 at 君. Oh, and yes, this 是, 中 真, 勿事 but 英語
01:53:35 <HackEgo> Learn English or else private will keep on making ing character at jun. Oh, and yes, this is, in truth, do not matter but in English
01:53:46 <pikhq> "Learn English or else I will keep on using kanji at you. Oh, and yes, this is, in fact, nothing but English."
01:54:03 <pikhq> `translatefromto zh en 習 英語 or else 私 will keep on 使ing 漢字 at 君. Oh, and yes, this 是, 中 真, 勿事 but 英語
01:54:05 <HackEgo> Learn English or else private will keep on making ing character at jun. Oh, and yes, this is, in truth, do not matter but in English
01:54:07 <uorygl> I will keep on learning English, or else use kanji ing at you. Oh, and yes, this Shi true medium, but 勿事 English
01:54:26 <uorygl> That's Google's actual translation from Japanese.
01:54:45 <uorygl> `translatefromto zh en ウィキプロジェクトのための名前空間新設に関する投票が行なわれています。投票期限は、5月30日です。
01:54:48 <HackEgo> Voting period は, May 30 です.
01:54:59 <uorygl> `translatefromto zh en ウィキプロジェクトのための名前空間新設に関する投票が行なわれています。
01:55:03 <HackEgo> ウ ィ キ プ ロ _ GUI ェ ク Suites Circular ta の の に former space off the new line na si ru voting ga i ma si わ れ て.
01:55:15 <uorygl> Former space off the new line!
01:56:02 <Sgeo_2> Hmm, PdaNet has a trial period thing. After the trial period, I can't use https:// anymore
01:56:21 <pikhq> uorygl: Try translating from Japanese?
01:58:19 <pikhq> Quadrescence: いいえ、日本語がどんどん上手になる。それが日本語じゃなくて、漢字で書いた英語だった。
01:59:22 <pikhq> I wonder how that gets mangled.
01:59:31 <pikhq> `translate いいえ、日本語がどんどん上手になる。それが日本語じゃなくて、漢字で書いた英語だった。
01:59:33 <HackEgo> Japanese, not that it was written in English characters.
02:00:06 <pikhq> What did it even do with the first sentence?
02:00:09 <pikhq> `translate いいえ、日本語がどんどん上手になる。
02:00:11 <HackEgo> No, the Japanese have become increasingly well.
02:00:25 <pikhq> *Wow* that mangles it.
02:00:54 <pikhq> I wasn't expecting perfection, what with Japanese being very context-dependent, but... Damn.
02:00:57 <pikhq> That's just *bad*.
02:01:08 <pikhq> `translate でも、これは本当に日本語だ。
02:01:10 <HackEgo> However, this is really Japanese.
02:01:24 <pikhq> That's a workable translation.
02:02:47 <Quadrescence> ( ﾟ∀ﾟ)ｱﾊﾊ八八ﾉヽﾉヽﾉヽﾉ ＼ / ＼/ ＼ 一二三┻━┻ 一二三┻━┻ 一二三┻━┻ 一二三┻━┻ 一二三┻━┻}。々°)ﾉ
02:05:08 <uorygl> `translatefromto jp en 阿波罗8号是阿波罗计划中的第二次载人飞行任务，三位执行此任务的宇航员分别为指令长弗兰克·博尔曼、指令舱驾驶员詹姆斯·洛威尔以及登月舱驾驶员威廉·安德斯。
02:05:11 <HackEgo> Members 威廉安德斯.
02:05:40 <augur> uorygl: thats chinese not japanese.
02:05:52 <augur> `translatefromto ch en 阿波罗8号是阿波罗计划中的第二次载人飞行任务，三位执行此任务的宇航员分别为指令长弗兰克·博尔曼、指令舱驾驶员詹姆斯·洛威尔以及登月舱驾驶员威廉·安德斯。
02:05:53 <HackEgo> Members 威廉安德斯.
02:05:55 <uorygl> I know. I'm wondering what will happen if I translate from the wrong language
02:06:23 <alise> Quadrescence: do you like exljbris?
02:06:30 <uorygl> `translatefromto zh en 阿波罗8号是阿波罗计划中的第二次载人飞行任务，三位执行此任务的宇航员分别为指令长弗兰克·博尔曼、指令舱驾驶员詹姆斯·洛威尔以及登月舱驾驶员威廉·安德斯。
02:06:34 <alise> I really want Calluna, http://www.josbuivenga.demon.nl/calluna.html
02:06:46 <uorygl> `translatefromto zh en 威廉安德斯
02:07:45 <alise> i'm just trying to get an excuse to be mock-raged at you :|
02:09:52 <alise> Quadrescence: please suck on multiple cacti
02:10:44 <Sgeo_2> ARGH, I hate this thing
02:12:39 <alise> Quadrescence: Still, at least you have one; I have no language preference, as I'm continually saddled with the troubling knowledge that all existing languages are inadequate, plus the desire to make new, better ones, which never works out quite as I plan. I have several language antipreferences, though.
02:12:58 <alise> One is definitely C. It can't even handle basic data structures without a page of memory handling and bookkeeping.
02:13:05 <Quadrescence> But at the time I need to hit the books, I use C
02:13:08 <alise> And that's simply the antithesis of what I want.
02:14:25 <Sgeo_2> How is Python inadequate? How is Haskell inadequate?
02:15:14 <alise> Python is inadequate because it suffers from the endless tangle of impurity, and GvR does not truly understand languages: he believes, for instance, that iteration obsoletes general tail-call elimination. He is an imperative programmer out of his league, designing something he has no idea how to design.
02:15:18 <pikhq> Haskell is probably the closest in common use to what alise would like. What with not sucking giant donkey balls and all.
02:15:26 <alise> The result is a mediocre mess with a glaze of cleanness.
02:15:52 <alise> Haskell is inadequate because its type system cannot express many things it should be able to, because its syntax is lacking in some ways, and a general bag of odds-and-ends of flaws.
02:15:56 <alise> It's good, but it doesn't sit right.
02:17:09 <Sgeo_2> How is PHP inadequate? *ducks*
02:17:38 <Quadrescence> alise: I think you'll find that if you ended up designing a language and making something marketable, you'd have a hard time
02:17:41 <pikhq> PHP is wrong in most of the ways that can be managed.
02:17:59 <Sgeo_2> Which is worse, PHP or Perl?
02:18:10 <alise> Quadrescence: Yes, but GvR is not so much detached from Rasmus, the creator of PHP, who hates languages and doesn't enjoy programming.
02:18:27 <alise> One level higher, certainly, but he certainly does not have the range of understanding of languages to make the right decisions.
02:18:32 <alise> And I believe Python is simply not a good language.
02:18:55 <alise> I could easily make a marketable language with all the API trimmings, I just don't want to. What's the point? It'd probably suck.
02:19:02 <Sgeo_2> Why would someone who doesn't enjoy programming make a programming language?
02:19:20 <Sgeo_2> To make things "easy" for non-programmers or something?
02:19:26 <pikhq> PHP is most definitely worse than Perl.
02:20:07 <pikhq> Perl suffers from having too much flexibility and a somewhat write-only syntax.
02:20:30 * Sgeo_2 finds that, at my level, Haskell is a bit read-only
02:20:39 <pikhq> PHP suffers from being "designed" by pretty much throwing shit together.
02:21:09 <Quadrescence> Fortunately in PHP and Python, you can Get Things Done
02:21:29 <Quadrescence> Not write some fancy "beautiful code" that is hard to use
02:21:43 <pikhq> Quadrescence: This is the only reason those languages have not been written off as complete and utter failures.
02:21:59 <Sgeo_2> pikhq, are you saying that Python should be a failure???
02:22:01 <Quadrescence> Of course, because they've done the opposite of fail for people
02:22:31 <pikhq> Sgeo_2: I'm saying I liked it better when it was called Tcl.
02:23:00 <alise> <Sgeo_2> Why would someone who doesn't enjoy programming make a programming language?
02:23:02 <alise> To get things done.
02:23:55 <alise> Anyway, "getting things done" is usually employed as a deus ex machina against good design; this crappy design works because people have used it for so long that they have figured out how to cobble together something that works; your superior design is more immature and so hasn't formed the framework it needs yet; therefore, the old way is better than the new way.
02:24:13 <alise> Not everyone who wants change and dislikes the status quo is a hopeless ideologue.
02:25:10 <Sgeo_2> My dad wants me to find a company that makes a universal laptop charger, something with Lithium-Ion or something. The company's name starts with Digi
02:25:20 <Sgeo_2> He wants me to find the company
02:25:34 * Sgeo_2 wishes he was more certain about the company's name
02:25:44 <alise> http://www.google.co.uk/search?hl=en&client=firefox-a&hs=VY8&rls=com.ubuntu%3Aen-GB%3Aunofficial&q=digi*+laptop+battery+charger+lithium+ion&aq=f&aqi=&aql=&oq=&gs_rfai=
02:26:03 <Sgeo_2> I didn't know you could do * in Google searches!
02:26:43 <Sgeo_2> Hm, I can't exclude digital for obvious reasons, but I don't want digi* to only match because of digitalk
02:29:52 * Sgeo_2 screams wildly at the slowness of his Internet connection
02:30:29 -!- Sgeo_2 has quit (Remote host closed the connection).
02:32:12 -!- Sgeo has joined.
02:32:55 <Sgeo> There are too many companies named Digitech!
02:32:57 -!- augur has quit (Remote host closed the connection).
02:35:53 -!- augur has joined.
02:37:07 * Sgeo decides to just look for chargers for his laptop's model
02:37:58 <Sgeo> Hm, that's twice that I typed charter instead of charger
02:48:11 <Sgeo> F My Internet connection
02:48:15 <alise> Higher-order incantations.
02:48:17 <Sgeo> I'm going off of IRC to play more Gish
02:49:24 -!- Sgeo has quit (Quit: Leaving).
02:49:59 -!- SgeoN1 has joined.
02:50:38 <SgeoN1> Gish > crappy Intetnet use
03:03:56 -!- poiuy_qwert has quit (Ping timeout: 260 seconds).
03:04:20 -!- poiuy_qwert has joined.
03:14:34 -!- Gracenotes has joined.
03:22:49 -!- Oranjer has left (?).
03:26:04 -!- lament has joined.
03:30:16 <alise> Quadrescence: incidentally how easy is it to create a new "style" in LaTeX?
03:30:53 <alise> logically, you should have to design everything from scratch rather than mangling an existing style; but on the other hand, that sounds like an incredibly laborious task
03:34:13 <Quadrescence> I'm still working on mine and I've been working at it for months
03:34:56 <alise> Quadrescence: And, presumably, I'd want to define one for the H2G2 books...
03:35:05 <alise> What have you gotten me into.
03:35:26 <pikhq> I hate packet-per-second Internet.
03:36:15 <alise> My printing of the books has some odd quirks; "CHAPTER N" is, if I recall correctly, set vertically to the left of the opening of the text.
03:37:54 <alise> Note to self; remember not to do any typesetting.
03:37:58 <alise> Quadrescence: Ah, but going too far is wonderful!
03:38:38 <Quadrescence> You can read all about the memoir class in the documentation, all the typographic theory
03:38:46 <alise> Quadrescence: I wouldn't want to produce anything of anything less than significantly greater quality than I've seen.
03:39:11 <alise> That means I'm up against my hardback edition; that means I'm going to have to tweak /something/ along the line
03:39:16 <lament> /mode #esoteric -b lambent
03:40:22 <alise> Quadrescence: I'll probably do that. But tweaking an existing style is error-prone, as you said.
03:40:42 <alise> Besides, I won't really feel pleased with the work unless I make it as good a I can, so I'll have to see...
03:40:53 <alise> Quadrescence: Do you want to live in a world where every book looks the same? :)
03:41:10 <Quadrescence> There are actually a lot of modifiable parameters
03:43:05 <alise> Quadrescence: I don't remember the videos; have you put any inscriptions on the cloth of the hardbacks? I'm not sure what the correct term is.
03:43:30 <Quadrescence> no I haven't. I don't have the right tools, so I leave the cloth blank
03:43:57 <alise> I vastly prefer them to dust jackets. I wonder how it's done
03:44:09 <alise> like maybe they embed little gnomes into the cloth...
03:44:28 <alise> How expensive, exactly, is penis
03:45:03 <alise> order of magnitude?
03:46:41 <alise> Now I can't believe that.
03:47:54 <alise> Quadrescence: I'm imagining a shelf filled with hardback books with nothing written on the spine
03:48:02 <alise> IN A WORLD... WHERE BOOK-PRINTERS RUN OUT OF MONEY...
03:48:25 <lament> ill sell you my penis for $5
03:49:38 <alise> see i can afford $5.
03:54:13 <alise> drop caps are weird
03:56:51 -!- Gregor has quit (Ping timeout: 252 seconds).
04:01:04 <alise> like seriously, I mean, what's the point.
04:17:11 <alise> SgeoN1: "Google Abandons Direct Sales of Nexus One"
04:18:32 <SgeoN1> How so? It's going to be available from retailers apparently
04:19:35 <alise> But locked to a carrier, I'll bet; and at a higher price.
04:20:16 <SgeoN1> And I guess AT&T won't offer it
04:21:08 <alise> That's a good thing
04:21:10 <SgeoN1> Although maybe I should have been patient and waited for the EVO
04:21:53 <alise> It's too big! And Sprint-only.
04:23:05 <SgeoN1> My dad said he'd be willing to switch to Sprint
04:23:49 <alise> Then do so! Sprint are the best network around.
04:24:08 <alise> That's my advice. The EVO will be nice, if you have hands big enough. Although apparently Sense isn't all that hot. Eh. You'll figure something out.
04:24:10 -!- augur has quit (Read error: Connection reset by peer).
04:24:27 <alise> I suppose I'll probably dream of typography and H2G2 now. Should be interesting. Bye!
04:24:29 -!- augur has joined.
04:24:30 -!- alise has quit (Quit: Leaving).
04:29:49 -!- augur has quit (Ping timeout: 265 seconds).
04:39:15 -!- Gregor-L has joined.
04:47:21 -!- SgeoN1 has quit (Read error: Connection reset by peer).
04:47:24 -!- SgeoN1 has joined.
04:55:30 -!- Sgeo has joined.
05:03:56 <Gregor-L> AnMaster, pikhq and anybody else who might care: http://codu.org/music/op13/GRegor-op13-wipp1.ogg
05:04:36 <pikhq> Wait, op13? What of op12?
05:04:45 <Gregor-L> I'm scavenging the good parts.
05:05:00 <Gregor-L> Including the beginning, lest ye be confused :P
05:05:03 <pikhq> So, op13 is your 12th opus.
05:05:25 <Gregor-L> Some people like to skip the number 13, I'm skipping the number 12 instead.
05:05:58 <pikhq> Anyways. Y'know what's totally awesome for learning a foreign language?
05:06:01 <pikhq> Watching shows in it.
05:06:07 <Sgeo> A bit jarring a bit before 1:30
05:06:32 <Gregor-L> Sgeo: Yes, that is intentional :P
05:36:46 -!- Gregor-L has changed nick to Gregor.
05:43:40 -!- lament has quit (Ping timeout: 240 seconds).
05:59:39 -!- uorygl has quit (Read error: Connection reset by peer).
05:59:58 -!- uorygl has joined.
06:00:03 -!- uorygl has quit (Read error: Connection reset by peer).
06:04:58 -!- uorygl has joined.
06:14:36 -!- augur has joined.
06:21:12 -!- augur has quit (Remote host closed the connection).
06:21:21 -!- augur has joined.
06:26:27 -!- poiuy_qwert has left (?).
06:43:13 -!- SgeoN1 has quit (Ping timeout: 265 seconds).
06:54:43 -!- SgeoN1 has joined.
06:56:25 -!- Sgeo has quit (Remote host closed the connection).
07:21:46 -!- ais523 has joined.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:17:53 -!- SgeoN1 has quit (Quit: Bye).
09:08:26 -!- Phantom_Hoover has joined.
09:21:06 -!- lament has joined.
09:43:05 -!- oerjan has joined.
09:46:35 -!- BeholdMyGlory has joined.
10:05:45 -!- tombom has joined.
10:08:50 -!- lament has quit (Ping timeout: 246 seconds).
10:09:27 <Phantom_Hoover> Zooming in on the Sierpinski Gasket is oddly difficult.
10:10:04 <oerjan> you keep hitting the empty spots? :)
10:11:22 <oerjan> maybe it's about dimension - i vaguely recall the mandelbrot border has hausdorff dimension 2, or something
10:11:38 <oerjan> (despite being a curve)
10:11:41 -!- FireFly has joined.
10:11:57 <oerjan> while the sierpinski gasket has less
10:12:10 <oerjan> so that might make the mandelbrot border easier to hit
10:12:40 <Phantom_Hoover> I think it's because the iterative process that makes the Sierpinski Gasket causes the filled-in areas to peter away to nothing.
10:13:29 <oerjan> oh you mean it's not drawing it perfectly? or something else
10:14:14 <Phantom_Hoover> No, it's because when you cut out the middle triangles, the filled-in area is 3/4 of what it was before.
10:14:19 <oerjan> it shouldn't be two hard to zoom in perfectly - it looks locally everywhere like a sierpinski gasket, or two joined in a corner
10:14:41 <Phantom_Hoover> So carrying on ad infinitum causes the filled-in bits to approach 0.
10:14:43 <oerjan> Phantom_Hoover: yes. but that's precisely what makes it have dimension < 2
10:15:39 <oerjan> i guess there's lot of combinatorics in it
10:16:00 <Phantom_Hoover> The even and odd numbers in Pascal's Triangle are in a pattern similar to the Sierpinski Gasket.
10:18:16 <oerjan> hm i guess that makes sense, if odds are at corners
10:18:25 <Phantom_Hoover> So logically, Pascal's Triangle contains infinitely more evens than odds.
10:18:59 <oerjan> well to get an odd number in the pascal's triangle, one of the parents must be odd
10:19:56 <oerjan> so if you have an all-even range, most of it is preserved on the next step
10:20:00 <Phantom_Hoover> So you would get large blocks of evens being encroached by odds as you move downwards.
10:20:11 <oerjan> while an all-odd range turns nearly all-even
10:21:05 <Phantom_Hoover> IIRC if you use a 1D automaton on a triangular grid where each cell is the XOR of the two above it, you get the gasket.
10:21:44 <oerjan> ah yes. come to think of it, i am sure i've tested that out some time.
10:23:43 <oerjan> (for one thing, i once went through all 2-cell 2-value 1d automata (well there are just 16 and some are isomorphic) to see how easy the long term behavior was to predict
10:24:27 <oerjan> and XOR and its equivalent dual (EQV ?) were the hardest since it needed summing binomials
10:25:09 <oerjan> (by predict, i mean calculating without simulating all intermediate steps)
10:26:55 <oerjan> (isomorphic = equivalent, using mathspeak here)
10:27:45 <Phantom_Hoover> You can get Sierpinskiesque behaviour in Life, under certain circumstances.
10:28:50 <oerjan> "The Hausdorff dimension of the boundary of the Mandelbrot set equals 2 as determined by a result of Mitsuhiro Shishikura."
10:30:01 <oerjan> hm i recall some discussion previously on the channel that made a sierpinski gasket using a long line
10:30:24 <Phantom_Hoover> Yeah, in Life superstrings (very long orthogonal lines of cells) create a messy gasket.
10:36:14 <oerjan> <uorygl> `translatefromto jp [...] <-- i think that's ja not jp?
10:45:57 <oerjan> <Quadrescence> AnMaster: Who is german in here <-- jix is
10:46:52 <oerjan> not precisely active though
10:49:01 <AnMaster> * [jix] idle 95:00:38, signon: Wed May 12 12:47:44 <-- nice idle time
10:51:26 <AnMaster> <Phantom_Hoover> So logically, Pascal's Triangle contains infinitely more evens than odds. <-- hm, but aren't there an infinite number of both?
