00:00:34 <kallisti> I guess I'll insert some mathspeak: intervals of wavelengths within the visible color spectrum such that a suitable sample size of wavelengths from this interval are indistinguishable to 100% of a suitable sample size of human test subjects.
00:00:47 -!- Frooxius_ has joined.
00:01:34 <fizzie> Color isn't even a wavelength (unless you think "brown" isn't a color); it's a spectral energy distribution.
00:02:33 -!- Frooxius__ has joined.
00:03:21 <kallisti> isn't a specific color a (wavelength, luminance) pair?
00:03:37 <Gregor> You don't get to be a color.
00:04:12 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
00:04:16 <fizzie> Not if you ask a regular person. There's no "brown" wavelength.
00:04:17 -!- Frooxius has quit (Ping timeout: 240 seconds).
00:04:23 -!- Frooxius___ has joined.
00:04:26 -!- Frooxius___ has changed nick to Frooxius.
00:04:39 <fizzie> You need at least three coordinates, given that there's (usually) three types of color receptors.
00:04:49 <Gregor> If we were dichromates, all colors we were capable of perceiving would be representable identically as (wavelenth, luminance) pairs.
00:04:53 <Gregor> But we're trichromates.
00:04:58 <Gregor> Hence fizzie's mention of three :)
00:05:20 -!- Sgeo|web has quit (Quit: Page closed).
00:05:24 <fizzie> Except for the tetrachromates. Hands up, everyone who's one.
00:05:43 -!- Frooxius_ has quit (Ping timeout: 244 seconds).
00:05:51 <elliott> OK: http://li278-81.members.linode.com/wiki/Main_Page
00:05:55 <elliott> Does the site run fast for you guys?
00:05:59 <fizzie> I was just about to say "(You'll be shot.)"
00:06:02 <elliott> It should do (all the caching and everything is on).
00:06:09 <elliott> Although I haven't yet turned the PHP cache up to the maxxx.
00:06:19 <elliott> Gregor: How much did you click around? :P
00:06:35 * elliott would be interested in numbers on reloads (just plain Ctrl+R) from Chrome's Network tab.
00:07:16 <fizzie> I just frontpaged, that's all.
00:07:29 <fizzie> The phone's not really best for speedtesting.
00:07:46 -!- Frooxius__ has quit (Ping timeout: 255 seconds).
00:08:33 -!- pikhq has quit (Read error: Operation timed out).
00:08:55 -!- pikhq has joined.
00:10:13 <kallisti> I apparently don't understand color at all.
00:10:21 <elliott> pikhq: Hey, how fast does http://li278-81.members.linode.com/wiki/Main_Page load
00:11:01 -!- Chef_ has joined.
00:11:37 <kallisti> http://en.wikipedia.org/wiki/File:Cones_SMJ2_E.svg
00:11:45 <kallisti> this looks a lot like a wavelength, luminance pair to me.
00:14:01 <fizzie> kallisti: Yes, those are the response curves for the three types of receptors; a single (wavelength, luminance) pair is not enough to produce all (R1,R2,R3) response triplets that the receptors generate.
00:14:43 <zzo38> Please do not use the guns no being shot please.
00:15:08 <elliott> zzo38: If you tell me how quickly http://li278-81.members.linode.com/wiki/Main_Page loads, I'll never use the guns!
00:15:25 <kallisti> fizzie: hmm, because you can be perceiving multiple sources of light at once right?
00:16:34 <fizzie> It's not so much multiple sources as it is non-monochromatic sources; light that has energy at multiple wavelengths.
00:16:39 <zzo38> How do I time it on Windows?
00:16:41 <fizzie> Admittedly two coordinates in the CIE xy chromaticity diagram might be enough to define a "color", if you think all colors are in http://en.wikipedia.org/wiki/File:CIExy1931.png -- but that means "red" and "dark red" aren't differen't colors.
00:18:15 <kallisti> wouldn't dark red just have lower intensity...
00:18:18 <fizzie> Normally you just designate three wavelengths as "primaries", and specify their amplitudes, and then you get a triangle on that graph that you can repesent using those.
00:18:19 <zzo38> The footer HTML comment says 0.048 seconds
00:18:37 -!- nooga has quit (Ping timeout: 240 seconds).
00:18:41 <elliott> zzo38: That's how long MW takes to generate it. But anything faster than "it took 5 seconds" is OK by me.
00:18:41 <zzo38> But I don't know if that is correct
00:19:01 <zzo38> It does take less than 5 seconds.
00:19:36 <fizzie> Well, if you have (x,y) coordinates from that graph and "intensity" (it's not quite it), you again have three coordinates.
00:19:57 <zzo38> I am unable to log in.
00:20:08 <kallisti> fizzie: oh, hm, you're saying that by lowering the intensity at the "red" wavelength you actually add more green.
00:20:16 <fizzie> Normally you just designate three wavelengths as "primaries", and specify their amplitudes, and then you get a triangle on that graph that you can repesent using those primaries. Then you print your triangle and show it's bigger than your competitor's triangle.
00:20:39 <fizzie> Then home-theater enthusiasts buy your projector instead.
00:20:44 <elliott> zzo38: Yes, that's correct. It doesn't have the user table and the database is locked./
00:20:53 <kallisti> fizzie: do these competitions often happen over the internet? I think I've heard about what you're talking about.
00:20:53 <elliott> zzo38: I'm going to show this to Graue, basically.
00:21:07 <elliott> fizzie: Are those nameservers still down?
00:21:30 * kallisti still doesn't completely understand the distinction
00:21:36 <zzo38> fizie: No I want to use color with any number of primaries. RGB is a useful approximation but for high-quality it is insufficient.
00:22:04 <kallisti> it seems that the responsivity curves would describe all of the perceivable colors...
00:22:28 <kallisti> what is the third input parameter that doesn't correspond to wavelength and luminance?
00:22:32 -!- Chef_ has quit (Remote host closed the connection).
00:22:57 <zzo38> Whether it is trichromate or tetrachromate, there is still potentially inaccuracy but if it is trichromate only, it is somewhat inaccurate and if tetrachromate as well it is slightly less inaccurate but will still be inaccurate regardless; combination is less inaccurate but still inaccurate too bad sorry
00:23:17 <fizzie> kallisti: You need to be able to generate all possible (R1,R2,R3) values, where Rn is the energy of your spectral distribution integrated over the spectrum, multiplied by that curve.
00:23:18 <zzo38> And then, it also depend on, whether it is screen or printer, you need different colors for making accurate!
00:23:48 <fizzie> kallisti: You can't do that with a distribution that's a Dirac delta at particular wavelength, multiplied by an amplitude.
00:24:52 <fizzie> For the same reasons as you can't span the R^3 space with just two R^3 vectors.
00:26:04 <fizzie> (If all your energy is at one wavelength, R1=k1*a, R2=k2*a, R3=k3*a, where a is your amplitude and k1, k2, k3 are values of those curves at your wavelength.)
00:27:27 <elliott> What's that sendmail-providing package people use?
00:27:29 <elliott> That doesn't provide anything else.