10:51:53 -!- copumpkin has joined.
10:52:12 <AnMaster> if so I would like to see a proof of that
10:52:40 <AnMaster> oerjan, then there can't be infinitely more evens than odds can it?
10:52:48 <oerjan> there is also density to consider
10:53:14 <AnMaster> oerjan, but you can get a bijection of the even and the odds in Pascal's triangle, no?
10:53:22 <oerjan> although being similar to sierpinski's triangle, it probably fluctuates from 0 to 1 and back
10:53:37 <oerjan> well of course, they're both infinite and countable
10:54:02 <AnMaster> oerjan, can't you get a bijection between uncountable of same cardinality?
10:54:14 <oerjan> in fact it goes from all odd to all but 2 even in one step, perhaps 2^n-1 to 2^n rows?
10:54:31 <oerjan> of course you can, that's the definition of same cardinality
10:54:49 <oerjan> but there's only one countably infinite cardinality
10:57:46 <oerjan> !haskell let pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt] in take 16 pt
10:58:00 <EgoBot> [[1],[1,1],[1,2,1],[1,3,3,1],[1,4,6,4,1],[1,5,10,10,5,1],[1,6,15,20,15,6,1],[1,7,21,35,35,21,7,1],[1,8,28,56,70,56,28,8,1],[1,9,36,84,126,126,84,36,9,1],[1,10,45,120,210,252,210,120,45,10,1],[1,11,55,165,330,462,462,330,165,55,11,1],[1,12,66,220,495,792,924,792,495,220,66,12,1],[1,13,78,286,715,1287,1716,1716,1287,715,286,78,13,1],[1,14,91,364,1001,2002,3003,3432,3003,2002,1001,364,91,14,1],[1,15,105,455,1365,3003,5005,6435,6435,5005,3003,1365,455,105,
10:59:08 <AnMaster> Quadrescence, was that supposed to be a joke?
10:59:57 <oerjan> AnMaster: if you look at those rows, numbered from 0, then the 2^n-1 rows are all odd and the 2^n are all even except the 1's at the end
11:00:19 <AnMaster> oerjan, are there rows with mixed odd/evens in the middle?
11:00:44 <oerjan> um all the other rows have both even and odd i believe
11:01:04 <oerjan> oh and the p'th row for a prime p has all but the ends divisible by p, i recall
11:01:56 <oerjan> hm it certainly _looks_ like row 9 has all divisible by 3, maybe it's a p^n thing
11:02:27 <AnMaster> oerjan, or maybe it is just coincidence?
11:02:41 <oerjan> well it's not coincidence for p^1 or 2^n
11:03:02 <oerjan> well let's check a few more
11:04:03 <oerjan> !haskell let pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt] in take 100 $ map (foldl1' gcd) pt
11:04:31 <oerjan> !haskell pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . take 100 $ map (foldl1' gcd) pt
11:04:55 <oerjan> !haskell pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . take 100 $ map (foldl' gcd 0) pt
11:05:18 <oerjan> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . take 100 $ map (foldl' gcd 0) pt
11:05:21 <EgoBot> [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
11:05:39 <oerjan> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . take 100 $ map (foldl' gcd 0) (init.tail$pt)
11:05:42 <EgoBot> [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
11:06:01 <oerjan> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . take 100 $ map (foldl' lcm 0) (init.tail$pt)
11:06:10 <EgoBot> [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
11:06:22 <copumpkin> !haskell main = print $ iterate (\row -> zipWith (+) ([0] ++ row) (row ++ [0])) [1]
11:06:49 <oerjan> why the _heck_ are they defining that as 0
11:07:08 <oerjan> it's far more useful to have lcm 0 n as n
11:07:10 <copumpkin> !haskell iterate (\row -> zipWith (+) ([0] ++ row) (row ++ [0])) [1]
11:07:27 <AnMaster> oerjan, it is programming not math. Even if it is haskell...
11:07:35 <oerjan> copumpkin: EgoBot does not handle infinite output well
11:07:45 <oerjan> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . take 100 $ map (foldl1' lcm) (init.tail$pt)
11:07:56 <copumpkin> !haskell main = print . take 100 $ iterate (\row -> zipWith (+) ([0] ++ row) (row ++ [0])) [1]
11:08:18 <oerjan> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . take 100 $ map (foldl1' gcd) (init.tail$pt)
11:08:18 <AnMaster> hm I think stderr goes to /dev/null for egobot
11:08:20 <EgoBot> [1,2,3,12,10,60,105,280,252,2520,2310,27720,25740,24024,45045,720720,680680,12252240,11639628,11085360,10581480,232792560,223092870,1070845776,1029659400,2974571600,2868336900,80313433200,77636318760,2329089562800,4512611027925,4375865239200,4247163320400,4125815796960,4011209802600,144403552893600,140603459396400,136998242488800,133573286426580,5342931457063200,5215718803323600,219060189739591200,214081549063691400,209324181306720480,20477365562613960
11:08:22 <EgoBot> [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
11:08:31 <AnMaster> oerjan, what was the second line?
11:09:06 <AnMaster> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . take 100 $ map (foldl1' gcd) (init.tail$pt)
11:09:09 <EgoBot> [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
11:09:18 <oerjan> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . take 100 . zipWith [0..] $ map (foldl1' gcd) (init.tail$pt)
11:09:32 <AnMaster> oerjan, see I pasted your line above
11:09:44 <AnMaster> oerjan, don't claim that the 1,1,1,1... line arrived first at your end
11:10:13 <AnMaster> oerjan, was it supposed to happen though?
11:10:19 <oerjan> AnMaster: actually i was confused by the fact i _made_ the 1,1,1,1... line first and thought it had timed out
11:10:52 <copumpkin> the question remains of who generated the 1,2,3,12 then
11:11:00 <oerjan> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . take 100 . zip [0..] $ map (foldl1' gcd) (init.tail$pt)
11:11:04 <oerjan> copumpkin: that was mine
11:11:04 <EgoBot> [(0,1),(1,1),(2,1),(3,1),(4,1),(5,1),(6,1),(7,1),(8,1),(9,1),(10,1),(11,1),(12,1),(13,1),(14,1),(15,1),(16,1),(17,1),(18,1),(19,1),(20,1),(21,1),(22,1),(23,1),(24,1),(25,1),(26,1),(27,1),(28,1),(29,1),(30,1),(31,1),(32,1),(33,1),(34,1),(35,1),(36,1),(37,1),(38,1),(39,1),(40,1),(41,1),(42,1),(43,1),(44,1),(45,1),(46,1),(47,1),(48,1),(49,1),(50,1),(51,1),(52,1),(53,1),(54,1),(55,1),(56,1),(57,1),(58,1),(59,1),(60,1),(61,1),(62,1),(63,1),(64,1),(65,1),(66
11:11:21 <AnMaster> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . take 100 $ map (foldl1' lcm) (init.tail$pt)
11:11:25 <EgoBot> [1,2,3,12,10,60,105,280,252,2520,2310,27720,25740,24024,45045,720720,680680,12252240,11639628,11085360,10581480,232792560,223092870,1070845776,1029659400,2974571600,2868336900,80313433200,77636318760,2329089562800,4512611027925,4375865239200,4247163320400,4125815796960,4011209802600,144403552893600,140603459396400,136998242488800,133573286426580,5342931457063200,5215718803323600,219060189739591200,214081549063691400,209324181306720480,20477365562613960
11:11:49 <AnMaster> oerjan, so did you find out the 9th row thing yet?
11:12:06 <AnMaster> oerjan, couldn't you just check 3 or 4 cases by hand?
11:12:24 <oerjan> the next case that's at all interesting is 25
11:12:30 <oerjan> so no, i wouldn't like that
11:12:41 <copumpkin> but was too lazy to fix it, it's past 6am :)
11:12:51 <AnMaster> copumpkin, yes it is. It is 12:12
11:13:08 <AnMaster> copumpkin, heck it is past it a lot of times this year alone
11:13:13 <fungot> Available: agora alice c64 ct darwin discworld europarl ff7 fisher ic irc jargon lovecraft* nethack pa speeches ss wp youtube
11:13:19 <fungot> AnMaster: recent british literature, besides including the three or four greatest fantaisistes of the present is that it's too damned stupid to question the people around the station; but his attitude toward the matter was by this time have found the other neighboring gate to the abyss, and these dark ruins were in truth primordial sarkomand.
11:13:33 <fungot> AnMaster: as we proceeded. out the window i had left seemed involved in a passage from any given dimensional plane to the next opening in the roof glittered the pale pole star, fluttering as if alive in the stone staircase had been encountered, and the
11:14:05 <AnMaster> hm copumpkin are you (relatively) new here or?
11:14:25 <oerjan> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . take 10 . zip [0..] $ pt
11:14:28 <EgoBot> [(0,[1]),(1,[1,1]),(2,[1,2,1]),(3,[1,3,3,1]),(4,[1,4,6,4,1]),(5,[1,5,10,10,5,1]),(6,[1,6,15,20,15,6,1]),(7,[1,7,21,35,35,21,7,1]),(8,[1,8,28,56,70,56,28,8,1]),(9,[1,9,36,84,126,126,84,36,9,1])]
11:14:37 <AnMaster> copumpkin, nah no issue, just wasn't able to find your nick in my mental lookup table
11:14:38 <oerjan> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . take 10 . zip [0..] $ (init.tail$pt)
11:14:41 <EgoBot> [(0,[1,1]),(1,[1,2,1]),(2,[1,3,3,1]),(3,[1,4,6,4,1]),(4,[1,5,10,10,5,1]),(5,[1,6,15,20,15,6,1]),(6,[1,7,21,35,35,21,7,1]),(7,[1,8,28,56,70,56,28,8,1]),(8,[1,9,36,84,126,126,84,36,9,1]),(9,[1,10,45,120,210,252,210,120,45,10,1])]
11:15:16 <oerjan> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . take 100 . zip [0..] $ map (foldl1' gcd . init . tail) pt
11:15:22 <EgoBot> [(0,input.26654.hs: Prelude.init: empty list
11:15:38 <copumpkin> is there usually this much haskell in here?
11:15:39 <oerjan> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . drop . take 100 . zip [0..] $ map (foldl1' gcd . init . tail) pt
11:15:56 <AnMaster> copumpkin, varies widely. But usually oerjan can figure out a working command faster than that
11:16:07 <AnMaster> copumpkin, btw fungot is written in befunge98 in case you didn't know
11:16:08 <fungot> AnMaster: " then you're not sure. you saw the fnord ancient and fnord rock strata fully verified all of lakes bulletins, and proved that these pinnacles had been towering up in exactly the same, that door's coming open," i answered, " i can't tell you yet which it was called. the being is spoken of as holding all knowledge, and its
11:16:09 <fungot> http://git.zem.fi/fungot/blob/HEAD:/fungot.b98
11:17:04 <AnMaster> copumpkin, oh and oerjan is our resident mathematician
11:17:33 <copumpkin> I come from #haskell, so I'm used to being surrounded by them
11:17:37 <oerjan> Phantom_Hoover: i'm trying to demonstrate something, just messing up the code
11:17:57 <AnMaster> Phantom_Hoover, of mathematician? Ask him, I don't remember... I saw some paper he had written once iirc. Didn't understand one sentence of it :P
11:18:09 <oerjan> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . drop . drop . take 100 . zip [0..] $ map (foldl1' gcd . init . tail) pt
11:18:24 <oerjan> why the _heck_ is that a type error
11:18:36 <AnMaster> oerjan, is it? where does egobot say so?
11:20:01 <oerjan> !haskell import Data.List; pt = [1]:[zipWith(+)(0:r)(r++[0])|r <- pt]; main = print . drop 2 . take 100 . zip [0..] $ map (foldl1' gcd . init . tail) pt
11:20:05 <EgoBot> [(2,2),(3,3),(4,2),(5,5),(6,1),(7,7),(8,2),(9,3),(10,1),(11,11),(12,1),(13,13),(14,1),(15,1),(16,2),(17,17),(18,1),(19,19),(20,1),(21,1),(22,1),(23,23),(24,1),(25,5),(26,1),(27,3),(28,1),(29,29),(30,1),(31,31),(32,2),(33,1),(34,1),(35,1),(36,1),(37,37),(38,1),(39,1),(40,1),(41,41),(42,1),(43,43),(44,1),(45,1),(46,1),(47,47),(48,1),(49,7),(50,1),(51,1),(52,1),(53,53),(54,1),(55,1),(56,1),(57,1),(58,1),(59,59),(60,1),(61,61),(62,1),(63,1),(64,2),(65,1),(
11:20:27 <oerjan> ok it's true for 25 and 49
11:20:35 <AnMaster> oerjan, but not for most other ones?
11:20:56 <oerjan> those are the only untested squares of primes in that list
11:21:11 <AnMaster> oerjan, why not just take the first 100 relevant numbers or such?
11:21:53 -!- tombom_ has joined.
11:22:02 * Phantom_Hoover notices that most c/4 diagonal Life spaceships seem to be glider tagalongs
11:22:24 <oerjan> AnMaster: because i'm having trouble enough making a list of all the gcd's
11:22:58 <copumpkin> what are you trying to compute? I think I missed the beginning of this
11:23:06 <oerjan> Phantom_Hoover: i'm testing the theory that all elements of the p^n'th row of pascal's triangle other than the 1's are divisible by p for a prime p
11:23:32 -!- tombom has quit (Ping timeout: 240 seconds).
11:23:38 <oerjan> AnMaster: i don't know how to filter on prime powers easily
11:23:40 <AnMaster> oerjan, so not just squares but ^3 and ^4 and so on too?
11:23:56 <AnMaster> oerjan, check if each number is a power of a prime?
11:24:00 <oerjan> also it's just for primes, 36 breaks as you can see
11:24:02 <AnMaster> might not be a one-liner any more...
11:24:09 <oerjan> AnMaster: that's horrible to do, i think
11:24:19 <copumpkin> there's a really short way to get primes :P
11:24:29 <copumpkin> prime powers in order probably requires more work though
11:24:32 <oerjan> AnMaster: it requires factorization?
11:24:54 <oerjan> it would be easier to generate all than test for them directly
11:24:57 <AnMaster> oerjan, well sure it isn't fast, but for numbers of reasonable size it can't be too hard
11:25:12 <oerjan> AnMaster: it's not a one-liner
11:25:54 <AnMaster> oerjan, after all something like the command line tool factor is fast for (random example) 347772662826551
11:26:12 <AnMaster> 347772662826551 is a prime apparently
11:26:50 <AnMaster> 987326589764578281478231293 is fast but 9873265897645782814782312932 takes several seconds
11:27:13 <oerjan> AnMaster: i'm trying to demonstrate, not check for myself. also a factor program would obviously use advanced algorithms
11:27:29 <AnMaster> maybe because it has two large prime factors instead of one much larger and many small?
11:27:41 <AnMaster> $ factor 987326589764578281478231293
11:27:41 <AnMaster> 987326589764578281478231293: 3 3 7 47 32265841 10334261253277693
11:27:42 <AnMaster> $ factor 9873265897645782814782312932
11:27:42 <AnMaster> 9873265897645782814782312932: 2 2 8005230950461 308337946735853
11:27:59 <oerjan> testing small divisors is much easier
11:28:01 <AnMaster> oerjan, it is still impressive that this just takes a few seconds on an old Sempron...
11:28:37 <AnMaster> oerjan, yet finding out that 10334261253277693 is prime didn't take a lot of time
11:28:46 <AnMaster> oerjan, doesn't haskell have some built in factoring thingy?
11:28:59 <AnMaster> mathematicians made it after all...
11:29:00 <oerjan> no prime testing has easy algorithms
11:29:18 <oerjan> _good_ factoring is _hard_
11:29:20 <AnMaster> oerjan, but surely there is import Something.Prime?
11:29:33 <oerjan> oh something on hackage, probably
11:29:39 <oerjan> it's not a _core_ feature
11:29:45 <AnMaster> oerjan, ah not part of the haskell platform?
11:30:26 -!- sebbu2 has quit (Ping timeout: 264 seconds).
11:30:33 <oerjan> time to google or wikipedia, i think
11:31:38 <AnMaster> oerjan, anyway the "semi-naive" way of doing this would be generating all primes up to, say, 500 using sieve of E., then trying a number of powers for them, say 2..20
11:31:47 <AnMaster> oerjan, this shouldn't be too bad complexity
11:32:54 <oerjan> well i'm passing on to checking references at this point
11:34:17 -!- myndzi\ has joined.
11:34:44 -!- MizardX has joined.
11:35:42 <oerjan> "It can be deduced from this that (n; k) is divisible by n/gcd(n,k)
11:36:33 <oerjan> ok if n is a prime power then gcd(n,k) is less when k is not 0 or n
11:36:55 <oerjan> so n/gcd(n,k) must contain at least one factor of the prime
11:37:23 -!- myndzi has quit (Ping timeout: 264 seconds).
11:38:01 <oerjan> (i couldn't paste the wp gif...)
11:38:47 <oerjan> "A somewhat surprising result by David Singmaster (1974) is that any integer divides almost all binomial coefficients."
11:38:59 <MizardX> (n; k) = n!/(k!*(n-k)!). n! | n, and k! | gcd(n,k), so (n; k) | n/gcd(n,k)
11:39:00 <oerjan> that's relevant to the odd/even case too
11:39:58 <oerjan> MizardX: are you writing those | backwards?
11:40:15 <oerjan> (well the first one at least)
11:40:53 <MizardX> a | b = a is dividable by b ... dunno if that is the normal order
11:41:03 <oerjan> no that's the opposite
11:42:01 <oerjan> and you still have the middle one reversed from the rest
11:42:16 <MizardX> (n; k) = n!/(k!*(n-k)!). n | n!, and gcd(n,k) | k, so n/gcd(n,k) | (n; k)
11:42:46 <oerjan> i don't think that's a correct proof fwiw
11:44:31 <AnMaster> (wrt: <oerjan> "A somewhat surprising result by David Singmaster (1974) is that any integer divides almost all binomial coefficients.")
11:45:08 <oerjan> AnMaster: http://en.wikipedia.org/wiki/Binomial_coefficient#Divisibility_properties
11:46:56 <oerjan> MizardX: you have a | b and c | d => ac | bd but what you're doing doesn't fit into that
11:50:30 <oerjan> (and anyway there's no way you can get this while ignoring the (n-k)! part)
11:53:20 <oerjan> AnMaster: ok, basically you take the first n rows, and count the fraction of that part of the triangle that is divisible by d. as n -> infinity, that fraction goes to 1.
11:54:33 <AnMaster> oerjan, and d is selected to be?
11:54:44 <AnMaster> oerjan, I mean if d > n then the fraction will be 0
11:54:59 <oerjan> you take the limit of n while keeping d fixed
11:55:09 <oerjan> so d << n most certainly in the limit
11:55:30 <oerjan> (well any nonzero integer)
11:55:35 <AnMaster> oerjan, right, if d = 1 then the fraction is 1
11:55:47 * oerjan swats AnMaster -----###
11:56:00 <oerjan> it's true for _all_ nonzero d.
11:56:02 <AnMaster> oerjan, hey, degenerate cases are fun
11:58:00 <AnMaster> oerjan, also I love how a circle is defined in terms of a cone and a cone is defined in terms of a circle
11:58:58 <oerjan> i will have none of your circular definitions!
12:01:05 <AnMaster> oerjan, hm wikipedia says "An integer n ≥ 2 is prime if and only if all the intermediate binomial coefficients $\binom n 1, \binom n 2, \ldots, \binom n{n-1}$ are divisible by n."