00:27:38 <kallisti> "because there can be more than one input wavelength" would have been sufficient I think. :P
00:27:39 <fizzie> (Okay, the above might not be quite mathematically accurate and I'm sure with some curve shapes (like if two curves are exactly the same) you could get all responses with one peak that you can with an arbitrary distribution, but never mind that.)
00:27:49 <elliott> There's ssmtp, msmtp... I can't figure out which one to use. :p
00:28:17 <fizzie> elliott: "ssmtp" is the one really small one that just provides a "sendmail" binary which speaks directly to a SMTP server.
00:28:44 <fizzie> ISTR that "msmtp" might do pretty much the same thing.
00:28:55 <elliott> The following packages will be REMOVED:
00:28:56 <elliott> exim4-base exim4-config exim4-daemon-light
00:29:02 <elliott> If I have those, shouldn't I have sendmail(1)?
00:29:25 <fizzie> Ye-s, I'd think. But maybe it's in exim4-something.
00:29:39 <fizzie> Also it's in some weird place always.
00:30:17 <elliott> Exim is a Mail Transfer Agent. It is normally called by Mail User Agents,
00:30:17 <elliott> not directly from a shell command line. Options and/or arguments control
00:30:17 <elliott> what it does when called. For a list of options, see the Exim documentation.
00:30:27 <elliott> fizzie: So, is it plausible that PHP mail() fails because it can't find that sendmail?
00:30:48 <fizzie> I... guess, though sbin is not really a strange place.
00:31:09 <fizzie> It's where the Postfix sendmail combattability wrapper is, too.
00:31:09 <elliott> Where the sendmail program can be found, usually /usr/sbin/sendmail or /usr/lib/sendmail. configure does an honest attempt of locating this one for you and set a default, but if it fails, you can set it here.
00:31:35 <fizzie> I seem to recall PHP could also deliver by smtp by itself, but I could be wrong.
00:31:46 <elliott> http://jann.is/daily/archives/741-Cannot-send-mail-with-PHP-and-exim.html
00:32:16 <elliott> The -i option prevents a line containing just a dot from terminating the message. Only an end-of-file (generated by typing CTRL-D if the input is from a
00:32:30 <fizzie> Well, they're all sendmail-"compatible".
00:32:43 <fizzie> In that they ignore a bazillion options.
00:32:59 <elliott> Esolang could not send your confirmation mail. Please check your e-mail address for invalid characters.
00:32:59 <elliott> Mailer returned: Unknown error in PHP's mail() function.
00:33:32 <fizzie> ssmtp needs some global configuration; namely, the smtp server to speak to.
00:33:45 <elliott> I just want something that connects to a server and blabs mail at it.
00:35:20 <fizzie> Do you mean a specific outgoing-mail server (that would be ssmtp), or one that actually does MX lookups and tries to deliver to the proper servers and keeps queues since they're down and so on (that would be Exim, Postfix, sendmail, any "real" MTA)?
00:35:55 <elliott> The latter. But I don't care about the queue thing.
00:36:17 <elliott> Postfix is a pain to set up, right?
00:36:48 <fizzie> I don't think there's really any middle ground there; if you want a thing that does MX lookups and falls back to secondaries and so on, it's going to be a "real MTA".
00:37:41 <fizzie> The Debian default Postfix configs (it's got about five "templates" of which you can select, at least with dpkg-reconfigure) I think should get things mostly right.
00:37:50 <fizzie> But Exim's the "default" one.
00:38:37 <fizzie> They will all accept "local" mail, though, they're like that.
00:39:05 <elliott> I don't want it to accept mail.
00:40:11 <fizzie> If someone sends to postmaster@your1234.linode.com (or whatever the IP was), it'll get deposited to /var/spool/mail/root or somesuch, by the default configs. At least that's my guess.
00:40:13 <elliott> Removing -i appears to have helped exactly nothing.
00:40:24 <elliott> fizzie: I thought one of the "default" configs stopped that.
00:40:44 <elliott> Exim can be set up for "local delivery only; not on a network".
00:41:02 <fizzie> That's pretty much the opposite.
00:41:11 <elliott> Oh. So is that why this isn't working?
00:41:14 <elliott> That's what it defaults to.
00:41:15 <fizzie> Then it'll only let you sendmail to local accounts.
00:41:23 * elliott chooses internet site; mail is sent and received directly using SMTP
00:42:01 <elliott> │ IP-addresses to listen on for incoming SMTP connections: │
00:42:05 <elliott> fizzie: What if I just set this to the empty string?
00:42:20 <fizzie> Localhost like that could be good enough.
00:42:22 <elliott> If it listens on 127.0.0.1, the internet won't see it, will it?
00:42:30 <fizzie> Then smtp-to-localhost will still work.
00:42:42 <fizzie> M'k, well, that should work.
00:42:56 <elliott> ┌───────────────────────────────────────────────────────────────────┤ Mail Server configuration ├───────────────────────────────────────────────────────────────────┐
00:42:57 <elliott> │ The Debian exim4 packages can either use 'unsplit configuration', a single monolithic file (/etc/exim4/exim4.conf.template) or 'split configuration', where the │
00:42:58 <elliott> │ actual Exim configuration files are built from about 50 smaller files in /etc/exim4/conf.d/. │
00:43:05 <elliott> │ Unsplit configuration is better suited for large modifications and is generally more stable, whereas split configuration offers a comfortable way to make │
00:43:08 <elliott> │ smaller modifications but is more fragile and might break if modified carelessly. │
00:43:13 <elliott> │ A more detailed discussion of split and unsplit configuration can be found in the Debian-specific README files in /usr/share/doc/exim4-base. │
00:43:20 <elliott> │ Split configuration into small files? │
00:43:28 <elliott> Good god, it is difficult to express how little I care, Exim.
00:43:56 <fizzie> (Though since your server shows up in the "Received:" headers, some 1970s-era people might assume that postmaster@yourbox can be contacted for abuse reports.)
00:44:39 <elliott> "This allows the site administrator to specify an email address that the wiki can contact if something goes wrong. Current versions of MediaWiki (1.13 and above) do not use this setting, having delegated its uses to $wgPasswordSender, but future versions or extensions might."
00:45:01 <elliott> People emailing esolang@linode-crap will not get a reply, but... too lazy to fix this.
00:49:50 <fizzie> Oh, you might still get some local mail somewhere, since when exim gives up trying to deliver someone's registration mail, it'll probably try to make a bounce.
00:50:50 <fizzie> (Maybe not very noticeable volumes.)
00:57:54 -!- elliott has quit (Ping timeout: 244 seconds).
00:59:19 <fizzie> Okays, slep now, have some bloodletting scheduled in five hours or so.
00:59:25 <zzo38> Something I wonder: What is the best compression scheme, which is not too complicated, for sokoban levels?
01:01:57 <shachaf> Why do you need to care about compression for sokoban levels?
01:02:21 <zzo38> Actually I am just thinking about it; it is not entirely very important.
01:07:26 <tswett> Say. Would anyone happen to know if the Thue-Morse sequence can be found by sampling periodically from a periodic function?
01:07:42 <tswett> Wait, of course it can.
01:08:45 <tswett> Can it be found by sampling periodically from... a periodic, Riemann-integrable function?
01:08:46 -!- calamari has joined.
01:22:17 -!- cheater_ has joined.
01:26:07 -!- cheater has quit (Ping timeout: 276 seconds).
01:49:26 -!- Frooxius has quit (Read error: Connection reset by peer).
01:49:49 -!- Frooxius has joined.
02:05:24 -!- augur has quit (Remote host closed the connection).
02:15:50 -!- Frooxius_ has joined.
02:17:28 -!- Frooxius has quit (Ping timeout: 244 seconds).
02:17:29 -!- Frooxius_ has changed nick to Frooxius.
02:20:08 -!- Frooxius_ has joined.
02:21:44 -!- Frooxius has quit (Ping timeout: 248 seconds).
02:21:46 -!- Frooxius_ has changed nick to Frooxius.
02:24:18 -!- elliott has joined.
02:29:54 -!- Frooxius_ has joined.
02:31:32 -!- Frooxius has quit (Ping timeout: 260 seconds).
02:31:36 -!- Frooxius_ has changed nick to Frooxius.
02:40:21 <elliott> Is anyone with IPv6 present?
02:41:29 <elliott> Does http://[2a01:7e00::f03c:91ff:fedf:9fdd]/wiki/Main_Page work correctly?
02:45:49 <shachaf> elliott: Oh, I only have it on a VPS.
02:45:55 <shachaf> What do the brackets mean?
02:47:49 <shachaf> Oh, they're used to disambiguate the port : or something?
02:48:14 <elliott> Just use curl or whatever.
02:49:18 <shachaf> Connecting to 2a01:7e00::f03c:91ff:fedf:9fdd:80... failed: Connection refused.
02:50:36 <elliott> "To attach an IP to your interface, issue the following command, making sure to replace the example IP with one of your pool addresses:
02:51:13 <elliott> OK, I should write the email to Graue.
02:51:40 <shachaf> elliott: Well, ping6 worked...
02:52:00 <elliott> http://esoteric.voxelperfect.net/wiki/800_numbers
02:52:14 <elliott> Probably I need to fiddle with nginx configuration.
02:52:38 <shachaf> I can't get it to accept a connection on any port.
02:53:11 <shachaf> I was thinking 800 numbers was an esolang.
02:59:36 -!- amca has joined.
03:05:18 -!- H3LLB0Y has quit (Ping timeout: 260 seconds).
03:05:49 -!- H3LLB0Y has joined.
03:06:27 <zzo38> How I have encoded a sokoban game is basically like this: The level is encoded as a stream of nybbles. First encode the size and the starting position (the size is offset by the minimum). And then encode walls using RLE but only the length; it automatically alternates wall/floor, and the player's position is skipped. And then, for the floors (including the player), encode RLE of the targets.
03:07:20 <zzo38> After that, the number of crates is already known because it equal number of targets. Encode number of empty spaces to the next crate, skipping walls, player, and spaces where a crate would get stuck. And that is finish.
03:08:41 <zzo38> How well would you think this work?
03:09:58 <zzo38> The edge of the board is automatically filled with walls and is not part of the encoding.
03:16:06 <elliott> Guses what I have ready to send?
03:19:35 <elliott> Gregor: Oioi. calamari: oi.
03:20:10 <elliott> monqy: You have an Esolang account, yes?
03:20:22 <shachaf> Why does Gregor get six times as many Ois as pikhq?
03:20:36 <monqy> elliott: yes its monqy
03:20:52 <elliott> monqy: Right. Unfortunately you've made two (2) non-spam-revert edits so I can't use you.
03:21:18 <monqy> oh right a talk page thing probably
03:21:21 <elliott> http://esolangs.org/w/index.php?title=User_talk:Billlam&diff=prev&oldid=23812
03:21:35 <elliott> http://esolangs.org/w/index.php?title=User_talk:Billlam&diff=next&oldid=23813
03:22:41 <monqy> http://esolangs.org/w/index.php?title=User:Monqy&oldid=22370 monqys proudest edit
03:23:16 <elliott> Why aren't those oi'd people awake?
03:23:28 <shachaf> Why didn't you Oi people who were awake?
03:23:38 <zzo38> Why do you need to know the number of non-spam-revert edits?
03:23:51 <shachaf> elliott: Did you know irssi only beeps when your nick is mentioned *right at the beginning* of a line?
03:24:55 <zzo38> shachaf: My client doesn't beep at all unless you tell it to; and if you do tell it to, you can configure your own pattern. Maybe irssi has a similar feature?
03:25:28 <shachaf> I'm talking about the default.
03:26:25 <zzo38> I want to figure out compressed sokoban encoding so that I can figure out how many will fit in various media, such as simple handwritten codes, QR codes, short radio signals, NES/Famicom cartridges, DVD, etc
03:26:33 <elliott> http://sfbay.craigslist.org/sfc/cpg/2854040585.html
03:26:37 <elliott> "To apply (required step):
03:26:37 <elliott> Please enclose your favorite one-liner in your favorite esoteric language :)"
03:26:40 <zzo38> And then can do similar thing for more complicated games too, such as tsume shogi
03:26:41 <elliott> shachaf: Y Combinator, man.
03:28:24 <shachaf> elliott: Hey, maybe I should go work there.
03:28:30 <shachaf> Except I hate all esoteric languages equally.
03:28:36 <shachaf> So I guess it's pointless.
03:29:14 <elliott> http://esolangs.org/wiki/Underload
03:29:23 <monqy> is it possible to hate underload
03:32:35 <shachaf> I hate all esoteric languages equally.
03:32:52 <elliott> What if Underload wasn't esoteric?
03:33:05 <shachaf> All programming languages are esoteric.
03:36:14 <shachaf> elliott: Did you double-check to make sure that the email doesn't consist of the words "HEY GRUEWAHTVERF I HATE U"?
03:36:31 <shachaf> That would be the wrong thing to write.
03:36:50 -!- Nisstyre has quit (Read error: Connection reset by peer).
03:37:24 <shachaf> elliott: You're well-known for telling people you don't like them very much, after all.
03:37:49 -!- Nisstyre has joined.
03:45:26 <elliott> calamari: Too late! But hi.
03:45:53 <elliott> shachaf: You sure do like YAML.
03:46:05 <elliott> calamari: (I was trying to find more people to support my presidential run.)
03:46:31 <shachaf> elliott: I don't like it as much as a certain other person.
03:47:28 <calamari> and by too late, does that mean you lost already?
03:47:51 <elliott> By "presidential run", I mean "server migration plan".
03:49:21 <calamari> ron paul is on the way out so looks like I'll be voting LP this primary
03:50:45 <calamari> don't feel like switching to republican party for the primary when he's already lost
03:53:08 <pikhq> elliott: You could've assumed my support by default. :P
03:54:23 <elliott> Yes, but then what if he came into #esoteric and went "pikhq!! DID YOU EXPLICITLY SUPPORT HIM" and you'd be all "No" and he'd be all "I WILL STRIKE HIM DOWN"
03:54:32 <elliott> Graue is a very scary man.