12:01:19 <AnMaster> (nice that the tex copies from the image
12:01:31 <AnMaster> (even though I had to replace two newlines with a $ on either side)
12:04:01 <MizardX> Anti-proof: 2/1 !| 2/(1*2)
12:06:26 <MizardX> ... of my claim of a proof
12:08:51 <oerjan> MizardX: oh i thought you had found a counterexample
12:09:08 <oerjan> but (2;1) = 2 so that's fine
12:09:38 <Phantom_Hoover> With Life, it'd really have to be anti-physics, though.
12:11:55 <oerjan> life breaks down quite frequently
12:13:43 <Phantom_Hoover> But there are some things which should be consolidated into an abstract theory.
12:14:51 -!- oerjan has quit (Quit: Later).
12:17:52 <AnMaster> Phantom_Hoover, btw I saw yesterday something mentioned about a wireworld variant that could "grow" new wires
12:18:03 <AnMaster> iirc self replication was possible
12:19:24 <AnMaster> Phantom_Hoover, didn't check the details
12:19:37 <AnMaster> was somewhere on googlecode, some golly-rule-repo thingy
12:19:55 <AnMaster> http://code.google.com/p/ruletablerepository/ from browser history
12:20:11 <AnMaster> http://code.google.com/p/ruletablerepository/wiki/TheRules#WireWorld_and_derivatives
12:21:29 <AnMaster> not sure if it is included by default
12:22:10 -!- ais523 has quit (Remote host closed the connection).
12:23:44 <Phantom_Hoover> AnMaster: Golly 2.1 gives access to online repositoried.
13:17:38 -!- Phantom_Hoover has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100423140709]).
13:23:36 -!- Tritonio_GR has quit (Read error: Connection reset by peer).
13:28:21 -!- alise has joined.
13:48:45 -!- Phantom_Hoover has joined.
13:59:13 <alise> Why does nobody print 1 as the I-type symbol any more?
13:59:18 <alise> I see it universally in old mathematical texts.
14:05:54 <alise> No, no, they're printed like Is.
14:06:04 <alise> http://upload.wikimedia.org/wikipedia/en/2/25/Quadrature_of_Circle_Cajori_1919.png
14:06:06 <alise> Look at the 1s here.
14:06:11 <alise> It isn't confusing!
14:08:28 <alise> Computer Modern certainly doesn't have that 1.
14:08:34 <alise> Nor does AMS Euler or any of the common mathematical fonts.
14:08:48 <alise> Or indeed any text fonts I know.
14:13:04 <alise> Like the regular 1 you see all the time.
14:13:18 <alise> Slanty top bit, serifed bottom bit.
14:13:21 <alise> Instead of serifed top and bottom bit.
14:13:37 -!- Phantom_Hoover has quit (Remote host closed the connection).
14:15:02 <alise> http://upload.wikimedia.org/wikipedia/commons/a/ab/Pm1234_Euler1755_I-V.png
14:15:14 <alise> Here is some text which shows it better; older, too; vintage 1755 Euler.
14:15:34 -!- Phantom_Hoover has joined.
14:16:21 * Phantom_Hoover notices that Golly has a script that supports 3D Life.
14:22:57 <alise> <alise> http://upload.wikimedia.org/wikipedia/commons/a/ab/Pm1234_Euler1755_I-V.png
14:22:57 <alise> <alise> Here is some text which shows it better; older, too; vintage 1755 Euler.
14:23:14 <alise> Does it scale up the neighbour stuff based on the dimensions or just keep it the same (which would be silly)?
14:23:49 <Phantom_Hoover> The script does the calculations, then draws the universe on a 2D CA>
14:24:09 <Phantom_Hoover> It's also fantastically slow, and they keys are unintuitive.
14:31:31 <alise> And I assume that the display is not very easy to see.
14:32:31 <alise> If there was an easy 3D library I'd make a proper 3D Life.
14:32:37 <alise> But, there is not.
14:32:41 <alise> Or at least, not that I know of.
14:33:09 <alise> Phantom_Hoover: hey do you know of any like big huge collections of interesting life patterns i could dump in a folder and have golly recognise?
14:33:13 <alise> downloading separate .rle files is so meh
14:36:02 <alise> Meh. Not fast enough.
14:36:51 <alise> Does Golly understand the organisation?
14:37:34 <Phantom_Hoover> IIRC you can just dump the decompressed directory into the Patterns directory and Golly can read it.
14:37:46 -!- maedhros777 has joined.
14:37:58 <maedhros777> Does anyone here know how to output a number in Brainfuck?
14:38:40 <alise> I think there's something on the Brainfuck algorithms page on the wiki for that.
14:39:08 <alise> Well, this is divmod: http://esoteric.voxelperfect.net/wiki/Brainfuck_algorithms#Divmod_algorithm
14:39:22 <alise> And a one or two digit number in decimal is just two divmods.
14:39:39 <alise> Admittedly you will need a lookup table.
14:39:55 <maedhros777> That's a long algorithm, though -- this is for a golf problem
14:40:39 <maedhros777> Actually...i could do it by checking if a cell is equal to 9, then put it in the next cell if it is
14:41:27 <alise> Phantom_Hoover: Hey, do you know what's the longest period oscillator? I know there's >200 but it has no page on LifeWiki.
14:41:38 <alise> I also wonder what the slowest spaceship is. Probably that replicator + a deletion machine, but that's chaeting :-)
14:43:51 <alise> Is there any "pulsating" spaceship? I mean something that moves like a slug; it contracts, then expands out again, slightly further forward.
14:44:14 <Phantom_Hoover> Though it tends to be on the order of one or two cells.
14:44:51 <alise> I was imagining this big slug getting all tiny.
14:45:02 <alise> (I thought you meant a one-to-two cell spaceship for a second there, and got really confused.)
14:47:30 <alise> Golly's default collection is so tiny, urgh
14:47:44 <alise> Phantom_Hoover: So has anyone emulated a non-Lifelike CA in Life?
14:48:07 <alise> I'm imagining one of those dot-matrix-printer type dealies printing each line of CA as it goes, then somehow "scanning" the printed output to do the next generation.
14:48:48 <alise> But that's 2D, anyway.
14:49:34 <alise> GOL is so rich, I'd like to life in a life simulation
14:50:30 <Phantom_Hoover> A vanilla cell, the metapixels and the Deep Cell, which simulates two universes at once.
14:50:48 <alise> Why simulate two at once? I'm curious.
14:51:27 <alise> Also, I don't have the metapixel in my golly version :( just a 512x512 one and the deep ceel
14:52:03 <alise> Memory storage still sucks in Life, yeah?
14:52:21 <Phantom_Hoover> The Spartan UCC has unlimited-size registers, I think.
14:53:03 <Phantom_Hoover> There's a script in Python that converts normal Life patterns into metapixels.
14:53:42 <alise> Phantom_Hoover: I meant as far as speed and stuff goes.
14:54:05 <alise> It'd be nice to have a Turing-machine style tape where the head can move at lightspeed.
14:54:26 <alise> (And thus retrieve things from memory and give them back to the main computer at lightspeed.)
14:54:29 <alise> Phantom_Hoover: Yeah, IMO that sucks :P
14:54:32 <alise> Lightspeed is awesome
14:55:04 <alise> We must harness the powah!
14:55:10 <Phantom_Hoover> There isn't even a way to turn a practical lightspeed signal around a corner, yet.
14:56:01 <alise> But you can have a lightspeed tunnel that emits and takes sub-lightspeed gliders, yeah?
14:56:31 <alise> I don't know; I'm just asking.
14:56:38 <alise> I assume it's possible. Otherwise it isn't really communication.
14:56:53 <alise> So have stretches of LHC^Wlightspeed tunnel for as long as you can make them, then into a sub-light-speed cornering device, which then feeds the gliders into another straight lightspeed tunnel.
14:56:54 <Phantom_Hoover> As I said yesterday, no-one has come up with a usable diagonal lightspeed wire.
14:57:07 <alise> A lightspeed tunnel
14:57:11 <alise> you feed in spaceships
14:57:17 <alise> bits are transmitted at lightspeed
14:57:22 <alise> then come out the other end as spaceships
14:57:36 <alise> All you need is a horizontal and vertical one of these, plus a very simple take-in-spaceship-and-emit-a-new-one-diagonally piece
14:57:41 <Phantom_Hoover> And yes, you can take the signal out of lightspeed to turn a corner, but the signal circuitry required would be huge, complex and slow.
14:57:43 <alise> Then just plug them together; voila, turner-corning.
14:57:53 <alise> Phantom_Hoover: So exiting lightspeed is difficult?
14:58:03 <alise> How is it really communication then, if everything has to go at light speed?
14:58:12 <alise> Circuitry can't always operate that quickly.
14:58:20 <alise> I guess this is why lightspeed is chaotic.
14:58:53 <alise> "Here is a sample Herschel-based oscillator with a 12-digit-prime period" ;; whoa
14:58:59 <alise> Phantom_Hoover: I saw it but did not understand it or run it.
15:00:23 <alise> What is the latest Golly?
15:00:43 <Phantom_Hoover> The version in the Ubuntu repositories is ancient, BtW.
15:00:52 <alise> Okay, so I can see the lightspeed telegraph and it looks like the LHC.
15:00:56 <alise> I don't understand it though.
15:02:38 <alise> Phantom_Hoover: Should I upgrade?
15:02:49 <alise> Also, the telegraph appears to have some sort of latency; after sending one blip of information, you must wait until it repairs itself to send another.
15:02:52 <alise> So it's not really lightspeed, is it?
15:02:58 <alise> For one bit, sure, but not sustained.
15:05:04 <alise> Heh, the machinery looks beautiful on hyperspeed.
15:05:40 <alise> If you zoom in you can see the cable jittering.
15:05:48 <alise> Phantom_Hoover: so I should upgrade Golly yeah?
15:13:12 <alise> [ ]golly_2.1-1.diff.gz09-May-2010 16:06 11K
15:13:13 <alise> [ ]golly_2.1-1.dsc09-May-2010 16:06 1.1K
15:13:13 <alise> [ ]golly_2.1-1_amd64.deb12-May-2010 13:05 2.0M
15:13:13 <alise> [ ]golly_2.1-1_i386.deb10-May-2010 12:05 1.9M
15:13:13 <alise> [ ]golly_2.1.orig.tar.gz09-May-2010 16:06 2.3M
15:14:54 <alise> jane@jane-desktop:~$ sudo dpkg -i --force-depends-version golly_2.1-1_amd64.deb
15:14:55 <alise> (Reading database ... 331562 files and directories currently installed.)
15:14:55 <alise> Preparing to replace golly 2.1-1 (using golly_2.1-1_amd64.deb) ...
15:14:55 <alise> Unpacking replacement golly ...
15:14:55 <alise> dpkg: golly: dependency problems, but configuring anyway as you request:
15:14:55 <alise> golly depends on libperl5.10 (>= 5.10.1); however:
15:14:59 <alise> Version of libperl5.10 on system is 5.10.0-19ubuntu1.1.
15:15:01 <alise> golly depends on libwxbase2.8-0 (>= 2.8.10.1); however:
15:15:03 <alise> Version of libwxbase2.8-0 on system is 2.8.9.1-0ubuntu6.
15:15:05 <alise> golly depends on libwxgtk2.8-0 (>= 2.8.10.1); however:
15:15:07 <alise> Version of libwxgtk2.8-0 on system is 2.8.9.1-0ubuntu6.
15:19:27 * alise just compiles golly herself
15:19:49 <alise> Phantom_Hoover: so a 3D life would use voxels, right?
15:21:32 <alise> http://www.ibiblio.org/e-notes/Life/Gliders.htm Gliders in 3D Life
15:22:52 <alise> 3D Life seems quite chaotic on random patterns.
15:22:56 <alise> A universe of insanity.
15:23:14 <uorygl> Surely there are multiple ways to 3Dify Life.
15:23:21 <alise> http://fbim.fh-regensburg.de/~saj39122/doefe/ another 3D Life
15:23:56 <alise> uorygl: only one obvious way; instead of (x,y) have (x,y,z); use 3D instead of 2D neighbours; and scale up the birth/death numbers accordingly
15:24:19 <uorygl> Well, yes, but there are multiple ways to choose those numbers.
15:25:44 <alise> Divide by 8, multiply by 26. Perhaps that's naive.
15:25:58 <uorygl> It is naive, but it may work. :P
15:26:16 * uorygl ponders spelling that naïve simply because he can.
15:26:29 <alise> The main issue is nice visualisation and editing.
15:26:35 <alise> Nave is a nice spelling.
15:28:06 <uorygl> Is crème brûlée a nice spelling?
15:28:59 <uorygl> Is that just a different spelling of the English word 'role', as in "What is his role in this ordeal?"?
15:29:12 <alise> His rle in the proceedings was primarily to mess everything up; he was too nave to handle his responsibility.
15:30:02 <alise> Phantom_Hoover: Python doesn't work with Golly on 64-bit machines. Will this be a problem?
15:30:49 -!- Warrior` has joined.
15:31:06 <alise> Oh noes a Warrior`
15:31:13 <alise> Who is pronounced only slightly differently to wareya
15:31:30 <alise> Another person in here; more to the point, who are you?
15:31:33 <uorygl> And significantly differently from Warrigal.
15:31:57 <alise> It's a relatively simple question; what brings you here?
15:32:08 <alise> Well, that's reasonable.
15:32:23 <Phantom_Hoover> alise: IIRC, you need to change the location of the Python .so files sometimes.
15:32:29 <alise> I was about to fire off the procedural "we're not about magick or any other kind of esoterica" but I don't suppose anyone in #brainfuck would be looking for that.
15:32:34 <alise> Unless they took the channel name a bit too literally.
15:32:41 <alise> Phantom_Hoover: one of the symbols is named diferently in the 64-bit one
15:32:53 <Warrior`> #brainfuck theres a refrence about #esoteric
15:33:25 <Phantom_Hoover> That would fit, given that we are ostensibly the umbrella channel for all esoteric languages.
15:33:53 <alise> But only ostensibly, mind.
15:34:03 <alise> Although I guess our current topic, the Game of Life, basically counts as an esolang.
15:34:10 <alise> More like an esouniverse, but there you go.
15:35:11 <alise> I have. As I said, I saw it, and it looked like the LHC; but I didn't understand it.
15:35:17 <alise> It certainly didn't seem to do anything with the information being transmitted.
15:35:43 <Phantom_Hoover> It does transmit a LWSS, but it's extremely inefficient.
15:38:03 -!- maedhros777 has quit (Remote host closed the connection).
15:39:06 -!- Warrior` has quit (Quit: Leaving).
15:39:34 <AnMaster> fizzie, I found something wonderful wrt. panoramic photos. A (possibly) cheap way to do panoramic head. It depends on if you have the needed stuff already, I happen to have that. Lots of it in fact. Will tell you more when you respond :P
15:41:11 -!- alise has quit (Remote host closed the connection).
15:45:09 -!- alise has joined.
15:50:21 <alise> Phantom_Hoover: so efficient lightspeed transmission is an open problem.
15:50:55 <alise> How hard can it be?
15:51:34 <Phantom_Hoover> It's not finding stable signals that's the problem, it's getting them into a manipulable form.
15:52:24 <alise> So, basically, the problem is that the way the signals are transmitted is chaotic
15:53:34 * alise gawps at metapixel-p216-gun
15:53:35 <Phantom_Hoover> Most of the LST's overhead comes from the fact that 10 signals are necessary per bit, and the tape needs to be repaired afterwards
15:53:36 <alise> It's... so huge...
15:53:56 <alise> Phantom_Hoover: Yes; I bet real lightspeed communication is impossible.
15:54:00 <alise> Because of the latency, the lag between signals.
15:54:04 <alise> You can send one bit at lightspeed, sure.
15:54:10 <alise> But sustained communication won't be at lightspeed.
15:54:59 <alise> has anyone used a metapixel to run a metapixel?
15:57:05 <alise> Phantom_Hoover: What do you think about my conjecture that true sustained lightspeed communication is impossible?
15:57:15 <alise> Maybe we could do something like quantum entanglement with a very long wire.
15:57:39 <alise> Admittedly, lightspeed != instant, but if we can recognise a very small group of cells and keep the "connector" which switches them far enough away not to interfere...
15:58:57 <Phantom_Hoover> By true sustained lightspeed communication do you mean sending data or processing?
15:59:13 <alise> I meant sending; there's lag between sending lightspeed communications due to repair.
15:59:21 <alise> So I bet that the maximum /sustained/ speed is far less than c, because of the lag.
16:00:05 <Phantom_Hoover> The LST can send repeatedly along the same wire; it only needs to repair a fixed length.
16:00:47 <alise> Yes, but if you monitor the right end: a blip gets fired, the wire is fucked up for a few generations, then it reforms, /then/ the next blip hits it.
16:00:52 <alise> There's that lag between sending two blips.
16:00:56 <alise> That lag is significantly slower than c.
16:01:05 <alise> So /sustained/ communication - non-stop, constant communication - cannot happen at c.
16:02:22 <uorygl> Well, what's non-stop, constant communication?
16:02:35 <uorygl> At best, you can transmit information only once per generation.
16:04:05 <uorygl> And it doesn't make any sense to say "that lag is significantly slower than c". That lag is measured in time per information sent, not time per distance.
16:04:19 <alise> Consider that we have a certain machine that, with the appropriate communication mechanism, wants to set one blip per generation going off.
16:04:35 <alise> You can't do this, since after setting one blip off, it takes several generations for the wire to repair itself sufficiently to accept another blip.
16:04:37 <uorygl> Then it can just use a bunch of these lightspeed telegraphs.
16:04:52 <uorygl> No new technology is required.
16:05:05 <alise> I'm trying to think of an occasion where it would need an infinite amount of them but I can't. Fair enough.
16:05:17 * alise causes an accident in the LHC
16:05:24 <alise> Black holes are sucking up all the gliders!!!
16:05:37 <alise> They're spreading and demolishing the wire at both ends!!
16:06:27 <alise> Whoa, I think this thing is actually eating the tape away at lightspeed
16:06:41 <alise> Truly a black hole.
16:07:01 <alise> Ah, it's just an information packet going backwards
16:07:26 <alise> I like how it's turning the tape into nice little evenly-spaced oscillators.
16:07:37 <Phantom_Hoover> It's not an information packet, it just looks like it.
16:07:40 <alise> With gigantic gobs of dark matter every so often.
16:07:43 <alise> Phantom_Hoover: okay
16:08:06 <alise> The dark matter appears to be when the destroyer collides with an information packet.
16:08:18 <alise> Phantom_Hoover: Pretty impressive that it survives the collision, or rather gets rebuilt just after it
16:08:49 <Phantom_Hoover> I think it's the natural way for a stretched line of beehives to burn.
16:08:52 -!- lament has joined.
16:08:59 <alise> The thing eating the wire away at lightspeed that looks exactly like a mirrored packet
16:09:30 <alise> Extraordinarily, the left piece of machinery is still trying to shoot out gliders
16:09:51 <alise> They're just being obliterated though.
16:10:08 <alise> Wait, this thing is eating the wire then pooping out oscillators.
16:10:56 <Phantom_Hoover> Interestingly, the ash of oscillators and still-lifes that is left behind when a pattern burns is very resilient.
16:11:19 <alise> I guess that is logic WHAT why did one of the oscillators appear one cell above the others
16:11:23 <Phantom_Hoover> I've tested it as a shielding material, and it can withstand a glider gun pointed at it.
16:11:28 <alise> Phantom_Hoover: Heh
16:11:32 <alise> Phantom_Hoover: I guess it makes sense
16:11:50 <alise> There was a bloody battle there, tons of cells fought it out; by definition, only those tough enough to survive chaotic conditions will remain at the end.
16:12:04 <alise> Otherwise they would have been destroyed. So it's a tautology that the debris of collisions is tough
16:12:24 <alise> Phantom_Hoover: Hey, you're using naturally-made resources from the Field to reinforce your products.