04:07:57 -!- FireFly has quit (Ping timeout: 240 seconds).
04:27:28 <elliott> Hmph, cached pages aren't being sent gzipped.
04:36:54 <zzo38> I am making a text adventure system in Haskell. Here is some types is this good to you? type GameFunction = (Game, TranscriptItem) -> (String, SystemRequest); data SystemRequest = UserInput !String !Game | UserDirectInput !String !Game | RewindTo !Int (Game -> Game) | RequestRandomNumber !Int !Game | GameOver;
04:38:49 <zzo38> Should it be designed differently?
04:48:07 -!- TeruFSX has joined.
05:18:17 -!- NihilistDandy has quit.
05:29:24 -!- azaq23 has quit (Quit: Leaving.).
05:39:33 -!- augur has joined.
05:39:51 -!- evincar has joined.
05:46:04 <evincar> Don't everybody talk at once.
05:50:51 <evincar> The people in #concatenative had mixed reactions to an article I wrote about concatenative languages.
05:51:11 <evincar> Wasn't the best thing I've ever written...
05:51:38 <monqy> you didn't embarrass them did you
05:51:42 <evincar> I dunno, it's embarrassing. Bask in my embarrassment.
05:51:51 <elliott> Things you write have differing levels of badness?
05:51:58 <evincar> I got new people to come to their channel.
05:52:01 <elliott> We saw it, by the way. Unfortunately.
05:52:18 <evincar> I go from mediocre to quite terrible.
06:08:14 -!- calamari has quit (Quit: Leaving).
06:10:28 -!- augur has quit (Remote host closed the connection).
06:10:36 -!- H3LLB0Y has left.
06:11:11 -!- H3LLB0Y has joined.
06:16:02 -!- H3LLB0Y has quit (Ping timeout: 260 seconds).
06:17:52 <Sgeo> How easy/difficult would it be to make a Smalltalk VM for LSL?
06:18:23 <Sgeo> Although I can't even begin to envision how to begin building anything like an IDE
06:18:40 <monqy> is that the second life thing? heh heh heh
06:18:53 <fizzie> Weren't you doing [insert language here] for LSL?
06:19:31 <monqy> you could do a concatenative language!!!!!
06:32:04 <evincar> They are all the rage, don't you know.
06:34:02 <elliott> by the way, "concatenative language" has a concrete definition.
07:01:42 <evincar> Look, I could have done a better job on that article.
07:01:54 <evincar> I had spent a while writing it and got fed up with it.
07:02:14 <evincar> So I just posted it, submitted it to HN, and waited to see what happened.
07:02:30 <evincar> What happened is way more people read it than I'd expected.
07:06:49 <elliott> just thought you should know
07:07:59 <evincar> Alright, explain it to me.
07:08:22 <evincar> If I really ought to know.
07:08:31 <evincar> Learning new things is great.
07:10:21 <elliott> a language in which syntactic composition is semantic composition. "A concatenative programming language is a point-free programming language in which all expressions denote functions and the juxtaposition of expressions denotes function composition." --Wikipedia or more formally (from [[Joy]]): "the meaning function is a homomorphism from the syntactic monoid onto the semantic monoid"
07:11:47 <elliott> for instance you can turn a language with f(x) style application into one with f(g) style composition but it would not be concatenative.
07:12:15 <evincar> That's a good way of putting it.
07:12:48 -!- zzo38 has quit (Remote host closed the connection).
07:17:51 <Sgeo> evincar, your article
07:18:13 <evincar> Ugh, fucking http://evincarofautumn.blogspot.com/2012/02/why-concatenative-programming-matters.html
07:19:25 <Sgeo> Oh, that's yours?
07:20:35 * Sgeo blames evincar for his newfound interest in Smalltalk (because I was looking at Factor thanks to that article, then Slava apparently stopped doing Factor stuff)
07:20:52 <evincar> Yeah, I just found out about that from the comments on HN.
07:21:04 <evincar> Apparently Factor development is to cease.
07:22:04 <evincar> Which I guess will be good for marketing Prog, if people end up looking for "the next concatenative language".
07:22:04 <Sgeo> I wonder if the reason that I like Pharo and its IDE so much might have more to do with me perceiving more mainstream IDEs as bloated
07:22:27 <evincar> "My current language project."
07:22:39 <evincar> Even though I sound like such an ass whenever I say that now.
07:22:41 <monqy> is that that language
07:22:47 <evincar> No it is not that language.
07:22:50 <evincar> It is a different language.
07:22:56 <monqy> what happened to that language :'(
07:22:58 <evincar> But hey, at least there's only one right now.
07:23:13 <Sgeo> " Built-in functions and flow-control statements are actually operators."
07:23:18 <evincar> The Smalltalky one with weird compositional semantics?
07:23:30 <monqy> every language of you'rs youv'e presented in here
07:23:40 <Sgeo> Misread that at first, I think.
07:23:58 <elliott> i don't think there is any indication factor development will cease completely.
07:23:59 <Sgeo> I do appreciate flow-control being actually ... library-like, but built-in functions are operators?
07:24:04 <Sgeo> That's .. a bit new to me
07:24:08 <evincar> Sgeo: I dunno what you're quoting from. Probably an old article on my blog that sucks?
07:24:14 <Sgeo> http://prog.sourceforge.net/cgi-bin/SyntaxOverview
07:24:15 <monqy> also what does "built in functions and flow-control statements are actually operators" even mean or do i not want to know
07:24:28 <Sgeo> Or am I looking at the wrong Prog?
07:24:30 <evincar> Ugh fuck don't look at that repo.
07:24:39 <evincar> Same name for a different thing.
07:24:50 <monqy> maybe you should pick a less already used name
07:25:05 <Sgeo> "Some types like to "decay" into other types. Any operation on a value of a radioactive type R results in an R containing a closure describing the operation. "
07:25:16 <monqy> that reminds me of what was that amazing language
07:25:42 <evincar> Radioactive types != lazy evaluation, if that's what you mean.
07:25:46 <coppro> http://www.amazinglanguage.com/
07:25:47 <monqy> http://alefpp.sourceforge.net/
07:25:56 <Sgeo> evincar, so it IS your thing?
07:26:14 <monqy> evincar made two things called prog?????
07:26:17 <evincar> Sgeo: I created that SourceForge project ages ago, yeah.
07:26:29 <monqy> you're your own worst enemy
07:26:45 <evincar> monqy: I decided Prog shouldn't be the ill-defined thing it was, and redefined it, borrowing the good stuff from the original.
07:26:55 <monqy> elliott: don't you remember??
07:26:59 <monqy> elliott: it's Alef++!
07:27:07 <Sgeo> "Alef++, has a crazy syntax."
07:27:28 <Sgeo> That's... never the most advertised thing on any language that is not an esolang.
07:27:31 <Sgeo> Pretty much ever.
07:27:45 <Sgeo> Sure, some languages have unusual seeming syntax, but that's usually not a selling point, is it?
07:27:52 <monqy> it's the closest i've found to the sourcereal of programming languages
07:27:56 <evincar> I think he means "crazy" in the sense that preteens mean "random".
07:28:20 <monqy> (my life goal is to find the sourcereal of programming languages)
07:31:05 <evincar> Sigh. No sense of comedic timing.
07:32:06 <evincar> In any case, the "hurr durr operators" thing was because the original thing that was called Prog could be parsed with an operator precedence parser.
07:32:44 <evincar> So everything built-in was an operator.
07:34:48 <monqy> eliot president of esolangs
07:40:20 <fizzie> I wonder if they already have "gets() 19xx-2011 never forget" T-shirts available.
07:40:41 <elliott> fizzie: Those afraid nameservers are still down, right?
07:41:04 <elliott> http://freedns.afraid.org/news/
07:41:05 <fizzie> elliott: Bad news, the email was just a string of profanities?
07:41:31 <fizzie> They said things are looking up, or something, when I checked in the morning.
07:41:48 <elliott> Good news, i.e. he thinks the plan is a good one.
07:42:20 <fizzie> You figured that out from the string of profanities? Good decoding. (I mean, I suppose this is one of those good-news-bad-news-good-news-bad-news routines.)
07:42:52 <elliott> They were very articulate profanities!
07:42:57 <fizzie> $ dig @ns4.afraid.org esolangs.org a | grep '^eso.*IN.*A'
07:42:57 <fizzie> esolangs.org. 3600 IN A 207.7.108.149
07:43:00 <quintopia> fizzie: like the chef sketch in the first key and peele. <3 that sketch.
07:43:04 <fizzie> It seems to work-ish. But ns3 doesn't.
07:43:19 <fizzie> So there might be delays and such.
07:43:44 <fizzie> (Dig's so verbose, it's annoying to paste.)
07:44:45 <elliott> There's that nslookup thing.
07:46:56 <fizzie> And the BIND9-utils 'host' tool, that's not too bad.
07:47:03 <ion> dig +short
07:47:24 <fizzie> Well, that's very short.
07:48:09 <fizzie> Isn't there any "+short-ish-but-not-quite-that-short-i-mean-like-one-line-but-maybe-still-repeat-the-name-and-type-or-something"?
07:48:27 <ion> dig +noall +answer
07:48:58 <fizzie> It's also shorter than my proposed name.
07:49:08 <elliott> fizzie: Why did you want a MW parser, by the way?
07:49:45 <fizzie> Oh, I didn't, now; I just recall wanting one a couple of years back and things weren't... good. Not sure what for.
07:50:17 <elliott> fizzie: That's a shame, I was going to bet you I could write a fully-compatible one.
07:51:05 <fizzie> Like you mean 32 megabytes or something!? (Are the caches in fact even larger nowadays? I really haven't been following.)
07:51:07 <ion> How much is huge cache * cash * cache? Is huge cache * cache = huge cache²?
07:51:19 <elliott> fizzie: They've been shrinkin'.
07:51:40 <elliott> i7 has 256-k-per-core-or-so and 8 meg L3.
07:51:50 <elliott> Yes, seems it's the same now.
07:51:59 <fizzie> "cache size : 6144 KB" on this old thing.
07:52:10 <fizzie> So I guess 32 megs is quite many megs.
07:52:29 <elliott> ion: Well, obviously juxtaposition is multiplication. So it's a huge cash cache^2.
07:52:33 <elliott> A cache of caches of cash.
07:52:46 <elliott> So if you store all your guns and money somewhere, and then have a big storage of those...
07:53:13 <ion> It’s caches all the way down.
07:53:44 <pikhq> Some recent Xeons go up to 30MiB L3.
07:54:09 <pikhq> And, yes, that's "really huge cache"
07:55:02 <elliott> Eventually we'll just have terabyte L1s or something.
07:55:12 <fizzie> Xeon E7-8867L, ten cores, 10x256k L2, 30M L3, I see.
07:55:25 <pikhq> fizzie: And up to 8 chips supported. :)
07:55:53 <pikhq> elliott: Yeah, but then we'll have petabyte RAM and still 200MHz RAM clock.
07:57:21 <fizzie> They installed a couple (16?) Nvidia Tesla M2090 cards to our cluster for testing, those are supposedly pretty spiffy.
07:58:06 <fizzie> Also two "fat nodes" with 1T of RAM each; I wonder if they feel bad about being called that.
08:00:50 <fizzie> The old ones were some kind of 12-core (2*6) Opterons; these new are some kind of Xeons, but I see no-one's bothered to update the documentation yet.
08:02:33 <fizzie> Ah, it's also 2*6 of Xeon X5650.
08:04:08 -!- augur has joined.
08:19:23 -!- cheater_ has quit (Read error: Connection timed out).
08:19:51 -!- cheater_ has joined.
08:23:20 <fizzie> "Footage has finally surfaced of Dean Dinnen's chainsaw-assisted attack at The Endyke Pub in Hull. Wasted and stoned out of his mind, the 24-year-old got sensitive after being told he couldn't smoke inside of the bar. Drowning in liquor and his feelings, he went back to his crib to grab his chainsaw.
08:23:25 <fizzie> Dinnen returned to the pub, wielding the weapon in a substance-fueled raged. Other customers tossed bar stools, kegs, and pool cues at him, but that didn't stop him from tearing through 32-year-old Andrew Pryor's arm."
08:24:00 <fizzie> In here people just use axes.
08:25:55 <fizzie> You should have slept and woken up, that's what I didded.
08:26:08 <ion> In Finland, the people tossing bar stools etc. would probably get fined or jailed for assault and battery.
08:29:34 <elliott> One wonders why he had a chainsaw in the first place.
08:30:32 <ion> Related (in Finnish, sorry): http://www.aamulehti.fi/Kotimaa/1194722981981/artikkeli/il+mies+hyppasi+liikkuvasta+taksista+kuljettajalle+vaadittiin+tuomiota+kuolemantuottamuksesta.html
08:30:34 <fizzie> Zombie survivalists all have.
08:32:10 <elliott> ion: Oh no, don't tell me you're Finnish too.
08:32:25 <fizzie> "The district court sentenced the driver only traffic from exile." Oh, Google Translate.
08:33:18 <ion> elliott: Ok, i won’t.
08:35:46 -!- amca has quit (Quit: Farewell).
08:36:54 <elliott> Man, I bet Graue is sleeping right now. Why amn't I?
08:50:45 -!- evincar has quit (Quit: Sleep.).
09:02:06 -!- monqy has quit (Quit: hello).
09:03:26 -!- Phantom_Hoover has joined.
09:16:02 -!- elliott has quit (Ping timeout: 245 seconds).
09:35:54 -!- oerjan has joined.
09:45:52 <oerjan> wow it's really happening
09:58:53 <oerjan> @tell kallisti <kallisti> but do we delineate an infinite spectrum of colors? <-- well relativity basically means it cannot be any more quantized than length itself, and no one has found any evidence length is quantized, although i think there are theories that it may happen at or near planck length scale
10:14:42 <oerjan> <tswett> Can it be found by sampling periodically from... a periodic, Riemann-integrable function?