16:12:40 <alise> "This strong material was formed from chaotic collisions millions of generations ago..."
16:12:59 <alise> "We mind it from a desolate part of the Inner Field and brought it to our products to make sure they have the best defences possible."
16:13:46 <Phantom_Hoover> It's impossible to synthesise, though, and creating it by burning junk would produce too many gliders to be safe.
16:14:12 <alise> Of course, I don't think there's anything wrong with doing it in a lab beforehand :-)
16:14:14 <alise> I just find it amusing.
16:14:21 <alise> How is it impossible to synthesise, though?
16:14:26 <alise> You mean while running the thing it's used in?
16:14:39 <alise> You could probably use a program to create a very resilient material.
16:14:47 <alise> The left side of the telegaph is now completely destroyed; just debris. It isn't even firing gliders any more.
16:15:10 <alise> The right side is as of yet undamaged, but the destroyer will reach it soon.
16:15:30 <uorygl> Wow, this looks really dangerous.
16:15:43 <alise> Phantom_Hoover: The funny thing is, the right side hasn't had any idea of any of this happening :-)
16:15:46 <uorygl> There are a bunch of queen bees here; some of them are emitting gliders that are colliding with each other and producing fireballs.
16:15:54 <uorygl> However, every time, the fireballs die out before causing any damage.
16:15:59 <alise> At least, assuming observers in the Game of Life obey regular laws.
16:16:06 <alise> uorygl: Best. Movie. Ever.
16:16:15 <alise> Queen bees shooting space ships that explode into fireballs.
16:16:35 <uorygl> What's more, a random glider recently flew in and managed to not make the reaction destroy itself.
16:16:58 <alise> Pause it, screenshot it, post the rle.
16:17:11 <uorygl> Oh, and occasionally, it emits a glider toward some debris...
16:17:22 <uorygl> This has to be artificial. I'm betting that the original pattern contained this exact thing.
16:18:08 <uorygl> http://pastebin.com/HEHLhKxK
16:18:17 <uorygl> I don't know what format that is; let's hope it's RLE. :P
16:18:42 -!- tombom__ has joined.
16:18:47 <alise> I'll let Phantom_Hoover investigate.
16:19:03 <alise> The destroyer is almost at the first component of the right end.
16:19:10 <alise> But it didn't affect it!
16:19:14 <alise> Here's the big one.
16:19:43 <alise> The transmitter is no longer working.
16:19:49 <alise> Here comes the next wave of spaceships.
16:19:49 <uorygl> My, this is making a dramatic exit.
16:20:12 -!- rodgort has quit (Quit: Coyote finally caught me).
16:20:13 <uorygl> It's emitting a glider every few hundred generations, and that glider is hitting some debris, causing it to draw ever closer...
16:20:16 <alise> The machinery is ... surviving, gliders and spaceships are still coming.
16:20:24 <alise> It's spreading, though
16:21:01 <Phantom_Hoover> Firing spaceships at fire tends to make the fire travel up the stream.
16:21:03 <alise> All the little personnel operating this transmission station are going to die eventually.
16:21:14 * alise puts it on hyperspeed
16:21:19 <alise> Just normal speed.
16:21:31 -!- rodgort has joined.
16:21:36 <alise> Whoa, I made it go far too fast.
16:21:48 -!- tombom_ has quit (Ping timeout: 260 seconds).
16:21:58 <alise> Well, spoiler, here's what happens to the right side: The entire machine is replaced by a bunch of oscillators and still lifes.
16:22:08 <alise> Complete oblivion. Even the bits at the outskirts.
16:22:51 <alise> I don't even know HOW it destroyed the bits at the outskirts.
16:22:57 <uorygl> This debris produced a ship right next to the machine's output.
16:23:06 <uorygl> I'm guessing that the next glider out will destroy the machine instantly.
16:23:16 <uorygl> Nope, it just deleted the boat.
16:23:17 <alise> Phantom_Hoover: There is a known oscillator that looks like the sun, right?
16:24:03 <alise> Yeah, just didn't recognise it.
16:24:08 <alise> No, perhaps 4 period.
16:24:14 <uorygl> And the machine's dead, turned into a single glider gun!
16:24:22 <uorygl> Let's see whether its beam moves forward or backward.
16:24:41 <alise> Phantom_Hoover: The oscillator that's a backwards S and a block and a little blip bats backwards and forwards, I like that one.
16:25:07 <alise> uorygl: is this queen-bee-turn.rle?
16:25:08 <alise> Phantom_Hoover: wat
16:25:17 <alise> Ooh, is it something unknown?
16:25:37 <Phantom_Hoover> If it's low period and in ash, then almost certainly no.
16:26:02 <alise> Somewhere in the vicinity of ~29-31.
16:26:10 <alise> Can't tell for sure; too complex.
16:26:51 <uorygl> alise: possibly; I don't know what that is.
16:26:55 <alise> WTF; it doesn't work out of context.
16:27:12 <alise> There are no alive cells near, but copying it to its own file makes it do something completely different.
16:27:31 <uorygl> I guess you're doing something wrong.
16:27:40 <alise> Phantom_Hoover: Sure, I'll give you the RLE it's in and the coords.
16:28:15 <Phantom_Hoover> If it's in DRH-oscillators, the information thing has a list of what they're all called.
16:28:28 <alise> It's in the debris.
16:28:58 <alise> Phantom_Hoover: http://filebin.ca/nmovkc/telegraph-in-peril.rle; there are multiple but the one I'm looking at is around XY= 308 1,392
16:32:33 <alise> Queen bees sure are popular today.
16:32:40 <alise> Phantom_Hoover: How come copying it out doesn't work?
16:34:05 <alise> I wonder what I did wrong.
16:34:27 <alise> Are there any methuselahs that, after their lifespan, kill every one of their cells?
16:34:31 <alise> That would be cool.
16:34:34 <alise> Like a really, really long fuse.
16:35:15 <uorygl> Does Golly come with the Caterpillar?
16:35:54 <alise> How do you get it to download patterns like it said it could?
16:35:59 <alise> Also, yeah, I was thinking of that one.
16:37:08 <alise> http://holyspiritvictorious4ever.blogspot.com/2009/03/1111-phenomenon-links-2012-to-baghdad.html
16:37:21 <alise> Phantom_Hoover: Help. Well that makes sense
16:37:35 <alise> is the lifewiki archive good?
16:38:07 <Phantom_Hoover> It's good for quick lookup of simple spaceships and oscillators.
16:38:25 <alise> Okay, I downloaded it; how do I install it?
16:38:28 <alise> Or does Golly not do that?
16:38:54 <Ilari> Caterpillar? That huge ship with odd speed?
16:39:11 <Phantom_Hoover> If you click on patterns in the archive Golly loads them. Ilari: Yes.
16:39:48 <alise> Yes, but I want it in my breeders directory
16:39:59 -!- Sgeo has joined.
16:40:20 <Ilari> Completely crazy: Caterpillar gun. :->
16:40:38 <alise> For every single one? No.
16:40:42 <alise> I'll just do it manually.
16:41:06 * alise organises Patterns folder into Golly and jslife
16:41:17 <Ilari> If one knows glider synthethis for some spaceship, can one transform that into gun for that spaceship?
16:41:36 <alise> It's just a restricted subset of a universal constructor, right?
16:41:42 <alise> Running in a loop using that synthesis.
16:41:54 <alise> So it's obviously possible, it's just a matter of making it... not a universal constructor-computer
16:42:21 <AnMaster> <AnMaster> any idea where fizzie may be? <alise> Running in a loop using that synthesis. <-- XD (yes I know it wasn't related, but...)
16:42:27 -!- Gregor has quit (Ping timeout: 252 seconds).
16:42:31 <alise> Cool, jslife/applications/unit110cell.lif.
16:42:34 <alise> Still not quite the nice output I'd like.
16:43:57 * uorygl scrolls through the Caterpillar at one-quarter scale using the down-arrow in Golly.
16:44:01 <Ilari> The idea I have for transforming glider synthethis into gun is to trace the pattern backwards into glider field and then use (high-perioid) glider guns plus reflectors to construct that field.
16:44:19 <alise> Phantom_Hoover: (10,2)c/10 -- what does /this/ notation mean?
16:44:28 <alise> I know Mc/N, but what is (X,Y)c/N?
16:44:36 <alise> It's a fuse; s-a bunch of 9s.lif.
16:45:03 <uorygl> It seems this ought to get up to speed pretty quickly.
16:45:21 <Ilari> Ah, its not a ship, ships can't have that high speed.
16:45:22 <AnMaster> <alise> Translates? <alise> As in rotates? <-- I thought translate == move x/y, not "rotate
16:45:26 <Phantom_Hoover> If it's a fuse, it's probably the period of the burning bit and how much it burns.
16:45:26 <alise> I love how straight lines make sierpinski triangles in Life.
16:45:41 <alise> Almost enough to make you believe Wolfram
16:46:08 <AnMaster> alise, they do? in the same line or?
16:46:14 <Phantom_Hoover> Straight lines are also the only things that can propagate into a vacuum at lightspeed.
16:46:18 <AnMaster> does it build one in a direction?
16:46:19 <alise> Just a straight line downwards creates two sierpinski triangles
16:46:27 <alise> Phantom_Hoover: We discussed it here a while ago.
16:46:31 * uorygl runs the Caterpillar at hyperspeed.
16:46:38 <Phantom_Hoover> AnMaster: The junk left behind is a Sierpinski triangle.
16:46:44 <alise> AnMaster: A slightly off-straight line makes a "mucky" Sierpinski triangle.
16:46:49 <Ilari> What's highest possible sum of velocities in directions for spaceship? 1/2 (LWSS, MWSS, HWSS and glider all have that velocity sum)?
16:46:57 <alise> This is made up of less defined debris, but still looks right.
16:47:04 <uorygl> Ilari: what makes that a sum?
16:47:14 <uorygl> In any case, yeah, 1/2 is the fastest possible spaceship.
16:47:18 <Ilari> uorygl: x and y component sum.
16:47:34 <uorygl> Then you might be able to go faster.
16:47:38 <alise> AnMaster: four or five or so
16:48:24 <Sgeo> http://gawker.com/5539717/steve-jobs-offers-world-freedom-from-porn?skyline=true&s=i
16:48:29 <alise> What's the RLE syntax for N-pixel down line?
16:48:50 * uorygl makes a little modification to the Caterpillar.
16:48:57 <AnMaster> Phantom_Hoover, which generation? or some average over several?
16:49:08 <AnMaster> also is there any "straight line drawing" in golly
16:49:10 <uorygl> Naturally, this modification destroys it.
16:49:45 <Sgeo> Why is c/2 the fastest possible spaceship?
16:49:49 <alise> Sgeo: Because it is.
16:50:05 <alise> Sgeo: Jobs is a hardass, we know this. However listening to Gawker about anything is idiotic, especially as they knowingly bought a stolen next-gen iPhone to write a review, then were total dicks about it when everyone found out
16:50:08 <alise> Also, it simply is so.
16:50:10 <Phantom_Hoover> A C spaceship is impossible on a non-toroidal universe.
16:51:38 <alise> Is this margin large enough to contain it?
16:51:43 <Ilari> How long caterpillar is front to back as its longest?
16:51:43 <alise> IRC is basically just distributed marginalia.
16:51:49 <alise> And yes, I just wanted an excuse to use the word marginalia.
16:51:49 <Sgeo> I was talking to a friend, mentioned how the SIM card is not in, and he suggests that I should have just bought an iPod Touch
16:52:11 <Sgeo> Bye in advance Phantom_Hoover
16:52:14 <alise> Sgeo: Well, he is wrong.
16:52:17 <AnMaster> Phantom_Hoover, I must be doing something wrong... I can't spot any sierpinski triangle...
16:52:25 <alise> Phantom_Hoover: When will you be back? I have so many irritating Life questions to ask you! :P
16:52:31 <alise> AnMaster: Is it totally straight and downwards? How long is it?
16:52:47 <AnMaster> alise, hm do golly allow you to measure that?
16:52:49 <alise> Phantom_Hoover: /Will/ you be back?
16:52:53 <alise> AnMaster: Just eyeball it.
16:53:02 <alise> Phantom_Hoover: Today? IN FIVE YEARS?
16:53:20 <AnMaster> alise, population at generation 0 is 404
16:53:31 <AnMaster> and I draw maybe 50 cells then copy-pasted
16:53:38 <Sgeo> Hm, how far in advance can "Bye in advance" be said?
16:53:46 <Sgeo> Could I say bye in advance to alise right now?
16:54:43 -!- poiuy_qwert has joined.
16:54:46 <uorygl> Bye in advance, everyone.
16:54:54 <AnMaster> alise, how many generations does it take to see sierpinski, I let it run until it became stable mostly
16:54:54 <alise> AnMaster: give rle
16:54:55 <uorygl> I say this not because I plan to leave, but because I'm counting on you leaving.
16:55:00 <alise> AnMaster: like a few
16:55:38 <AnMaster> alise, it went stable around 30 000 generations (eye balling from fast speed) not a lot of sierpinski patterns there
16:56:05 <alise> you did something wrong
16:56:16 <AnMaster> wait not all still lifes, was doing acceleration still
16:57:02 <AnMaster> alise, is RLE text only? pastebin would be easiest
16:57:05 <alise> Here is a life-like rule for you all: B38/S23
16:58:02 <AnMaster> alise, hm wait there is a sierpinski-like patterns early on but it goes away quickly
16:58:10 <Ilari> I guess the maximum length is about 330 800. At 17c/45, it takes about 876 000 generations to travel its own length. Quite huge perioid for gun...
16:58:14 <alise> AnMaster: The Sierpinski is there, you just have not noticed it
16:58:20 <alise> Make a bigger line, then zoom out
16:58:27 * Sgeo wants to play Gish on a computer that can actually handle it
16:58:38 <AnMaster> alise, you know, it is very tedious to draw a straight line in golly
16:59:12 <Phantom_Hoover> You can get RLEs in Golly with select/copy, and I must go now.
16:59:23 <AnMaster> yes but that was tricky to align
16:59:29 <Ilari> Worse, that would mean constructing glider gun with perioid of about 900 000 for caterpillar gun.
16:59:33 <AnMaster> alise, check that line after 5000 generations
17:01:01 <Sgeo> Bye Phantom_Hoover
17:02:10 <AnMaster> alise, okay I can see it early on while it is still growing now
17:02:50 <AnMaster> a bit harder to see it once it went stable... the pattern is "fuzzier" then
17:04:09 -!- kar8nga has joined.
17:04:20 <alise> bye Phantom_Hoover
17:04:34 <alise> AnMaster: i think a slightly non-straight line (moves to the right one pixel at some point) gives the best results
17:05:01 <AnMaster> alise, is it possible to get one where it isn't cut in the middle of a given "level" if you see what I mean.
17:06:16 <Sgeo> http://youtube.com/watch?v=fJEpxPWJAh8
17:06:28 <AnMaster> alise, if you look at it when it went stable (initial length of line is now 3035) you can see the middle looks like a extremely wide and squashed hour glass, right?
17:06:51 <alise> "Random fill percentage: 50"
17:06:53 <alise> How can I do a random fill?
17:07:03 <alise> AnMaster: I didn't look at it for too long
17:07:17 <AnMaster> alise, like the two triangles meet in the middle of a "level" of the outermost triangles
17:12:05 <alise> How fast can a spacefiller fill a patterned background, I wonder?
17:12:09 <alise> As in, not already filled.
17:15:48 * alise attempts to make a B/S rule based on silly idioms and truisms.
17:15:56 <alise> Well, two's company and three's a crowd; so obviously S2 and not S3.
17:16:14 <alise> You need a loving, stable family with a mother and father to survive, obviously, so you can't be born with 0 or 1, has to be 2.
17:16:29 <alise> Umm, you need a nice community around you, which can be of size seven or eight, so S78.
17:16:39 <alise> And you can also be born by massive orgies between the community, so B78.
17:16:48 <alise> Funny. It's chaotic.
17:16:50 <AnMaster> alise, wrt spacefiller: golly contains some example
17:17:03 <alise> Yes, but not with a pattern.
17:17:06 <AnMaster> notes mention theoretical upper bound
17:17:19 <Sgeo> half of space?
17:17:19 <AnMaster> (for the all-cardinal-directions one at least)
17:17:39 <Sgeo> Or was it a speed bound? I think I remember seeing half of space as a bound
17:17:44 <AnMaster> (iirc it can be faster if only growing in left/right _or_ up/down but not both
17:18:03 <alise> AnMaster: c/4 all directions, c/2 half
17:18:06 <alise> but that doesn't answer my question
17:18:09 <alise> I asked about patterned backgrounds
17:18:27 <Sgeo> What if it grows diagonally?
17:19:05 <AnMaster> alise, what do you mean c/4 all c/2 half?
17:19:13 <AnMaster> #C Spacefillers are the fastest-growing known pattern in Conway's
17:19:13 <AnMaster> #C Game of Life (probably the fastest possible). They fill space
17:19:13 <AnMaster> #C to a density of 1/2, conjectured to be the maximum density,
17:19:13 <AnMaster> #C and they do it at a speed of c/2 in each of the 4 directions,
17:19:13 <AnMaster> #C which has been proven to be the maximum possible speed.
17:19:20 <AnMaster> is what the info says for that
17:19:22 <alise> Hmm, I thought it was c/4.
17:19:29 <AnMaster> alise, and what do you mean patterned bg exactly?
17:19:37 <alise> Normally, the background is a pattern of all dead cells.
17:19:41 <Sgeo> Oh, "conjectured"
17:19:43 <alise> I'm sure you can figure out what I mean.
17:20:07 <AnMaster> alise, err? no it isn't. That space filler draws dead/alive/dead 1-width strips
17:20:07 <alise> Sgeo: It's very likely to be true, I think.
17:20:11 <Sgeo> If you play Night and Day, I think it is, you can say that they're all live
17:20:15 <alise> Trust it as you trust the Riemann hypothesis or the Collatz conjecture.
17:20:21 <alise> Yeah, Day and Night.
17:20:29 <alise> AnMaster: Stop talking until you understand the very plain meaning of what I said.
17:20:31 <AnMaster> alise, it is however a still life it draws. All dead cells doesn't really fill space, it is empty
17:20:42 <AnMaster> alise, oh you mean prior existing bg?
17:20:54 <Sgeo> What program are you using?
17:21:18 * Sgeo had Mirek's Celebration once
17:21:30 <AnMaster> alise, I suspect it depends on the pattern. Plus you would have to fit the filler to the specific pattern probably so they didn't destroy each other
17:21:56 <Sgeo> *Mirek's Cellebration
17:22:20 <alise> AnMaster: Of course.
17:22:24 <alise> Phantom_Hoover: You are back?
17:22:44 <Phantom_Hoover> Sgeo: Cellebration is good, particularly w/r/t Margolus rules
17:22:45 <alise> Control-5 doesn't work here
17:23:12 * Sgeo doesn't even know what a Margolus rule is
17:23:21 <Sgeo> Also, was surprised to see Golly has this: http://golly.sourceforge.net/win-loops.png
17:23:44 <alise> Phantom_Hoover: ah, no
17:23:48 <alise> Phantom_Hoover: can't I fill infinite space???
17:23:52 <Phantom_Hoover> It only added multi-state CAs a couple of versions ago
17:23:54 <Sgeo> It's something I saw in Cellebration, and for some reason, thought it unique to Cellebration
17:24:00 <alise> Phantom_Hoover: Why not?
17:24:04 <alise> It's perfectly mathematically consistent!
17:24:27 -!- oerjan has joined.
17:24:39 * Sgeo plops Phantom_Hoover and alise on a torus
17:24:53 <alise> Phantom_Hoover: No!