10:23:50 <oerjan> ^ul ((0)(1)):^!S(~:^:S*a~^~*a*~:^):^
10:23:50 <fungot> 011010011001011010010110011010011001011001101001011010011001011010010110011010010110100110010110011010011001011010010110011010011001011001101001011010011001011001101001100101101001011001101001011010011001011010010110011010011001011001101001011010011001011010010110011010010110100110010110011010011001011010010110011010010110 ...too much output!
10:24:16 * oerjan things his fingers might be starting to remember that
10:26:42 <fizzie> oerjan: Doesn't UP say something about measuring the energy (i.e. wavelength) of photons? ∆E∆t > one of the h's, or something.
10:27:07 <oerjan> yes. but that doesn't mean those are quantized separately.
10:29:41 <oerjan> basically, relativity means you can always find a different frame of reference in which the energy is whatever value you want above the minimum, so it cannot have a discrete set of values for particles moving freely
10:30:52 <fizzie> Sure, sure, but the point was that can we differentiate between systems A and B which e.g. emit photons at wavelength x and x+e for arbitrarily small e?
10:31:19 <fizzie> (Though I think the actual point was whether unaided humans can, with our built-in sensory apparatus.)
10:31:44 <oerjan> well iiuc, if we use long enough time for the measurement the ∆t part can become as large as we wish...
10:32:44 <oerjan> basically, the UP doesn't restrict the precision for just a single of the observables
10:32:49 <fizzie> But if it's a finite time.
10:36:11 <fizzie> Anyway, I don't think ∆t is just how long we spend for measuring things. But I'm very much not a physicist.
10:41:52 <oerjan> <Sgeo> How easy/difficult would it be to make a Smalltalk VM for LSL? <-- so let me guess, my joke about you using LSL instead of Smalltalk for whatever it was you were doing has now mutated.
10:42:22 <oerjan> oh right, implementing haskell
10:42:35 <oerjan> so, haskell in a smalltalk vm in lsl. check.
10:42:37 -!- tzxn3 has joined.
11:05:29 -!- nooga has joined.
11:18:46 -!- H3LLB0Y has joined.
11:21:29 <kmc> http://dotsies.org/ cool, another programmer reinventing the alphabet without even a basic knowledge of how human vision works
11:27:04 -!- Taneb has joined.
11:29:05 -!- Taneb has quit (Remote host closed the connection).
11:29:22 -!- Taneb has joined.
11:30:16 <Taneb> A scary thought occurs
11:30:31 <Taneb> I am younger than the proof of Fermat's Last Theorem
11:32:06 <Taneb> Also, Jackson has space for 72 more commands
11:44:48 <oerjan> i'm sorry, that hello took far too long. you are clearly a Taneb impostor.
11:47:50 <Taneb> Curses, foiled again.
11:48:00 <Taneb> For I am secretly elliott in disguise.
11:48:48 * oerjan suddenly gets a vision of a villain trying to rob a tinfoil factory and accidentally getting wrapped up in the machinery
11:48:56 <Taneb> But yeah, I just had things on my mind
11:51:53 -!- Taneb has quit (Remote host closed the connection).
11:52:46 -!- Taneb has joined.
11:53:29 <Taneb> oerjan: this better?
11:54:43 <oerjan> i'm afraid you are still foiled.
11:56:08 <oerjan> i wonder if it would be a good idea to plant a rumor that tin^Walumini?um foil has mind control chemicals added
11:56:39 <Taneb> That could have hilarious consequences
11:57:00 <Taneb> Although it took me a second to think what Walumini was
11:57:21 <kmc> http://berkeley.intel-research.net/arahimi/helmet/
11:58:19 <kmc> "Using a $250,000 network analyser, we find that although on average all helmets attenuate invasive radio frequencies... certain frequencies are in fact greatly amplified. These amplified frequencies coincide with radio bands reserved for government use according to the Federal Communication Commission"
12:08:00 <Taneb> I'm looking at the photos of the UV Rave the other night.
12:08:09 <Taneb> I look like a flourescent chemistry zombie.
12:10:41 <fizzie> There's a rebuttal to that, too.
12:11:06 <fizzie> http://zapatopi.net/blog/?post=200511112730.afdb_effectiveness
12:11:30 <fizzie> "-- there are serious flaws in this study, not the least of which is a complete mischaracterization of the process of psychotronic mind control. I theorize that the study is, in fact, NWO propaganda designed to spread FUD against deflector beanie technology, and aluminum shielding in general, in order to disembeanie paranoids, leaving them open to mind control."
12:13:21 <ion> Where are buzzie and fizzbuzzie?
12:13:36 <kmc> probably interviewing for jobs
12:13:48 -!- H3LLB0Y has quit (Quit: Leaving.).
12:13:55 <ion> At a blackboard
12:14:35 <oerjan> fizzbuzzie only shows up every second week or so
12:19:53 <Taneb> buzzie only once a twice a week
12:22:58 -!- Taneb has quit (Quit: lunch).
12:31:52 -!- oerjan has quit (Quit: leaving).
12:38:55 -!- Taneb has joined.
12:44:34 <Phantom_Hoover> Well, today's xkcd did have a fairly funny joke, as evidenced by the fact that it was funny when Black Books did it a decade ago.
12:58:23 <itidus20> kmc: yeah i think you really need to engineer such helmets based on known weaknesses of mind-control technology.. not troll-physics
13:01:22 <Taneb> That is a real scientific field.
13:05:20 <fizzie> But it doesn't have anything to do with spiders. :/
13:05:33 <Taneb> That would be Arachnoacoustics
13:05:48 <fizzie> Well, anyone doing that?
13:19:13 -!- Taneb has quit (Ping timeout: 252 seconds).
13:30:44 -!- Taneb has joined.
13:31:43 -!- augur has quit (Remote host closed the connection).
13:38:18 -!- nooga has quit (Ping timeout: 260 seconds).
13:54:35 -!- Taneb has quit (Remote host closed the connection).
13:57:39 <fizzie> Sometimes it's a bit confusing.
13:59:28 -!- Taneb has joined.
14:05:31 -!- Taneb has quit (Remote host closed the connection).
14:05:38 <lambdabot> http://code.haskell.org/lambdabot/COMMANDS
14:09:40 -!- Phantom_Hoover has quit (Remote host closed the connection).
14:10:52 -!- Phantom_Hoover has joined.
14:27:58 <fizzie> Only up to edit distance of two, sorry.
14:28:14 <fizzie> And as long it's unambiguous.
14:41:49 -!- augur has joined.
15:17:38 -!- Vorpal has joined.
15:19:02 <tswett> la_fen: tell elliott pong
15:23:50 -!- Taneb has joined.
15:25:39 <Taneb> If we define a function that I will narcisistically name Taneb_2 (n) such that Taneb_2 (n) is the lowest power of 2, x, such that n! mod x is 0.
15:26:45 <Taneb> This goes 0,1,1,3,3,4,4,7,7,8,8,10,10,11,11...