17:24:58 <alise> Just compute the random numbers on the fly!
17:25:31 <alise> And, where is ctrl-5 documented?
17:26:19 <alise> Why are so many B/S rules similar to Life?
17:26:28 <alise> Phantom_Hoover: Oh, I didn't even notice that
17:27:07 * Sgeo tries in vain to find out what CyTime it is
17:27:24 <alise> Quadrescence: I don't know that we have any Germans or German speakers.
17:27:35 <Sgeo> Look, alise! A shitty ancient 3d thing that I'm not paying money for because I'm not THAT nostalgic.. ok, so I am nostalgic
17:27:50 <Quadrescence> alise: I need a book translated and I'm trying to get as many people to help out as possible
17:28:15 <alise> Quadrescence: "Too many cooks" instantly appeared in my head.
17:28:21 <alise> Having many, many people translate one book?
17:28:41 <Quadrescence> alise: Yes, one person on the internet won't translate 500 pages for me
17:28:55 <Quadrescence> Well maybe there is one who would, but for the most part, they won't
17:29:14 <alise> Quadrescence: But the result will probably not be such good quality; mixed styles and such.
17:29:24 <alise> It's like getting many people to typeset one book.
17:30:02 <Quadrescence> I just can't afford to hire a single translator to do all of the work unfortunately.
17:30:10 <Phantom_Hoover> http://wiki-translation.com/tiki-index.php might be of help.
17:30:21 <Quadrescence> I am not doing this solely for myself; in fact, someone else brought it up
17:30:54 <Sgeo> "Abuse of this function will get you reported to Security"
17:31:01 <Sgeo> Hopefully asking what time it is isn't "abuse"
17:32:21 <alise> Quadrescence: What book?
17:32:45 <Quadrescence> A book on continued fractions: die lehre von den kettenbrüchen von dr. oskar perron
17:33:37 <alise> I like continued fractions. A whole book on them seems excessive though :)
17:34:28 <alise> You're pretty deep and mysterious. Or something
17:35:21 <alise> Phantom_Hoover: I take it it's an open problem as to what pattern is the most "useful" for things like lightspeed travel?
17:35:28 <alise> Like, the easiest for things to survive in and what not.
17:35:35 <alise> The most lubricant fabric of space.
17:35:59 <AnMaster> <alise> Just compute the random numbers on the fly! <-- that works for 1 generation
17:36:12 <alise> AnMaster: you store them after computing
17:36:16 <Phantom_Hoover> For lightspeed travel, the stripy agar is probably best.
17:36:25 <alise> Any page about the stripy agar?
17:36:32 <AnMaster> alise, yeah but you need to consider all of space for each generation
17:36:44 <AnMaster> meaning you need infinite space after 1 generation already
17:36:58 <alise> AnMaster: hmm... but truly random cells in life will produce truly random results, right?
17:37:04 <alise> Phantom_Hoover: I did that, but it exploded.
17:37:14 <Sgeo> Ooh, lightspeed travel?
17:37:24 <alise> Sgeo: Only across a pre-prepared space.
17:38:02 <alise> Sgeo: There's also the lightspeed telegraph, which is a machine with two very slow and inefficient machines at each end, and a very long strip of simple wire connecting them; one machine sends blips along this wire; they then travel at lightspeed, and the other machine absorbs them.
17:38:20 <alise> The receiving endpoint is very inefficient, though, and the blips carried are very chaotic; so it's not practical in its current form.
17:38:36 <Sgeo> Chaotic howso?
17:38:42 <alise> Not easily detected.
17:38:51 <alise> But if you need some particles to travel across a very, very long distance, then the lightspeed telegraph will be helpful.
17:38:57 <Sgeo> And surely, no matter the inefficiency, it's practical given a long enough space...
17:38:58 <alise> We're talking LONG here.
17:39:46 <Phantom_Hoover> On the scale of the example that comes with Golly, sending c/2 spaceships would be hugely faster.
17:40:09 <alise> You'd need a cord of several million beehives long for it to be faster than c/2 spaceships, wouldn't you?
17:40:17 <alise> Since the spaceship-emitter at the receiving end is just so slow.
17:40:22 <alise> Maybe thousands rather than millions.
17:40:34 <Sgeo> Since when is a beehive a unit of length?
17:40:44 -!- pikhq has quit (Read error: Connection reset by peer).
17:40:51 <alise> http://www.conwaylife.com/wiki/index.php?title=Beehive
17:40:57 <alise> Phantom_Hoover: Say, why aren't blocks useful for the wire?
17:41:14 -!- lament has quit (Ping timeout: 264 seconds).
17:41:23 <Phantom_Hoover> You need a reaction that burns along a track and leaves a copy of the track behind.
17:41:24 <oerjan> alise: iiuc basically lightspeed communication isn't really giving you anything if the machinery is so complicated you have to expand the machine by a factor >= 2 to compensate
17:41:47 <oerjan> because then why not use the smaller machine with c/2 stuff
17:41:56 <AnMaster> Phantom_Hoover, btw that "FTL tunnel" illusion based on building the back of an already constructed front... couldn't it still be used for communication?
17:41:57 <alise> oerjan: well actually the two ends are constant-size in the lightspeed telegraph
17:42:06 <AnMaster> Phantom_Hoover, if all you want is a one-shot signal or such
17:42:08 <alise> and no matter what length the wire they emit the received stuff at the same (slow, slow) pace
17:42:14 <alise> so if you have a really, really long wire, it'd be faster
17:42:27 <alise> Phantom_Hoover: is there not a spaceship that survives blocks and poops one out?
17:42:34 <oerjan> alise: yes but if you used it for memory, you'd have to have many receivers wouldn't you
17:42:38 <Sgeo> I guess it would be impossible to have the complexity of the machinery depend on the length of the wire
17:42:40 <alise> AnMaster: The lightspeed telegraph is usable for communication.
17:42:43 <alise> Given a sufficiently long wire.
17:42:56 <alise> oerjan: hmm... not neccessarily, it'd just be slower
17:43:08 <alise> but if you had a, say, several million beehives long one...
17:43:13 <alise> oerjan: then it'd be faster
17:43:53 <Phantom_Hoover> If you had lots of LSTs next to each other you might be able to squeeze a decent bitrate.
17:43:56 <Sgeo> AnMaster, presumably, by the time you could correctly detect whether or not a signal was sent through, the average speed dropped to c or lower
17:44:09 <AnMaster> Phantom_Hoover, didn't it detect just the front iirc?
17:44:17 <AnMaster> and assembled the back of a pre-assembled front?
17:44:24 <alise> Phantom_Hoover: Do you think, in your expert (:P) opinion, that practical lightspeed communication will ever be feasible in GOL?
17:44:40 <AnMaster> Phantom_Hoover, and it ends up moving the head somewhat ahead of c?
17:45:00 <alise> AnMaster: No; it's an illusion.
17:45:06 <Sgeo> AnMaster, and it ends up destroying the pre-assembled head if there is no ship
17:45:18 <alise> An example is the "Star Gate", an arrangement of three converging gliders that will mutually annihilate on collision. If a lightweight spaceship (LWSS) hits the colliding gliders, it will appear to move forwards by 11 cells in only 6 generations, and thus travel faster than light.[3] This illusion happens due to the fact that the glider annihilation reaction proceeds by the creation and soon-after destruction of another LWSS. When the incoming LWSS hits th
17:45:19 <alise> e colliding gliders, it isn't transported, but instead modifies the reaction so that the newly created LWSS can survive. The only signal being transmitted is that determining whether the outgoing LWSS should survive or not. This does not need to reach its destination until after the LWSS has been "transported", and so no information needs to travel faster than light.
17:45:24 <oerjan> Phantom_Hoover: hm i guess for a RAM the actual bit rate needed is just logarithmic in the size...
17:46:02 <Phantom_Hoover> The LST couldn't be used for memory storage; you'd use it for long-distance communication.
17:46:13 <AnMaster> alise, so that hit must be timed?
17:46:27 <Phantom_Hoover> All of the known signal circuitry uses spaceships and gliders.
17:46:31 <alise> In my opinion, the most awesomest thing would be a Turing machine tape where the head is mounted on a surface that allows it to move at light-speed, and that also has a lightspeed communication line to the CPU; then, it could receive instructions from the CPU at lightspeed, modify the memory, read it, then send the read value back at lightspeed.
17:46:36 <oerjan> Phantom_Hoover: well i'm thinking of when the memory is so huge you need long-distance communication to interact with it
17:46:40 <alise> It would be the ultimate Turing machine tape in Life.
17:46:54 <AnMaster> alise, couldn't you have a clocked signal then? with some guns or whatever they are called
17:46:58 <alise> Except for that the ULTIMATE ultimate one would be where the whole tape moves so that reading and modifying the current cell is always near-instant.
17:47:09 <AnMaster> alise, and make sure to send the signal synced
17:47:13 <Sgeo> "spaceships and gliders" sounds redundant
17:47:27 <AnMaster> alise, what happens if the LWSS is not there?
17:47:28 <Phantom_Hoover> alise: You'd need a receiver that could be destroyed and reconstructed at c.
17:47:45 <Sgeo> AnMaster, then a partial LWSS is made, but then dies
17:48:04 <alise> Phantom_Hoover: Or you could embed the memory-instruction-processor into every cell of the tape.
17:48:10 <alise> Phantom_Hoover: So that when you extend the tape it copies itself slowly.
17:48:20 <AnMaster> alise, it can be used to forward with a clocked signal
17:48:31 <AnMaster> since the partial LWSS will go away
17:48:35 <alise> Phantom_Hoover: Then all the communication line needs to do is, when it gets a lightspeed communication to the tape, move it into the processor; when it receives one from the tape, send it on to the CPU.
17:48:42 <AnMaster> just tested that by not sending a LWSS there
17:48:47 <alise> AnMaster: you can't do FTL communication
17:48:50 <alise> theoretical impossibility
17:49:04 <Phantom_Hoover> alise: Making a c signal stop at a set distance might be impossible.
17:49:08 <alise> Please don't say "quantum entanglement".
17:49:10 <alise> I will have to kill you.
17:49:14 <AnMaster> alise, but in GOL you can if you can have a clocked signal
17:49:14 <Sgeo> AnMaster, try reviewing the rules of GoL
17:49:21 <alise> Oh, impossible in reality
17:49:29 <AnMaster> Sgeo, I do see the issue but... hm
17:49:34 <alise> Phantom_Hoover: AnMaster believes he has come up with a way to do FTL communication in GOL; laugh at him to make him shut up or something
17:50:25 <Sgeo> AnMaster, by the time you can actually use something to detect whether the LWSS is there or not, the signal has no longer seemed to have taken lightspeed, I think
17:50:29 <Phantom_Hoover> AnMaster: It's patently impossible. The axioms only allow propagation of information at 1 cell/generation. END OF.
17:50:31 <Sgeo> *faster than lightspeed
17:50:33 <alise> Sgeo is pretty much correct
17:50:36 <AnMaster> Phantom_Hoover, open Sginal-Circuitry/stargate.rle in golly examples. Then remove the middle LWSS in the upper line. Check the results after running it a few generations
17:50:49 <AnMaster> (enough to let all go to the other side)
17:51:04 <AnMaster> Phantom_Hoover, now, tell me how that does not seem to be FTL
17:51:13 <Sgeo> AnMaster, seem to be is the operative phrase
17:51:34 <AnMaster> Sgeo, yes but in which was is it not then
17:51:55 <AnMaster> Phantom_Hoover, hm? what do you mean by that
17:52:17 <Sgeo> AnMaster, Has anyone really been far as decided to use even go want to do look more like?
17:52:22 <alise> But anything in THE UNIVERSE
17:52:24 <alise> Trying to PROCESS THIS OUTPUT
17:52:28 <alise> Could NOT see it all as FTL
17:52:34 <alise> We are God, we are omnipotent, we can observe everything at once
17:52:36 <alise> Cells in the universe can not
17:53:04 <AnMaster> Sgeo, eh? I can't parse that grammar
17:53:13 <Sgeo> AnMaster, neither could I process yours
17:53:29 <Sgeo> <AnMaster> Sgeo, yes but in which was is it not then
17:53:32 -!- Gregor-L has joined.
17:54:14 <Sgeo> It's not, because the signal to kill the new LWSS or not propagates at c or less
17:54:31 <Sgeo> If you try to detect the ship before that signal propagates, it will always be there
17:54:36 <Phantom_Hoover> alise: Did you get the thing about it being impossible for something at C to stop?
17:54:37 <alise> We're GOL hobby physicists making a cruel mockery out of this crackpot's ideas! Mwahahaha!
17:54:47 <alise> THIS MUST BE WHAT IT'S LIKE TO FEEL TRULY INTELLIGENT
17:54:52 <alise> Phantom_Hoover: Wow, no.
17:54:54 <Sgeo> If you wait until the ship is either destroyed or not destroyed, then the signal travelled at c or less
17:54:55 <alise> Phantom_Hoover: That is cool. Is it true?
17:55:04 <alise> Phantom_Hoover: What if you put an amazing wall of blockingness in its path? I guess that doesn't count.
17:55:10 <AnMaster> alise, I'm not saying I am right. I just want a good explanation of why I'm not
17:55:18 <alise> Phantom_Hoover: I guess the tick where it decided whether to destroy itself would make it c/2.
17:55:19 -!- Gregor-L has changed nick to Gregor.
17:55:31 <jix> Quadrescence: yes
17:55:33 <Sgeo> AnMaster, there's a pre-made ship front. Trying to detect it to get a signal faster than light will always result in a detection, whether or not there was a signal
17:56:07 <Sgeo> Wait, things at c can't stop without a wall?
17:56:28 <Phantom_Hoover> The front of the signal can't be accessed by things behind it
17:57:38 <Sgeo> How to propagate a signal in Life faster than c:
17:57:41 <alise> Phantom_Hoover: But can you have "stargates" at various points, which absorb the c-ship, then decide whether to fire another one on or not?
17:57:46 <Sgeo> Make a GoL simulation in GoL
17:57:52 <alise> Sure, it slows it down from c for a bit, but it'd allow conditional destruction every now and then.
17:58:02 <alise> Wow, this is exactly like science-fiction space engineering.
17:58:09 <Sgeo> Corrupt one of the cells, so it sends a signal to a cell that's not a neighbor
17:58:42 <alise> Phantom_Hoover: But still a lot better than c/2, assuming your travel distance is long enough.
17:58:43 <alise> No, not each cell.
17:58:59 <alise> For the Turing machine.
17:59:10 <alise> Hmm... and each cell would be right next to each other, almost.
17:59:15 <alise> So the c wire wouldn't get you much there, would it?
17:59:23 <alise> Or, well, you do need a c wire, but only for the return trip.
17:59:26 <alise> When accessing memory.
17:59:32 <alise> So accessing memory would be a lot faster than writing it.
18:00:00 <Phantom_Hoover> So now we need to work out how to create a c wire which can have signals injected at regular intervals.
18:00:02 <alise> Phantom_Hoover: If mounted on an appropriate surface, how fast could a whole tape move itself so that one tape-cell is at the (non-moving) head?
18:00:05 <alise> Presumably not at lightspeed.
18:01:03 <alise> Phantom_Hoover: How could reading/writing be at lightspeed, if it has to stop at every cell?
18:01:09 <alise> The read response can go at lightspeed, but the read request can't.
18:01:21 <alise> And the write has no return journey.
18:03:15 <alise> How would it know when to intercept?
18:03:19 <alise> It doesn't know if it's The One.
18:03:27 <alise> And if it collided and then decided, it'd be significantly sub-c.
18:03:42 <alise> But that's the whole problem; telling things to the cell!
18:03:47 <alise> That requires communication that stops!
18:03:47 <AnMaster> Phantom_Hoover, oh right that is the issue. That LWSS travels at c/2... Why didn't you just say that
18:03:52 <alise> That's what we're trying to achieve :P
18:03:54 <AnMaster> THAT would have been the clear answer
18:03:58 <alise> AnMaster: Because that's fucking obvious???
18:04:17 <AnMaster> alise, depends on how well you know life.
18:04:39 <alise> If it travelled at c do you think we'd be saying all this stuff about complicated light-speed telegraphs?
18:04:46 <Phantom_Hoover> alise: Well, telling the cell it's The One will be much less data than read/writes.
18:05:11 <AnMaster> alise, it could have been c/2 < speed < c for all I knew
18:05:31 <AnMaster> in theory for all I knew enough to make the increase worth it..
18:05:50 <alise> Phantom_Hoover: Reading: "Cell, you are the one." Cell then poops off the data at lightspeed to the CPU.
18:06:17 <alise> Phantom_Hoover: Writing: "Cell, you are the one. (1 bit that only the one will examine)" Cell then poops the data onto the tape.
18:06:34 <Phantom_Hoover> Writing: "Cell, you are the Write One", cell taps into lightwire and reads any signal on it.
18:06:38 <alise> You can't tell the cell it's the one any faster than that, and certainly not at lightspeed because it can't stop, and if you stopped it it's <c.
18:06:47 <AnMaster> Phantom_Hoover, what are those light speed bubbles? It doesn't seem mentioned on the life wiki. saw it at some other link posted here before
18:06:53 <alise> Phantom_Hoover: Err, why?
18:06:58 <alise> That wouldn't be significantly faster.
18:07:06 <alise> Phantom_Hoover: Tape cells only store 1 bit of information.
18:07:10 <alise> AnMaster: Bubbles?
18:07:29 <AnMaster> some page about light speed stuff
18:08:14 <alise> Phantom_Hoover: I mean, sure, it'd be /one/ less thing for the cell's intercept-and-interrogate handlers to copy out.
18:08:17 <alise> But that's, like, a few cells.
18:08:17 * Sgeo loves pathological edge effects
18:08:42 <alise> Phantom_Hoover: Reading isn't quite at c, since you still need interception. It's <c for the part where the "You, gimme your value" signal travels down the wire; then =c when the read signal travels back.
18:09:15 <alise> Probably the "oi, it's you; (and if this is a write here is a bit)" should be done with c/2 spaceships, because intercepting and re-pooping (technical term) a c blip every single cell would end up really slow.
18:10:34 <Phantom_Hoover> Although tapping into the lightwire to the CPU would likely have a significant overhead.
18:11:04 <Phantom_Hoover> Since lightwires are by their nature very easy to burn.
18:11:05 <AnMaster> alise, oh not GOL cell but memory cell (made up of several GOL cells)?
18:11:40 <AnMaster> alise, why not call it memory block such
18:12:05 <alise> Phantom_Hoover: Yes, we're assuming that we've come up with a relatively efficient lightspeed receiver.
18:12:16 <Sgeo> Why is the Java thing on the Wiki not working?
18:12:22 <AnMaster> anyway. why not put it into several banks. Sure it wouldn't be exactly a Turing machine. But it might be more practical
18:12:49 <Phantom_Hoover> The sender needs to be able to send to a lightwire at any point on its length.
18:13:04 <AnMaster> plus if you could branch the signal you could have it check several banks in parallel
18:13:35 <alise> Sgeo: Works for me
18:13:43 <alise> AnMaster: No faster
18:13:44 <AnMaster> and sure, if one of them didn't return, wouldn't matter would it? The signal would die when it hit the (current) end of the tape
18:13:50 <Sgeo> Hm, I think it's because I clicked Next
18:13:51 <alise> Phantom_Hoover: ah, i see
18:13:55 <Sgeo> That broke something somehow
18:13:59 <AnMaster> Phantom_Hoover, okay do it before it transmits? thus the multiple banks
18:14:00 <alise> Phantom_Hoover: so we need a lightwire that can accept communications at any part
18:14:04 <alise> doesn't the telegraph have one of them?
18:14:07 <alise> it's uniform all the way along
18:14:10 <alise> so given enough space...