15:26:59 <Taneb> @oeis 0 1 1 3 3 4 4 7 7 8 8 10 10
15:27:04 <lambdabot> n minus (number of 1's in binary expansion of n). Also highest power of 2 di...
15:27:04 <lambdabot> [0,0,1,1,3,3,4,4,7,7,8,8,10,10,11,11,15,15,16,16,18,18,19,19,22,22,23,23,25,...
15:27:37 <Taneb> @oeis 0 0 1 1 1 2 2 2 4 4 4 5 5 5
15:27:47 <lambdabot> [0,0,0,1,1,1,2,2,2,4,4,4,5,5,5,6,6,6,8,8,8,9,9,9,10,10,10,13,13,13,14,14,14,...
15:28:37 <fungot> Taneb: it's an infinite list to me.). it would be a fnord? hehe.
15:28:42 <fungot> Available: agora alice c64 ct darwin discworld europarl ff7 fisher fungot homestuck ic irc* iwcs jargon lovecraft nethack pa qwantz sms speeches ss wp youtube
15:28:50 -!- nooga has joined.
15:29:01 <fungot> Selected style: jargon (UNIX-HATERS mailing list archive)
15:29:09 <Taneb> fungot, give me advice
15:29:10 <fungot> Taneb: many existing implementations or add unnecessary complexity to merit a higher level programming interface consider sendmail.cf. perhaps, for example, are we always destined to reinvent the same port.
15:29:50 <oklofok> so here i am reading my morning nonsense from #philosophy
15:30:03 <oklofok> and i accidentally click on #esoteric
15:30:26 <oklofok> and read fungot's line and i'm like finally, something that makes sense.
15:30:26 <fungot> oklofok: what bothers me even more than 13 blocks long, at most cs labs, and
15:31:07 <oklofok> ah, that reminds me of Kant's thesis about the existence of 14 blocks long without an inherent substantial being.
15:32:05 <cheater_> oklofok: i also think that this
15:32:38 <Taneb> I had a really good formula forTaneb_2(n) but it doesn't account for multiples of four that are not a power of 2
15:33:25 <Taneb> The highest power of 2 dividing n!
15:34:11 <Taneb> Pending a better name
15:35:07 <cheater_> "power of 2 in the prime decomposition"
15:35:37 <fizzie> fungot: I think that was an infinite list to everyone.
15:35:37 <fungot> fizzie: would anyone be interested in running restricted mailing-list software that caused all our software even more than ten minutes. mailing lists were expected to follow in a
15:39:14 <oklofok> Taneb: "dividing (n!)" and not "(dividing n)!" right?
15:39:39 -!- Taneb has quit (Ping timeout: 252 seconds).
15:40:06 <oklofok> i'll take that as a "FUCK MY HOUSE IS ON FIRE"
15:40:26 <kmc> highest power of 2 dividing! n
15:41:01 <fizzie> It doesn't just divide, it divides.
15:43:17 <itidus20> http://www.youtube.com/watch?v=6wJzo5v8wYQ
15:45:33 <oklofok> 2^k \emph{divides} n is a good way to say k is the highest power that divides n
15:51:52 <fizzie> \frac{\mbox{divi}}{\mbox{des}}.
15:58:31 <Gregor> @tell elliott Wow, it's almost like we're in distant timezones!
16:00:17 -!- calamari has joined.
16:00:20 <fizzie> lambdabot: Are you doing some
16:04:36 <Gregor> Oh, mail tracking. Apparently my package sat in Los Angeles, CA for a week for no reason, then teleported to Lafayette, IN for delivery.
16:08:48 <fizzie> Mail tracking, the leading cause of the "F5 finger" symptome.
16:09:42 <lambdabot> kallisti: You have 2 new messages. '/msg lambdabot @messages' to read them.
16:09:46 <calamari> my favorite is when the package was already delivered and I check the tracking and it says it wasn't
16:14:29 <fizzie> My favorite was when the tracking says the package was already given out of the post office, but I hadn't picked it up yet.
16:20:04 <Gregor> My favorite is when the package says it went to the wrong city, and is now out for delivery.
16:20:11 <itidus20> wikipedia's disambiguation page says "Function (mathematics), an abstract entity that associates an input to a corresponding output according to some rule"
16:20:26 <itidus20> this to me sounds like what is also called a process
16:20:50 -!- sebbu2 has changed nick to sebbu.
16:20:53 <Gregor> Proc{esses,ocedures} have side effects.
16:21:08 <Gregor> Of course, the definition of "side effect" is up to you.
16:21:24 <fizzie> Once they routed my from-the-next-city package into somewhere far up there in Northern Finland, and then said there has been an "irregularity".
16:21:28 <Gregor> Oh, also functions terminate :)
16:21:58 <itidus20> yeah i can't cheat this topic very easy
16:22:16 <Gregor> If your input is a machine state and your output is a machine state, then a procedure is a function ^^
16:23:33 <Gregor> fizzie: I recall some friend posting a screenshot of the tracking web site showing that their package was looping.
16:23:48 <Gregor> It had done something like Portland -> Eugene -> Portland -> Eugene
16:24:37 <fizzie> Quite a few languages also have terminology where a "procedure" is pretty much a "function" which just doesn't return a value.
16:25:18 <itidus20> fizzie: yeah but thats not what people come here to hear
16:26:38 <Gregor> Well, yeah, in programming languages the words all get blurry.
16:26:49 <Gregor> But the quote from above did say "Function (MATHEMATICS)" (caps added)
16:27:24 <itidus20> so is a function a state transition table?
16:28:16 <itidus20> maybe table isn't the right word here since i wouldn't really quite know what i am referring to
16:30:12 <itidus20> so i could have a machine with 2 states.. a and b.. and a function which accepts either a or b... and i could call this function identity and say a->a, b->b
16:31:21 <itidus20> but in this case the type of x would be a machine state instead of a function
16:32:21 <itidus20> and it all starts to fall apart and wishful thinking that it could be so easy
16:36:36 <itidus20> i think it is not such an easy topic :-D
16:37:39 -!- boily has joined.
16:48:21 <itidus20> it seems that probably the reason i fill my head with misconceptions is my urge to take shortcuts
16:51:14 <itidus20> kallisti: in any case i will read up on functions in my own time.. any other topic you have in mind?
16:54:15 <oklofok> itidus20: in mathematics, the difference between a function and a process, to me, is not that a process has side-effects (that's meaningless in math), it's that a process can be computed in some way (the "how" is left implicit)
16:55:12 <kallisti> itidus20: many kinds of state transitions can be described as functions, but functions themselves have nothing to do with state.
16:55:28 <Gregor> Well, I don't think the term "process" is meaningful in mathematics, but if we take the concept "function" more broadly, then you have to introduce state to be able to describe what a process is.
16:56:12 <kallisti> for example finite state automata have a state transition function : (states, alphabet) -> states
16:56:25 <kallisti> for example finite state automata have a state transition function : states x alphabet -> states
16:57:27 <oklofok> "<Gregor> Oh, also functions terminate :)" i don't think this is very relevant either
16:57:55 -!- zzo38 has joined.