18:14:28 <AnMaster> Phantom_Hoover, if you have 1024 places to check you might now have 2 parallel checks of 512 places instead
18:14:47 <AnMaster> seems like a probable win to me even if branching has a certain overhead
18:15:01 <AnMaster> as long as the overhead is smaller than what you gain
18:15:02 <Phantom_Hoover> You'd need a way to break the wire without burning it, then transmitting a signal, then fixing the wire perfectly.
18:15:21 <AnMaster> Phantom_Hoover, I suggested branching *before* the transmitter at the CPU
18:15:32 <AnMaster> as in, using multiple transmitters
18:15:47 <Phantom_Hoover> It'll need to be synchronised, which is another issue with this design.
18:16:11 <AnMaster> Phantom_Hoover, local clocks synced at the start? It isn't as if life has any issues with clock drift :P
18:16:36 <alise> Phantom_Hoover: It'd be nice if there was an agar that allowed something to pass through such that the bit around the agar's edge would easily transform it into a spaceship of some sort
18:16:50 <alise> It's nice that Life has no clock drift or anything :P
18:17:02 <alise> And no nondeterminism
18:17:03 <AnMaster> Phantom_Hoover, hm, make a pentium4 style pipeline? ;P
18:17:15 <alise> I wonder what would happen if I wrote a Life simulator with entanglement.
18:17:21 <AnMaster> Phantom_Hoover, nah, depends on what you mean. Sure some latency will happen but in which part did you mean here
18:17:31 <alise> That is, you can mark certain regions of space as entangled and pair them with another region.
18:17:32 <Sgeo> How does Maze work?
18:17:36 <alise> Then any change on one side happens to the other.
18:17:40 <AnMaster> Phantom_Hoover, well it was a bad joke
18:17:48 <alise> This also means that a glider could innocently walk into the entangled region, then when it walks out it'll be TWO
18:17:55 <alise> one in the original location, one in the other entangled region
18:17:57 <alise> Phantom_Hoover: COOL OR WAT.
18:18:00 <AnMaster> Phantom_Hoover, anyway, where is the latency in this case. In the sender?
18:18:34 <alise> Phantom_Hoover: COOL OR WAT.
18:18:43 <alise> Of course you would
18:18:43 <Sgeo> Can patterns cause arbitrary regions to be entangled?
18:18:50 <alise> It wouldn't just be entanglement, it would be AWESOME entanglement.
18:18:50 <Sgeo> That would be cooler
18:18:55 -!- pikhq has joined.
18:19:01 <alise> Basically, a fixed region of space is entangled (by act of God).
18:19:15 <alise> Then if a glider moved in, it would cause cells in the entangled region to change, and thus the glider would appear weirdly in the other one.
18:19:23 <alise> first only the first part of the glider that hits the entangled region would appear
18:19:29 <alise> and then quickly die out because it wouldn't have enough substance
18:19:31 <alise> which would mangle the glider
18:19:38 <alise> Phantom_Hoover: ^^ see that
18:19:42 <AnMaster> Phantom_Hoover, well sure, "lower" addresses would be faster
18:19:47 <alise> Phantom_Hoover: can you think of any way to make something go through the entangled region and not be obliterated?
18:20:09 <alise> Phantom_Hoover: maybe have some sort of spaceship such that part of it is still life, such that when it enters, only the bottom row is there, but it connects with the previously-still top of it
18:20:13 <alise> and starts being a spaceship?
18:20:22 <AnMaster> Phantom_Hoover, but that is where the banks come into it. Basically you check multiple cells at the same time that way
18:20:22 <Phantom_Hoover> Change the transition rules, as cells in the entangled area effectively have 16 neighbours?
18:20:24 <alise> that would then obliterate the one that entered, but the one that comes out would work
18:20:31 <alise> Phantom_Hoover: nope, keep them the same
18:20:37 <alise> Phantom_Hoover: they're treated as different cells
18:20:39 <alise> it's just that you add one more rule
18:20:49 <alise> "whenever a cell in an entangled region changes, both entangled regions change accordingly"
18:20:58 <alise> only when it changes
18:21:01 <alise> "stay the same" would not conflict
18:21:02 <AnMaster> Phantom_Hoover, if you are doing <c addressing and that allows branching a signal easily then you should do it like a tree or such to have just a few levels and be able to check a lot of memory cells at once
18:21:06 <alise> and also, life takes precedence over death i think
18:21:22 <AnMaster> balanced binary tree sounds terribly complicated in life
18:21:49 <alise> Phantom_Hoover: You know that dot matrix printer?
18:21:59 <alise> http://tlrobinson.net/blog/2009/02/07/game-of-life-generator/
18:22:05 <alise> Using the same thing as the golly generator
18:22:14 <alise> Phantom_Hoover: I bet you could make one of them that takes dynamic input at simulation-time.
18:22:17 <alise> Then it'd be a generic output device.
18:23:12 <alise> Phantom_Hoover: Although...
18:23:20 <alise> Phantom_Hoover: With a universal constructor, you could just construct still lifes to do the same thing.
18:23:23 <alise> But a UC would be a lot slower, right?
18:23:38 -!- capitancar has joined.
18:23:41 -!- sebbu has joined.
18:23:48 <alise> Phantom_Hoover: Ooh, what about a Life unit cell with a UC? The life can hit a special region to control the UC >:)
18:24:43 <alise> universal constructor
18:24:44 <Phantom_Hoover> Again, you just need to emulate the stored instructions.
18:25:08 <alise> <capitancar> que mas
18:25:16 <alise> <capitancar> comentame
18:25:19 <alise> capitancar: english please
18:25:28 <Sgeo> "It is probably the most well-known one cell thick pattern"
18:25:47 <Sgeo> http://www.conwaylife.com/wiki/index.php?title=One_cell_thick_pattern
18:25:55 <Sgeo> I think blinkers might be a bit more well known
18:26:33 <pikhq> Well, for once my Internet makes sense.
18:26:44 <AnMaster> Sgeo, I didn't know of that pattern. I knew of blinkers though
18:26:56 <pikhq> It drops out because of heavy rain.
18:26:57 <AnMaster> )(of course that is just one data point)
18:27:04 <pikhq> Y'know, what you'd *expect*.
18:27:47 <alise> pikhq: your packets go into space man
18:27:51 <alise> space is one cold motherfucker.
18:27:54 <AnMaster> pikhq, well for any sort of wireless communications it makes sense that weather have an effect
18:28:14 <AnMaster> alise, that is what it's mum said!
18:28:24 <alise> AnMaster: I see :P
18:29:21 <alise> *its, also, *mom or mother for the joke
18:29:22 <Phantom_Hoover> "That's the end of that one. Atmospheric conditions in outer space often interfere with transmitting. "
18:30:09 <alise> All that solar wind has an affect.
18:30:33 <Sgeo> Hm, trying to remember how I found out about GoL
18:30:44 <Sgeo> I remember some program, that had a file 1103
18:30:54 <Sgeo> Didn't give it a name or anything, so I called it 1103
18:31:04 <Sgeo> (the R-pentomino)
18:31:09 <alise> Sgeo: Do you really think you need /more/ nostalgia?
18:31:13 <Phantom_Hoover> I found out about it when I found a rather crappy GoL simulator for the Mac Dashboard.
18:31:25 <pikhq> AnMaster: Like I said, *it makes sense* for once.
18:31:34 <alise> Phantom_Hoover: KIDDO
18:31:36 <AnMaster> pikhq, how often does it rain there?
18:31:42 <pikhq> AnMaster: Too often.
18:31:47 <Sgeo> I think space was green
18:31:47 <Phantom_Hoover> I then forgot about it until I read about it again in one of Ian Stewart's books.
18:31:53 <alise> AnMaster: every time Obama cries
18:31:53 <AnMaster> pikhq, ah then it is scared of the rare clear weather
18:31:56 <pikhq> It's been raining for the past week.
18:31:58 <alise> cries. for freedom.
18:32:00 <AnMaster> pikhq, doesn't know how to handle it
18:32:28 <pikhq> I feel like I'm British or something.
18:35:29 <alise> pikhq: Tell me! Have you ever experienced DRIZZLE?
18:35:50 -!- FredrIQ has joined.
18:36:04 <alise> Answer, pikhq! Answer, you foul-mouthed cretin!
18:36:18 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
18:36:59 <alise> pikhq: HAVE. YOU. EVER. EXPE-- oh, hello FredrIQ, sorry for your (first?) impressions of this channel, I'm just chastising silly pikhq here-- RIENCED. DRIZZLE?!
18:37:40 <alise> Oh my god, qotile!
18:37:43 <Sgeo> "Drizzle" reminds me of CT
18:37:54 <alise> Phantom_Hoover: "Time is equated to the Z-axis"
18:37:55 -!- poiuy_qwert has joined.
18:37:59 <AnMaster> alise, you know what, you should be able to do?
18:38:16 <AnMaster> alise, copy a pattern from golly then paste it into the search box on lifewiki
18:38:21 <alise> pikhq: Are you SURE? Drizzle is not something you know to be drizzle, because its very virtue is its uncanny ability to make you feel rotten and get things wet without making an impression on anything.
18:38:23 <AnMaster> to search for the name of some specific pattern
18:38:28 <alise> It is not rain, it is worse than rain! It is of the devil and you never notice it!
18:38:40 <alise> It is not malicious, it is merely so passively, mediocrely HORRIBLE that it destroys your soul!
18:38:46 <alise> HAVE YOU _EVER_ EXPERIENCED DRIZZLE?
18:39:09 <Sgeo> http://www.qotile.net/blog/images/image5.jpg that looked physically real at first
18:39:09 <pikhq> alise: All I know is, I have certainly gone out and came back in and noticed that I was both wet and feeling awful.
18:39:16 <pikhq> Without noticing the wettening itself.
18:39:23 <alise> Wet, you see, wet, drizzle doesn't make you wet, it makes you drizzledupon.
18:39:47 <Sgeo> I thought that maybe it was 3d printed\
18:39:56 <alise> You have not experienced drizzle. You are scum to call yourself British! Never be ungrateful for your weather, for we have the powers contained within our atmosphere to crush your soul should one ever decide to introduce you to the real British climate!
18:39:58 <pikhq> I also like-a the tea.
18:40:05 <alise> REMEMBER THIS, HEATHEN!
18:40:53 <pikhq> alise: Is YOUR last name the name of a shire?
18:41:14 <alise> Elliott "Hird" Oxfordshire
18:41:39 <pikhq> Dammit, you got the good shire.
18:41:55 <alise> I don't actually live in Oxfordshire but it would be cool if I did
18:42:15 * pikhq is inexplicably named "Worcester".
18:42:19 <AnMaster> Phantom_Hoover, what do you call a thing that looks like a LWSS but is one longer in the "longer" side? Thus making the gap in the opposite side 3 cells instead of 2. In the middle, and one up, in that gap there is an extra cell
18:42:29 <AnMaster> can't think of a better way to describe it
18:42:42 <AnMaster> Phantom_Hoover, yep looks like it
18:42:50 <alise> pikhq: Worcester, pah. It's Worcestershire.
18:42:56 <AnMaster> Phantom_Hoover, there should be some pattern→name lookup/search
18:42:59 * Sgeo his AnMaster with an OWSS
18:43:03 <alise> Also, anyone who calls Worcestershire Sauce "Worcester sauce" is living in a state of Sin.
18:43:21 -!- capitancar has left (?).
18:43:32 <pikhq> alise: True, true.
18:43:33 <Sgeo> OverWeight Space Ship
18:43:37 <alise> pikhq: Also note, I'm not actually called Oxfordshire
18:43:37 <pikhq> There's more to the shire than Worcester.
18:44:16 -!- marchdown has joined.
18:44:32 <alise> Totally a real shire
18:44:33 <Sgeo> alise, are you a HURD of HIRDS?
18:44:45 <alise> Sgeo: Go away, Stallmanshire.
18:45:01 <alise> I hereby decree that all surnames henceforth spoken in this channel must be suffixed, if they are not already, with "-shire".
18:45:03 <pikhq> alise: Go away, Shireshire.
18:45:06 <Sgeo> erm,yeah, hird not herd
18:45:21 -!- Phantom_Hoover has changed nick to Phantom_Hooversh.
18:45:25 <alise> This is an ex cathedra declaration that calls under Papal infallability.
18:45:32 -!- Sgeo has changed nick to Sgeoshire.
18:45:33 -!- Phantom_Hooversh has changed nick to Phantom_Hoover.
18:45:37 <alise> In fact, infallability is now the official spelling.
18:45:44 <alise> Did I ever tell you guys I'm the Pope btw?
18:45:46 -!- Phantom_Hoover has changed nick to Hoovershire.
18:45:48 -!- alise has changed nick to aliseshire.
18:46:35 <oerjan> `translatefromto en no drizzle
18:46:50 <aliseshire> oerjan: you don't have it, you can't possibly have a name for it
18:46:57 <oerjan> `translatefromto no en yr
18:47:14 <oerjan> in fact we have two names for it
18:47:28 <oerjan> but google only knows one of them
18:47:59 <pikhq> oerjan: Yeah, but we're English-speakers.
18:48:05 <pikhq> WE TAKE YOUR WORDS
18:48:17 <AnMaster> Hoovershire, how do I fill random to some other density than 50% in golly?
18:48:40 <pikhq> "Duskregn" and "yr" are now English words!
18:48:45 <pikhq> Ain't it awesome speaking English?
18:48:50 <AnMaster> aliseshire, eh? seems weird to make it a preference like that
18:50:05 -!- aliseshire has changed nick to kreivikunta.
18:50:11 <kreivikunta> I'd use this name, except it has "kunt" in it
18:50:58 <Deewiant> That Japanese just says "shaia" (I think) i.e. it only transliterated
18:51:12 <kreivikunta> I did notice it had a striking lack of kanji
18:52:07 <Sgeoshire> AnMaster, http://www.conwaylife.com/wiki/index.php?title=Teleportation
18:52:10 <oerjan> `translatefromto en no shire
18:52:32 <oerjan> ok we have drizzle but apparently we don't have shires
18:54:01 <AnMaster> strange, golly must be leaking memory, swap trashing. Restarting it and it goes away
18:54:02 <oerjan> ah if i believe the german, shire would be "grevskap" in norwegian. and our constitution prohibits nobility, so...
18:54:23 <oerjan> so in fact shires are illegal here :)
18:55:00 <AnMaster> Hoovershire, I did set a memory limit in preferences
18:55:09 <AnMaster> Hoovershire, for both hashlife and quicklife
18:55:13 <AnMaster> Hoovershire, and I was using quicklife
18:55:30 <Hoovershire> You should probably notify the devs, then.
18:55:41 <AnMaster> Hoovershire, yeah mem limit was set to 500 MB for each. System has 4 GB RAM. Golly was according to htop using 80% of total memory
18:56:17 <oerjan> kreivikunta: certainly
18:56:18 <Hoovershire> The access times make it too slow, apparently.
18:56:19 -!- poiuy_qwert has quit (Read error: Connection reset by peer).
18:56:19 <AnMaster> Hoovershire, well 500 MB should be much less that the limit for swap
18:57:46 <AnMaster> Hoovershire, sure there is some overhead. And then there is X and Gnome. But without golly running it is like 200 MB RAM used (including disk cache, which was almost completely eaten due to the swap trashing, kernel prefers to throw away disk cache iirc
18:57:47 <oerjan> kreivikunta: i think that may have been one of the original provisions from 1814, even
18:57:54 -!- poiuy_qwert has joined.
18:58:03 <Sgeoshire> How would a spherical playing field for GoL work?
18:58:23 <AnMaster> Hoovershire, I was studying random fill. I believe it might have happened when switching between hash algorithms
18:58:26 <kreivikunta> oerjan: what, forbidding shires?; or allowing amendment?
18:58:32 <kreivikunta> if allowing amendment, by definition it must have been one of the first
18:58:42 <oerjan> kreivikunta: well both presumably
18:58:46 <AnMaster> Hoovershire, I found hashlife was too slow until it has become mostly stable. But then hashlife works fine
18:58:48 <kreivikunta> If spherical, it's just a regular wrapping.
18:59:08 <AnMaster> Hoovershire, so I used quicklife for some 2000-3000 generations then switched to hashlife. I believe the mem leak is related to that
18:59:21 <AnMaster> Hoovershire, also I got a nice pattern that looks kind of like a face here
18:59:31 <pikhq> kreivikuntashire should be a shire.
18:59:37 <oerjan> kreivikunta: iirc as far as land is concerned what the constitution prohibits is any form of inheritance which requires the heir not to sell the property
18:59:42 <AnMaster> Hoovershire, a collision of gliders leaving the random region produced it
19:00:09 * Hoovershire checks whether gliders can produce Pi-heptominoes
19:00:22 <oerjan> this has the effect of preventing land-based nobility families
19:00:36 <kreivikunta> pikhq: Kreivikuntashire is just shireshire :P
19:00:49 <AnMaster> Hoovershire, well I think it was gliders. A bit hard to tell: initial population was >500,000
19:00:49 <oerjan> but it also forbids inherited noble titles separately, i think
19:01:35 <Sgeoshire> http://www.conwaylife.com/wiki/index.php?title=Aircraft_carrier those islands look further away then they are
19:01:40 <oerjan> (those who were nobility at the time the constitution went into effect stayed so, but it didn't pass to their children. iiuc.)
19:02:23 <oerjan> if i understand correctly
19:03:01 <AnMaster> Hoovershire, it was a 2-glider collision
19:03:32 <Hoovershire> Yeah, pi-heptominoes create a debris cloud that looks like a face and are 2-glider synthesisable
19:03:33 <AnMaster> Hoovershire, two ponds, 6 blocks, 5 blinkers. Arranged kind of like a face
19:04:06 <AnMaster> Hoovershire, probably you know the collision already. Saw some list of possible 2-glider-collisions on the wiki
19:04:21 <AnMaster> Hoovershire, how rare is that in random 25% fill I wonder
19:04:27 <pikhq> kreivikunta: NEED MOAR SHIRE
19:04:29 <AnMaster> Sgeoshire, too busy watching pretty patterns ;P
19:04:55 <pikhq> Deewiant: Also. Yes, of course it only transliterated. "Shire" is a loanword in Japanese. :)
19:05:06 <pikhq> 13:04 [freenode] -!- シャイアshir Erroneous Nickname
19:05:17 <AnMaster> Hoovershire, sure but not quite as common outside the random block from my tries so far
19:06:01 <Deewiant> pikhq: It's not "of course", there could be an equivalent concept it translates to. Although I guess that since it originates in European nobility it should be obvious that there isn't.
19:06:22 <AnMaster> Hoovershire, hm zooming a lot causes swap trash too. But yes memory limit is set and population is "only" 487141 now
19:06:36 <oerjan> kreivikunta: hm it looks like the constitution only prohibited establishing _new_ titles, while a later law of 1821 abolished the old ones (except for people keeping titles until they died)
19:06:41 <AnMaster> Hoovershire, initial population was 4090522
19:07:12 <AnMaster> kreivikunta, hm? why are you repeating what I said?
19:07:56 <oerjan> kreivikunta: apparently norway only had 3 noble titles previous to this, anyway (which presumably made it easy to abolish)
19:08:50 <AnMaster> oerjan, heh. And how many had each of those types of title?
19:08:57 <Hoovershire> AnMaster: Again, if you're getting persistent memory leaks, you should tell the devs.
19:09:27 <AnMaster> Hoovershire, it does seem related to switching algorithm on the fly. Not completely sure. It is a bit random...
19:09:45 <AnMaster> Hoovershire, but yes I will. Probably not this evening. Got some other things to do really...
19:10:25 <oerjan> AnMaster: well three Stamhus (shires?), really. actually one of them (Jarlsberg) still exists, the constitution only prohibits making new ones
19:11:06 <pikhq> Deewiant: Even if there *was* an equivalent concept, the Japanese would still probably have this as a loan word. They treat foreign languages a *bit* like English does.
19:11:16 <oerjan> AnMaster: this is mostly from the norwegian wikipedia article on Adel, which is marked as not very good though
19:11:22 <pikhq> It, too, has native words as a *minority* of the words in the language.
19:13:55 <pikhq> Oh, hey, copumpkin.
19:14:01 <pikhq> Didn't realise you were in here.
19:15:30 <kreivikunta> Where the neighbours are are determined how small the cell number is, order of magnitude; so (0,0) would have neighbours (-1,0), (0,-1), (1,0), (0,1), (1,1), (-1,-1), etc.
19:15:45 <kreivikunta> But (0.1,0.1) would have (-0.2,0.1), (0.1,-0.2), etc
19:15:52 <oerjan> copumpkin: are you dual to a pumpkin?
19:15:54 <kreivikunta> Exercise: Proof this actually works out to anything consistent
19:30:46 <Hoovershire> You can send a signal at 2c/3 to anywhere in the universe.
19:30:55 <Hoovershire> http://pentadecathlon.com/lifeNews/2009/03/working_2c3_signal_elbow.html
19:31:58 <kreivikunta> ah, is it because it's 2/3 instead of 1/2 of light?
19:32:12 <Hoovershire> It means that you can have a 2c/3 signal wire going to any point in the universe,
19:33:42 <augur> you can't hide forever, mister peebles!
19:34:27 <AnMaster> Hoovershire, on life wiki (polyomino page): "Charles Corderman discovered the switch engine by running an exhaustive computer search on all decominoes. The machine that discovered the decomino seed was unique, possessing an unusual architecture. "
19:34:43 <AnMaster> there is no citation or any further details
19:34:50 <AnMaster> as to what this special architecture would be
19:35:04 <AnMaster> google didn't turn up anything useful
19:35:11 <Hoovershire> It means that the search program can't be run on your own computer.
19:35:19 <AnMaster> Hoovershire, so in what way was it special
19:35:34 <Hoovershire> It's probably just some custom machine he built for the purpose.
19:35:38 <oerjan> AnMaster: alien technology, duh
19:35:53 <kreivikunta> yeah probably just a custom searching machine
19:35:54 <AnMaster> Hoovershire, FPGAs? Doesn't say anything about when... so no idea if it was FPGAs or something else
19:36:15 <kreivikunta> Hoovershire: I think my hunch is that a c turn is impossible
19:36:39 <AnMaster> "A switch engine (or Corder engine[1]) is a methuselah that was found by Charles Corderman in 1971. It is unstable by itself, but it can be used to make c/12 diagonal puffers and spaceships. "
19:36:51 <kreivikunta> Hoovershire: because a turn basically has to deconstruct and then reconstruct a vehicle since life has no notion of "turning" as such
19:37:18 <kreivikunta> Hoovershire: My conjecture is that the complexity of the operations required to rebuild it rotated are sufficiently complex that they will always take longer than a straight line
19:37:32 <Sgeoshire> Wait, couldn't you terminate the wire and retransmit before a second elbow
19:37:35 <kreivikunta> Hoovershire: However, it may be possible to build an elbow that, while running slower than c, can plug directly on to two lightspeed wires.
19:37:39 <Hoovershire> kreivikunta: Why can't a signal induce another signal in a wire at 90 degrees?
19:37:48 <kreivikunta> Hoovershire: It can, but it'll take more than one generation.
19:38:24 <AnMaster> why is it an issue if there is some delay in that?
19:38:47 <AnMaster> as long as most of the time there is no delay I mean
19:39:12 <AnMaster> build it with mostly straight segments and turning isn't an issue
19:39:18 <kreivikunta> Because without a delay it's as simple as having a receiver and emitter right next to each other.
19:39:25 <kreivikunta> Which, sure, is possible, but the point is something that's practical, unlike that.
19:39:41 <AnMaster> kreivikunta, see ^ "<AnMaster> err, no more than c*"
19:40:40 <AnMaster> kreivikunta, so as long as you have mostly straight lines with just a few corners it will still be roughly the same speed
19:40:53 <kreivikunta> and we can theoretically make corner-turners now
19:41:10 <kreivikunta> AnMaster: as slow as the receiving and sending ends of a telegraph put together
19:41:21 <Hoovershire> kreivikunta: Suppose you have a signal with an exposed spark, which can induce a signal cleanly in a wire at 90 degrees.
19:41:30 <kreivikunta> receive it, send it into a receiver that sends out the same signal, but on another wire, that's tilted
19:41:33 <AnMaster> kreivikunta, so how long does the cable need to be for that to be useful?
19:41:52 <AnMaster> kreivikunta, well depends on what you need to do with it
19:41:54 <kreivikunta> the LST is already hopelessly impractical.
19:42:11 <AnMaster> kreivikunta, LST seems mostly useful for high speed, very long, links?
19:42:26 <kreivikunta> AnMaster: yes, which is nothing in practice.
19:42:29 <Hoovershire> Only if you need to get across tens of thousands of cells.
19:42:41 <kreivikunta> Hoovershire: Golly/Miscellaneous/lightspeed.rle -- this is just a fuse isn't it??
19:42:42 <AnMaster> Hoovershire, yes and can't that happen?
19:43:00 <AnMaster> fuse would work fine for a one-time signal no?
19:43:04 <Hoovershire> Remember, each pulse is one bit, so the information is dribbling.
19:43:05 * kreivikunta notes something... Most things in Life move not that much slower than lightspeed.
19:43:12 <kreivikunta> Those guys' doppler shift must be fucking insane.
19:43:38 <Hoovershire> AnMaster: A fuse is impractical, because you need to rebuild it at <c.
19:43:55 <AnMaster> Hoovershire, but that only applies if you need to rebuild it
19:44:05 <kreivikunta> one-bit communication at a specified time?
19:44:05 <Sgeoshire> http://conwaylife.com/wiki/index.php?title=Lightspeed_wire the bit about 2c/3 turns seems.. obsolete?
19:44:22 <AnMaster> kreivikunta, yes why not. A "stop, we found the prime we looked for" signal
19:44:26 <kreivikunta> Come to think of it, I've never built a clock in Life. Or, well, much at all.
19:44:33 <kreivikunta> I think I'll do that. This isn't an unsolved problem is it Hoovershire?
19:44:51 <AnMaster> kreivikunta, well you could want it
19:45:11 <kreivikunta> Hoovershire: There has to be some sort of simpler clock.
19:45:14 <AnMaster> kreivikunta, you don't _need_ lightspeed for anything
19:45:26 <Hoovershire> *All* signal circuitry has a clock, in the form of the shuttle type it uses.
19:46:00 <AnMaster> kreivikunta, you seldom build clocks in non-CAs
19:46:18 <AnMaster> because you use an external clock signal probably
19:46:28 <AnMaster> you don't build one in VHDL except for simulation
19:46:41 <Sgeoshire> Oh, multiple elbows are useful for closing loops?
19:46:46 <kreivikunta> [[In the Game of Life, for every p greater than or equal to 14 it is possible to construct a glider gun in which the gliders are emitted with period p.[1]]]
19:46:52 <kreivikunta> Is there a LWSS clock that can emit faster?
19:46:58 <Hoovershire> kreivikunta: Gosper makes sense, because the smallest would be easily-discovered.
19:47:02 <kreivikunta> Or even something which can emit some sort of pulse at c?
19:47:10 <kreivikunta> (Just be emitted every c... or multiple of c)
19:48:19 <AnMaster> Hoovershire, can't gun and bullets travel at same speed? If they are traveling in opposite directions at least
19:48:26 <Hoovershire> But anyway, clocks are unnecessary in Life.
19:49:13 <Hoovershire> Because most technology is innately p30 or p46, so it's all synchronised.
19:49:34 <kreivikunta> Hoovershire: I'd like a clock that goes faster than p14 though
19:49:48 <kreivikunta> even just something that emits some sort of pulse every 13 generations would be cool
19:51:06 <AnMaster> kreivikunta, define a clock? Oscillator which emit something?
19:51:33 <kreivikunta> It is acceptable if the oscillator needs a nearby wire to emit stuff onto.
19:52:16 <Sgeoshire> http://www.yucs.org/~gnivasch/life/lightspeed/index.html
19:52:33 <AnMaster> oh that was the page with the bubbles
19:52:41 <Sgeoshire> Is lightspeed transmission stuff that inefficient, that a p91090 gun is useful?
19:53:10 <Sgeoshire> Oh, that's just transmission stuff
19:54:24 -!- augur has quit (Remote host closed the connection).
19:54:34 <Sgeoshire> Wait, the page actually talks about makign stripe agar?
19:54:49 <kreivikunta> http://www.yucs.org/~gnivasch/life/lightspeed/index.html
19:54:56 <kreivikunta> All the alive/dead alternating rows -- that's stripe agar.
19:55:19 <Hoovershire> kreivikunta: Are you still on the p13 clock?
19:55:59 <kreivikunta> Let's see... at the left and right ends of the white lines, just on the border; on the top and bottom, alive cells with two dead cells in-between, such that there are alive cells on both very edges at both top and bottom... and both the top and bottom non-bordered lines should be alive.
19:56:04 <kreivikunta> Yep, I think I know how to make stripe agar now.
19:56:56 <Hoovershire> kreivikunta: You might want to look at Greyships and spacefillers.
19:57:10 <Hoovershire> They work by creating agars, normally striped.
19:57:28 <AnMaster> Sgeoshire, I can't find any page on the lifewiki about p91090
19:57:33 <Hoovershire> AnMaster: A spaceship containing an agar with 1/2 density.
19:57:45 <Sgeoshire> AnMaster, search http://www.yucs.org/~gnivasch/life/lightspeed/index.html
19:58:07 <AnMaster> Sgeoshire, "phrase not found", did you typo it?
19:58:45 <AnMaster> I copy-pasted from <Sgeoshire> Is lightspeed transmission stuff that inefficient, that a p91090 gun is useful?
19:59:28 <kreivikunta> Hoovershire: I just made some stripe agar, but it isn't stable.
19:59:39 <kreivikunta> Do I have to do something to it after I prepare the base material? Does it have to be a certain size? It looks right...
20:00:01 <Hoovershire> Stripe agar needs regular bookends at the boundaries.
20:00:48 <AnMaster> Hoovershire, got a link to an example grayship? the life wiki basically have a stub
20:01:44 <Hoovershire> AnMaster: Cambrain-Explosion.rle has some greyships at the top.
20:02:11 -!- calamari has joined.
20:02:23 <calamari> Gregor: http://4.bp.blogspot.com/_4LYV_SFknMc/S9H7jdd_tQI/AAAAAAAAAQk/dKoK6biWn9Y/s1600/KFC-KRISPY-KREME-LUTHER-DOUBLE-DOWN-DOUGHNUT-BUN-D.jpg
20:02:46 -!- augur has joined.
20:02:49 <AnMaster> Hoovershire, which directory is it in?
20:03:13 <Hoovershire> kreivikunta: The bookends on the sides need to be on every other line.
20:03:52 <Sgeoshire> What, exactly, makes it so inefficient?
20:04:11 <kreivikunta> Sgeoshire: the only signals that like travelling along wire at light-speed are chaotic and hard to both produce and detect
20:04:19 <kreivikunta> to produce you need to like collide gliders together at precise timings
20:04:21 <AnMaster> Hoovershire, which site is the front?
20:04:30 <kreivikunta> and detecting them at the end requires very complicated, large and controlled machinery
20:04:58 <AnMaster> Hoovershire, are that entire group greyships or?
20:05:11 <kreivikunta> http://1.bp.blogspot.com/_TO9A8O1Azeg/RnZs8QADCGI/AAAAAAAAAIM/11c9lJax9Z0/s1600-h/Pi-tracks-DRH-17Feb1997.GIF CREEY FACE CIRCUITS
20:07:27 <Hoovershire> AnMaster: There's a pattern that stretches a striped agar horizontally at -123, 22
20:09:12 <kreivikunta> It carries a whole galaxy of life organisms.
20:10:49 <Hoovershire> There were the beginnings of a (13,1) spaceship on the caterpillar page.
20:11:34 <kreivikunta> I still don't understand the (a,b) notation.
20:12:01 <Hoovershire> It moves 13 cells left for every 1 it moves up.
20:12:08 -!- Sgeoshire has quit (Remote host closed the connection).
20:12:14 -!- Hoovershire has changed nick to alise.
20:12:26 -!- alise has changed nick to Hoovershire.
20:12:35 <AnMaster> does anyone know if fizzie is away for the weekend or something?
20:14:29 <Deewiant> /whois says he's been idle for almost 24 hours, so maybe he's away for today.
20:16:33 <Ilari> Would that be first spaceship tha does not move in one of 8 main directions? Or has some such ship aleady been built?
20:17:12 <Gregor> Jan Vitek: Flash is to computing, what porn is to cinema?
20:17:12 <Gregor> Gregor Richards: ... YES. I agree with that statement 115%.
20:17:33 <pikhq> Why hello, Richardsshire.
20:17:40 <Hoovershire> Also, everything interesting in Life is found with searches these days.
20:17:41 <kreivikunta> Hoovershire: So what's wrong with my agar?
20:18:32 <kreivikunta> Gregor: Jan Vitek's Facebook photo: http://www.facebook.com/profile/pic.php?oid=AAAAAwAgACAAAAAJOxKn1xL6DjDHhndQyKszQjZLwByUy-T-E0I5xvwVi4D_tpmGTj9TFtyFVsNHHm5cWyKXCFE9IzkwvJZABz70l0zVm61rAYnYoZxN3j_FXfjxVH0OgTyfYPdvehsrEVuG&size=normal
20:18:39 <kreivikunta> ARE YOU SURE YOU WANT TO AGREE WITH THIS MAN
20:18:54 <Gregor> kreivikunta: He's my advisor :P
20:18:57 <fizzie> (I'm not away, I'm just busy.)
20:19:09 <calamari> Gregor: I knew you'd like that
20:19:20 <pikhq> kreivikunta: That's Jan Vitekshire, sir.
20:19:29 <calamari> looks like a yummy breakfast sandwich
20:19:38 <Hoovershire> kreivikunta: I told you: the end bookends need to be on alternate blank rows.
20:19:42 <pikhq> And thats Worcestershire to you!
20:19:50 -!- Gregor has changed nick to GregorRichardsvi.
20:19:59 -!- GregorRichardsvi has changed nick to Richardsville.
20:20:10 <kreivikunta> Josiah "Pick Headquarters" Worcestershire, amirite
20:20:41 <pikhq> kreivikunta: Lawls
20:21:10 <kreivikunta> Because the end bits need to attach to an on row.
20:21:55 -!- Sgeo has joined.
20:21:58 -!- Richardsville has changed nick to Richardsgrafstvo.
20:22:43 <Sgeo> http://www.wovencroft.com/Images/tm51610a.gif how helpful to have the time and date disappear and reappear like that
20:22:48 <oerjan> amirite would be a badass mineral
20:23:20 <oerjan> kreivikunta: shirely you can't be serious
20:24:21 <kreivikunta> But adding another row cannot possibly work!
20:26:14 <kreivikunta> I would like to inform you that that cannot possibly work.
20:27:38 <Sgeo> Fun. The mayor posts an rtf, and everyone has problems
20:28:49 <kreivikunta> Hoovershire: So I need there to be off rows before the final on row that do not haev a pixel, right?
20:29:24 <kreivikunta> Hoovershire: I don't just want to use your fixed version because I want to figure out what the heck I did wrong!
20:30:12 <kreivikunta> <Hoovershire> http://pastebin.com/pqjVb0mg
20:31:04 <Hoovershire> OK, you see how the row ends have dead cells with 3 live neighbours?
20:32:24 <Hoovershire> But the last row can't have a bookend on it.
20:33:11 -!- Hoovershire has quit (Read error: Connection reset by peer).
20:33:32 -!- Hoovershire has joined.
20:34:10 -!- chickenzilla has quit (Quit: leaving).
20:34:13 <kreivikunta> Note to self; make a program to generate stripe agar.
20:34:21 <kreivikunta> Hoovershire: now i just need to figure out how to transmit signals :)
20:34:54 <Hoovershire> The simplest signal for a stripe agar is made as follows:
20:34:56 -!- Richardsgrafstvo has changed nick to Gregor.
20:35:50 <Hoovershire> At one end of the empty bit, put a cell one up from the stripe
20:36:19 <Hoovershire> Add a cell one down from the stripe at the next cell.
20:36:41 <Hoovershire> The cells up and down are at the back of the blip.
20:37:28 <kreivikunta> and the middle of the c has blanks after it
20:38:23 <Sgeo> There is someone here using Win98
20:38:47 -!- poiuy_qwert has quit (Quit: This computer has gone to sleep).
20:39:32 <kreivikunta> It works; but when it reaches the end it destroys the agar.
20:40:16 <Sgeo> At this big, important event: 34 people
20:40:21 <Sgeo> 43 people online total
20:40:25 <Sgeo> Used to be hundreds
20:41:29 -!- pikhq_ has joined.
20:42:25 <Sgeo> Hoovershire, no
20:42:28 <kreivikunta> No, on some shitty old 3D VR game he likes
20:42:33 <Sgeo> At this "Town Hall" meeting
20:42:47 <Sgeo> kreivikunta, um, are you sure you're not alise in dis.. you're alise, aren't you?
20:43:17 <Hoovershire> kreivikunta: Yes, the blip is unstable at beginning and end.
20:43:32 <kreivikunta> Hoovershire: There's no machine you can surgically attach to the end of the agar to keep it stable?
20:43:39 -!- pikhq has quit (Ping timeout: 276 seconds).
20:43:40 * Sgeo doesn't like it enough to pay $5/month
20:44:19 <Sgeo> Some areas of the city have been "fixed". DOn't know what could have possibly been broken, other than the fact that it's dying
20:44:20 <Hoovershire> I doubt we'll be able to find one without a search program.
20:44:24 <kreivikunta> Hoovershire: So agar is not that useful, unless we can make stable, stationary machines that work inside agar.
20:44:33 <kreivikunta> (Then agar would essentially become a mini-CA in itself.)
20:44:51 <kreivikunta> Hoovershire: Because you can't detect any signals, it's just chaos
20:45:18 <Sgeo> What about <thing that the new parent company has been promising for years>? "Dead"
20:46:02 <Sgeo> Seriously, I will say that I _HATE_ IVN
20:46:12 <Sgeo> (The company that owns Cybertown)
20:48:08 -!- pikhq has joined.
20:48:39 -!- pikhq_ has quit (Ping timeout: 240 seconds).
20:49:04 <Sgeo> The most deadly thing to Cybertown was the newly added cost, and the fact that non-payers can do very, very little
20:49:06 <Sgeo> They can barely chat
20:51:19 -!- marchdown has quit (Quit: marchdown).
20:51:51 <oerjan> Hoovershire: XOR only gives you linear functions
20:52:38 <oerjan> you also need a 1 or EQV to get all of them
20:52:42 -!- augur has quit (Ping timeout: 252 seconds).
20:53:50 <oerjan> http://en.wikipedia.org/wiki/Post's_lattice is all about what sets of boolean functions generate which others
20:54:44 <oerjan> XOR gives AP_0, with 1 too you get A
20:55:34 -!- marchdown has joined.
20:55:36 <oerjan> all affine boolean functions, i.e. f(x1,...,xn) x_(i_1) + ... + x_(i_k) + C
20:56:10 <oerjan> P_0 are the 0-preserving ones, AP_0 is their intersection
20:57:51 -!- pikhq_ has joined.
20:58:11 <Sgeo> http://www.integratedvirtualnetworks.com/index1.php?is_ct=1 <<===CYBERTOWN KILLERS
20:58:27 -!- pikhq has quit (Ping timeout: 240 seconds).
20:59:07 <Sgeo> Forced Cybertown to move to a subscription system
20:59:09 <oerjan> kreivikunta: top symbol, apparently
20:59:15 <oerjan> it's a lattice after all
20:59:50 <Hoovershire> http://www.yucs.org/~gnivasch/life/lightspeed/index.html
20:59:52 <oerjan> kreivikunta: it's the top symbol, i said, it requires unicode
21:00:02 <oerjan> in that article's notation
21:00:25 <oerjan> kreivikunta: T is used for something else, although only with indexes
21:00:28 <Hoovershire> I've asked on the forum what the state of the art is, as that page is from 2008.
21:00:54 <kreivikunta> Hoovershire: link to the forum? think I've seen it before
21:00:58 <kreivikunta> also, is there a game of life irc channel?
21:01:20 <Hoovershire> http://www.conwaylife.com/forums/index.php
21:01:55 -!- kreivikunta has changed nick to alise.
21:01:56 <oerjan> anyway there are 5 maximal subclasses of the top, M, A, P0, P1 and D. to generate everything you need to have a function outside each of them. note that one function can be outside several, e.g. NAND obviously isn't in any of them
21:02:29 -!- Hoovershire has set topic: Howsabout we put "esoteric programming languages" SOMEWHERE in the topic? | Esperanto is still bannable! | Je peux utiliser une langue étrangère aussi. | xn--v8jad0f7b6z4eoa6v0hk534a7hlwhnnl8s. | Jag älskar Unicode i mina topicar. | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
21:02:58 -!- Hoovershire has set topic: For Game of Life discussion go to #b2s23 | http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
21:03:28 -!- pikhq has joined.
21:03:33 -!- impomatic has joined.
21:03:37 <impomatic> Has anyone played with one of these? http://en.wikipedia.org/wiki/SheevaPlug
21:03:44 -!- Hoovershire has changed nick to Phantom_Hoover.
21:03:51 <alise> I know people who have
21:04:30 -!- pikhq_ has quit (Ping timeout: 265 seconds).
21:05:33 <oerjan> M are monotonic functions. AND, OR, 0 and 1 generate all of them
21:08:54 <alise> copumpkin: #b3s23!
21:08:57 <alise> I could have done that in one message! :P
21:09:32 <oerjan> if you know which class each of your basic functions generate, that hasse diagram in the article tells you everything else about what it generates
21:09:42 <Sgeo> "We invite you to log in and learn more about Silhouette and the Nexos."
21:10:05 <alise> I DID NOT SAY JOINING WAS OPTIONAL
21:10:08 -!- pikhq_ has joined.
21:10:35 <oerjan> AND is /\P, OR is \/P, NOT is UD and XOR is AP_0
21:11:14 -!- pikhq has quit (Ping timeout: 264 seconds).
21:11:36 -!- ParaSait has joined.
21:11:44 <alise> what about op(a,b,c,d,e,f,...) := (a /\ c /\ e /\ ...) something (b \/ d \/ f \/ ...)
21:11:50 <oerjan> that table to the left of the diagram tells you a set of generators for each class
21:11:53 <alise> for different values of something
21:12:17 <ParaSait> *can somebody explain me FukYorBrane? :P*
21:12:28 <alise> ParaSait: no but Gregor can
21:12:59 <oerjan> alise: i think that depends a lot on "something"
21:12:59 <alise> oerjan: what do you mean er what
21:13:14 <alise> op(a,b,c,d,e,f,...) := (a /\ c /\ e /\ ...) xor (b \/ d \/ f \/ ...)
21:13:34 <oerjan> well op(a,b) is XOR itself
21:14:09 <Gregor> I'm not sure I can X-D
21:14:26 <ParaSait> Just had a few questions bout it
21:14:32 <Gregor> ParaSait: It's sort of like Brainfuck, except your "tape" is the other player's code, so you want to ... y'know ... break it.
21:14:36 <oerjan> alise: op(a,a) is 0, then you get op(a,0,c) = a /\ c and op (0,b,0,d) = b \/ d
21:15:29 <Gregor> ParaSait: It switches your "write head" from one program to the other.
21:15:32 <alise> oerjan: can you get not
21:15:40 <alise> op(a,1) would do it
21:15:41 <Gregor> I, err, don't recall how it decides where to land you in the other program >_>
21:15:52 <alise> we have 0, and, or, xor...
21:15:58 <ParaSait> Heh. That's what I was wondering too :P
21:16:02 <alise> Gregor: same location?
21:16:05 <oerjan> alise: you get P_0. it does not contain not.
21:16:18 <oerjan> alise: you need a 1 to get not from xor
21:16:21 <Gregor> alise: Could be, mabye modulo program size or something? I'd have to look at the code to recall.
21:17:03 <oerjan> alise: (P_0 has \/ and + listed as generators)
21:17:32 * Sgeo decides that he may pay $5 every few months
21:17:43 -!- pikhq has joined.
21:18:03 <alise> op(a,b,c,d,e,f,...) := (a xor c xor e xor ...) and (b xand d xand f xand ...)
21:18:03 <oerjan> alise: basically P_0 is every function which maps all 0's to 0
21:18:10 <alise> does xand even make any sense
21:18:20 <alise> a xor b = (a or b) and not (a and b)
21:18:23 <alise> a xand b = (a and b) and not (a and b)
21:18:24 <oerjan> alise: do you mean <-> ?
21:18:38 <alise> oerjan: yeah let's go with that
21:18:39 -!- pikhq_ has quit (Ping timeout: 240 seconds).
21:18:43 <alise> op(a,b,c,d,e,f,...) := (a xor c xor e xor ...) and (b eqv d eqv f eqv ...)
21:18:57 <ParaSait> Gregor: Well I could do some research in the source of that fyb implementation. But I can't actually run that program cause for some reason it crashes asa I exec it with those 2 args
21:19:14 <oerjan> well op(a,b) = a and b
21:20:06 <Gregor> ParaSait: AFAIK it's just ./fukyorbrane <program> <program
21:20:08 <oerjan> op(a,a) = a and a, that doesn't help much
21:20:28 <oerjan> op(a,a,a,a) = (a xor a) and (a <-> a) = 0
21:20:34 <Gregor> ParaSait: And E_WORKSFORME. What's your platform?
21:20:42 <alise> op(a,b,b,c) := (a xor b) and (b = c)
21:20:56 <alise> = (a or b) and not (a and b) and (b = c)
21:21:53 <alise> op(a,b,c,c) := (a xor c) and (b = c)
21:22:15 <oerjan> op(a,0,c,0) = (a xor c) and 1 = a xor c
21:22:27 <ParaSait> Gregor: Vista. Compiled with codeblocks
21:22:34 <oerjan> we have and, xor and 0
21:23:53 <oerjan> we cannot ensure the xor part is 1
21:24:09 <alise> hasn't it been proven that nand and nor are the only universal ones
21:24:32 -!- marchdown has quit (Quit: marchdown).
21:24:34 <oerjan> only for binary operators, i think
21:24:58 <oerjan> and this is not even a single function really
21:26:14 -!- pikhq has quit (Ping timeout: 264 seconds).
21:26:57 <oerjan> i think this is also P_0, we cannot get anything more
21:28:20 <oerjan> hm or wait and is just /\P
21:29:48 <Gregor> ParaSait: laaaaaaaaaaaawlz
21:29:56 <Gregor> ParaSait: Not guaranteed to work on shitty OSes :P
21:29:58 -!- pikhq has joined.
21:30:05 <alise> op(a,b,c) := (a nor c) xor b
21:30:19 <Gregor> ParaSait: That being said, it really still ought to >_>
21:30:27 <alise> = (not (a or b)) xor b
21:30:34 <Gregor> Do you have more details than "it crashed" as per why it crashed?
21:30:58 <alise> = ([not (a or b)] or b) and not ([not (a or b)] and b)
21:31:04 <oerjan> op(a,a,a) = (not a) xor a = 1
21:31:42 <alise> = (not (a or c)) xor b
21:31:49 <alise> = ([not (a or c)] or b) and not ([not (a or c)] and b)
21:32:37 <alise> (a nor c) xor b seems to be the simplest way to phrase it.
21:32:51 <alise> oerjan: you can't get 0 without one 0 already
21:33:11 <oerjan> simple enough. we get 1, or
21:33:36 -!- FredrIQ has quit (Quit: - nbs-irc 2.39 - www.nbs-irc.net -).
21:33:49 <alise> op(a,b,c) := (a xor b) nor (a xor c)
21:34:03 <alise> op(a,a,a) := (a xor a) nor (a xor a) = 0 nor 0 = not (0 or 0) = 1
21:34:04 <ParaSait> Gregor: Well I run it with the line: fukyorbrane.exe prog1.fyb prog2.fyb, then I get a console window and immediately a stopped working message
21:34:11 <oerjan> er is this still the same one?
21:34:23 <Gregor> ParaSait: I don't know what a "stopped working message" is ...
21:34:31 <oerjan> alise: i wasn't quite finished
21:34:44 <alise> but i proved it isn't complete!
21:35:24 <ParaSait> Gregor: Well you know. Windows detects a runtime error and then gives a little window that says it's looking for a solution and eventually wants to make an error report, etc etc
21:35:27 <oerjan> well sure, but 1 and or only give us VP_1 which has the whole T tower above it
21:35:37 <Gregor> ParaSait: Ah ... well that's quite useless.
21:35:46 <oerjan> so it could be any of those
21:35:51 <Gregor> So anywho, EgoBot can run FYB :P
21:35:54 <EgoBot> Sorry, I have no help for fyb!
21:35:59 <Gregor> EgoBot: WELL SCREW YOU THEN
21:36:03 <oerjan> (well the right T tower)
21:36:19 <EgoBot> Sorry, I have no help for fukyorbrane!
21:36:49 <Gregor> ParaSait: It has a full-on FYB hill, but I apparently was too stupid to put any of that in !help :P
21:37:06 <oerjan> <alise> op(a,b,c) := (a nor c) xor b
21:37:52 <EgoBot> Score for Gregor_badmangler: 0.0
21:38:00 <Gregor> Now where does it put the hill ...
21:38:10 <Gregor> http://codu.org/eso/fyb/in_egobot/
21:38:58 -!- pikhq has quit (Read error: Connection reset by peer).
21:39:01 -!- copumpkin has left (?).
21:39:12 -!- pikhq has joined.
21:39:33 <ParaSait> The bot might be able to run fyb, but I still don't get it xD
21:39:46 <Gregor> Well, you can read the code even if you can't make it work on Windoze :P
21:43:38 -!- marchdown has joined.
21:43:50 * pikhq lurves using the stack pointer as a general purpose register
21:49:06 <Gregor> The stack is for the weak.
21:51:21 <oerjan> alise: ok i've confirmed that is not in T_1^2 and now my head hurts :D
21:51:27 <HackEgo> Runs arbitrary code in GNU/Linux. Type "`<command>", or "`run <command>" for full shell commands. "`fetch <URL>" downloads files. Files saved to $PWD are persistent, and $PWD/bin is in $PATH. $PWD is a mercurial repository, "`revert <rev>" can be used to revert to a revision. See http://codu.org/projects/hackbot/fshg/
21:51:31 <oerjan> which means it must be P_1
21:51:38 -!- pikhq has quit (Read error: Connection reset by peer).
21:52:01 -!- pikhq has joined.
21:52:35 <Phantom_Hoover> Can we run it on #b3s23 with bgolly as one of the programs?
21:52:53 -!- Oranjer has joined.
21:53:13 <Gregor> You can put whatever programs you want to on it.
21:53:20 <Gregor> So long as you do it yourself :P
21:53:29 <Gregor> As per putting it on another channel, sure.
21:57:17 <oerjan> 22:33 alise> op(a,b,c) := (a xor b) nor (a xor c)
21:57:17 <oerjan> 22:33 alise> op(a,a,a) := (a xor a) nor (a xor a) = 0 nor 0 = not (0 or 0) = 1
21:57:23 -!- HackEgo has quit (Remote host closed the connection).
21:57:23 -!- EgoBot has quit (Remote host closed the connection).
21:57:26 -!- EgoBot has joined.
21:57:26 -!- HackEgo has joined.
21:58:20 <oerjan> alise: on to the next one you said
21:58:40 <oerjan> i don't think you can get 0 in it
21:58:55 <alise> you know we could just write a computer program to figure out all this
21:59:01 <alise> i mean it is just a truth table...
21:59:20 <oerjan> basically to check for containment in P_0 and P_1 all you need to do is apply the function to all 1's and all 0's respectively
21:59:53 -!- pikhq has quit (Read error: Operation timed out).
22:00:19 <alise> well, yeah, exactly
22:00:22 -!- pikhq has joined.
22:00:39 -!- poiuy_qwert has joined.
22:00:49 <oerjan> M, D and A aren't too hard to check either
22:01:11 <oerjan> so yeah a program could check if everything is generated
22:01:23 <alise> eh, sounds fun, /me writes it
22:01:26 <alise> not with your freaky post names though
22:01:30 <alise> with durned good easy names
22:02:21 <oerjan> i learned this stuff from post's original book, he used another completely different notation
22:03:05 <oerjan> this notation is actually an improvement :D
22:03:59 <oerjan> (much less unexplained indexes, most combinations are written with intersection/concatenation instead)
22:05:27 <alise> (p == q) == ~((p\/q) /\ ~(p /\ q))
22:05:27 <oerjan> it's pretty clearly possible to write a program to check exactly which class is generated, too
22:05:34 <alise> ~(~p /\ ~q) = p \/ q
22:05:39 <alise> ~(p /\ ~q) = ~p \/ q
22:05:52 <alise> (p == q) == ~(p \/ q) \/ (p /\ q)
22:05:56 <alise> (p == q) == (p nor q) \/ (p /\ q)
22:06:15 <oerjan> um, i guess class = clone in the modern notation
22:08:42 * alise gets bored of writing the program :P
22:09:27 <Sgeo> AnMaster, "If you're still not convinced, then construct such a pattern for yourself. You'll find that it either functions below lightspeed, explodes, or otherwise detects the 'ghost' LWSSes as being actual ones, and emits a spaceship regardless of whether a spaceship actually entered the device."
22:14:06 -!- Phantom_Hoover has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100423140709]).
22:25:53 -!- ParaSait has quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]).
22:27:42 <alise> http://www.reddit.com/r/sml SML FUCK YEAH
22:30:15 * Sgeo might get a 30 day renewal of Cybertown access
22:31:01 <alise> http://www.list4everything.com/list-of-idioms-to-use-at-workplace.html
22:31:05 <alise> LIST OF IDIOMS TO USE AT WORKPLACE
22:31:36 <alise> http://www.list4everything.com/sentences-that-are-used-repeatedly-by-professionals.html
22:31:40 <alise> SENTENCES THAT ARE USED REPEATEDLY BY PROFESSIONALS
22:32:13 <oerjan> oh wait, that's not a sentence.
22:33:02 <alise> It's missing mathematicians
22:33:11 <alise> Perhaps "I conjecture that ..."
22:33:30 <oerjan> well then it is presumably correct, modulo mathematicians.
22:33:46 * Sgeo is in CyberTown!
22:34:36 <oerjan> "We leave it as an exercise for the reader..."
22:34:50 <oerjan> all mathematicians love that one. trust me.
22:35:36 -!- calamari has quit (Quit: Leaving).
22:36:41 -!- kar8nga has quit (Read error: Connection reset by peer).
22:38:39 <alise> oerjan: is the proof left as an exer-
22:40:00 <oerjan> as much as possible of it, anyhow
22:40:57 <oerjan> my advisor sometimes referred to "proofs by general nonsense", although i'm not sure that term is very popular
22:41:24 <oerjan> this applied especially to heavy use of category theory :)
22:42:04 -!- marchdown has quit (Quit: marchdown).
22:42:24 <alise> oerjan: category theory = general abstract nonsense
22:43:15 <oerjan> hm there are actually google hits to apparently real papers using the term
22:44:10 <alise> well it's a very common term
22:44:18 <alise> old too (1940s says wp)
22:44:22 <alise> i love category theory
22:44:32 <alise> it's... everything
22:45:05 <alise> http://languagelog.ldc.upenn.edu/myl/AbstractNonsense1.png
22:45:12 <alise> http://languagelog.ldc.upenn.edu/myl/AbstractNonsense2.png
22:45:15 <alise> oerjan: ^ relevant
22:46:55 <oerjan> my first published paper had some nice abstract nonsense in it, i recall
22:47:07 <oerjan> (not using that term though)
22:47:41 <oerjan> "by the naturality of the pimsner-voiculescu exact sequence" iirc
22:48:04 <alise> one day I will undersatnd your papers
22:48:39 <oerjan> we didn't actually find a reference that it _was_ natural, although i did check that it was, and we didn't bother to elaborate on that... :)
22:50:21 <oerjan> (i guess we left the proof as an exercise for any reader smart enough to notice it was needed)
22:51:45 * Sgeo decides that Cybertown "freemail" is poorly named
22:52:52 <alise> oerjan: mathematicians basically hate each other don't they
22:53:00 <alise> "I DO have a proof for this, but I'll let you do it instead..."
22:54:17 -!- impomatic has quit (Quit: ChatZilla 0.9.86 [Firefox 3.5.9/20100315083431]).
22:54:36 -!- marchdown has joined.
22:55:14 <oerjan> in our case it wasn't so much that we wanted them to do it as that it was obvious it _should_ be true and would require a lot of work to write down that was really unrelated to the rest of the paper
22:55:38 -!- tombom__ has quit (Quit: Leaving).
23:08:09 <alise> I WANT IT TO BE TRUE
23:20:07 * pikhq has code that segfaults when not run under valgrind
23:22:03 <oerjan> alise: consider 0 and a -> b = (not a) or b
23:22:26 <oerjan> 0 is not in P_1 or D, while -> is not in P_0, M, A or D.
23:22:35 <oerjan> thus together they generate everything.
23:23:42 <alise> hmm how do you do and without quantifiers...
23:24:29 <alise> it makes sense, but ugh
23:24:46 <oerjan> we _are_ talking boolean logic here
23:26:19 <alise> I'm just so used to the constructive-function meaning of ->
23:26:26 <alise> so the ~a \/ b one freaks me out sometimes
23:27:48 -!- oerjan has quit (Quit: Reboot).
23:30:09 -!- pikhq_ has joined.
23:30:51 -!- oerjan has joined.
23:32:48 -!- pikhq has quit (Ping timeout: 276 seconds).
23:34:35 <oerjan> "The relative simplicity of Post's lattice is in stark contrast to the lattice of clones on a three-element (or larger) set, which has the cardinality of the continuum, and a complicated inner structure."
23:35:39 * oerjan wonders if even determining whether a set of 3-valued functions generate _everything_ is hard
23:39:33 -!- pikhq has joined.
23:40:59 -!- pikhq_ has quit (Ping timeout: 264 seconds).
23:44:37 -!- ballard45 has joined.
23:46:41 <pikhq> Mmm, Brainfuck-to-assembly compiler...
23:46:53 <alise> Assembly-to-Brainfuck compiler
23:49:56 -!- pikhq_ has joined.
23:50:51 -!- pikhq has quit (Ping timeout: 240 seconds).
23:55:04 <Sgeo> How does a p2 spaceship travelling at c make sense??
23:56:24 <alise> non-standard analysis is so cool
23:56:31 <alise> Sgeo: it can't unless it has some sort of agar or wire...
23:56:44 <alise> it moves right all the time
23:57:14 <alise> so it's two c-travelling spaceships that change into each other every generation
23:58:04 <Sgeo> http://fano.ics.uci.edu/ca/rules/b0236s2345/g1.html
23:58:52 <Sgeo> What would be cool is different areas of space following different rules