16:57:59 <oklofok> and you can't really even talk about whether a function terminates. that's meaningless.
16:58:07 <kallisti> terminate implies a computation in the first place.
16:59:12 <kallisti> hm, you could probably determine if computing the result of a function from an input terminates.
16:59:30 <kallisti> but it's probably not always straightforward how to compute a function.
17:00:01 <fizzie> Not all functions are, you know, computable.
17:00:05 <oklofok> "<kallisti> terminate implies a computation in the first place." exactly
17:00:11 <itidus20> ok suppose our machine is a tictactoe board which ignores symmetry and whose turn it is. curiously a valid game of tictactoe is only those state transitions where only one tile on the board is changed from an empty square
17:00:19 <oklofok> perhaps Gregor was really talking about programming in the first place
17:00:38 <fizzie> oklofok: Give that it was about "processes" in the first place, that doesn't sound entirely surprising.
17:00:43 <itidus20> uhmmm.. not sure where i'm going with this
17:01:37 <fizzie> Though I suppose e.g. a stochastic process is a mathematical creature and has a definition for a "process".
17:02:06 <oklofok> fizzie: well i suppose you could say that a function is "computable" if it's on some level of the arithmetical hierarcy. that means it's computable as long as you can quantify over infinite subsets of N (afaiu). and so on.
17:02:10 <kallisti> oklofok: do people actually use monads when describing different kinds of computational processes in a mathematical setting? or do they mostly use the more "standard" models of computation like production systems, logical combinators, turing machines, and so forth.
17:02:30 <kallisti> oh, well, I guess monads would fall under "logical combinators":
17:02:49 <itidus20> im really a very slow thinker.. and i have inadvertently created chaos by these ambiguous definition topics. but i hope that it is a useful chaos for some of you.
17:03:23 <zzo38> What exactly do you mean by "logical combinators"?
17:04:09 <Slereah> Did someone say logical combinators?
17:04:13 <zzo38> And how to monads fall under it?
17:04:34 <kallisti> hmm, well I meant things like SKI and lambda calculus. But I guess LC isn't really combinatory
17:04:52 <oklofok> kallisti: well that's computer science / logic stuff, we don't use monads for anything.
17:05:03 <fizzie> oklofok: I suppose you could also say that a function is "computable" by referring to some particular model of computation, and don't those all pretty much end up being the same set of functions?
17:06:44 <fizzie> It's because computer scientists can't math, and math people don't like computers.
17:06:57 <oklofok> fizzie: yeah, but that's different in that the analytical hierarcy is a fuckload less computable than anything in the arithmetical hierarchy, which is a fuckload less computable than anything that an actual machine can compute.
17:07:23 <zzo38> fizzie: People in this channel might like both
17:07:39 <fizzie> zzo38: Those are then some kinda freaks.
17:08:08 <itidus20> maybe it is because i am too impartial that i can't figure things out
17:08:12 <zzo38> (It includes myself too)
17:08:56 <oklofok> kallisti: well i don't use L-functions either, and those are way more important than monads.
17:09:17 <itidus20> that is to say, from my point of view i have to take everyones comments as equally valid
17:09:34 <kallisti> oklofok: aren't L-functions not even proven or anything...
17:09:37 <fizzie> kallisti: Don't listen to oklopol, ain't no one who can tell you what you can or cannot do! Believe in yourself! And other such phrases!
17:10:08 <fizzie> I was so tempted to add a "
17:10:12 <fizzie> *vomits a rainbow*" at the end.
17:10:18 <itidus20> i think you mean don't_listen_to(x)
17:10:20 -!- nys has joined.
17:10:44 <oklofok> kallisti: yeah i don't think you can prove they have analytic continuations
17:11:03 <oklofok> erm wait a sec, i'll elaborate
17:11:38 <kallisti> it would be lame if they were disproven completely, and then all of that work will vanish.
17:11:44 <oklofok> (because what i said was wrong)
17:11:44 <kallisti> silly mathematicians are playing it risky. :>
17:12:04 <itidus20> for example, no matter how smart people are, nor how much consideration is put into writing laws... no legal system will ever be perfect
17:12:43 <itidus20> so all participants in the legal system have to sort of get on with their work and ignore the fact their system is fundamentally imperfect
17:13:12 <oklofok> kallisti: that work would be part of the work that proves they exist
17:13:19 <itidus20> and, someone who couldn't face this fact can't really enter the legal system
17:13:36 <oklofok> the proof of a theorem doesn't vanish because the theorem is reduced to "true"
17:13:45 <oklofok> well, naturally some of it would be just plain useless.
17:14:26 <itidus20> and in the same way, if i am too preoccupied with getting the right understanding than getting some understanding same thing really
17:16:06 <kallisti> oklofok: I was talking about the situation where it was disproven.
17:17:24 <kallisti> that the work going into proving the theorem
17:17:30 <oklofok> kallisti: well the whole theory could then be thought of as an unnecessarily proof by contradiction. "assume we can analytically continue this L series here"
17:17:33 <kallisti> becomes part of the theorem of its contradiction in that case.
17:17:46 <oklofok> "we could then develop this awesome theory and OH FUCK A CONTRADICTION"
17:18:24 <kallisti> ..there are proofs in my math homework that I wish I could prove my contradiction
17:18:38 <kallisti> but my professor is one of those where he won't let you use methods that haven't been discussed yet.
17:19:18 <kallisti> a very annoying aspect of math classes.
17:19:39 -!- derdon has joined.
17:19:49 <kallisti> it's not quite the same in programming, in my experience. Usually as long as you meet the assignment requirements you can use whatever language features and standard libraries you want.
17:21:43 <kallisti> for example, back in my intro Java class we were covering nested if-else statements, but to shorten my code I just used an array, even though we hadn't covered them.
17:21:49 <kallisti> and the professor even commented on it positively. :P
17:24:28 <oklofok> you can use any methods you like in pretty much all math classes in our uni
17:24:36 <oklofok> the prof will show an answer based on the course though
17:24:57 <kallisti> yeah my professor just nitpicky about certain things
17:25:06 <kallisti> he's probably the best math prof. at the university as far as I can tell.
17:25:44 <kallisti> for example, he doesn't like the re-use of the same variable in different, unrelated quantified statements.
17:26:04 <kallisti> obviously he's not a fan of local scope. :P
17:27:07 <kallisti> but he doesn't count off if you do that.
17:28:42 <kallisti> ...random thought. it would be awesome to see in, a math paper, a huge integral in parentheses
17:28:58 <kallisti> in a text portion of the paper
17:29:12 <kallisti> rather than using some TOTALLY LAME VARIABLE.
17:29:13 <fizzie> Are you sure he'll let you ever prove something by contradiction? You know, there are people who just don't think that's kosher at all.
17:29:42 <kallisti> well, he's teaching from the book, to the point he uses whatever notational conventions the book uses, even if he doesn't like them.
17:29:51 <kallisti> for example, our natural numbers start at 0.
17:30:34 * kallisti is, as a programmer, prone to off-by-one errors in every facet of daily life.
17:30:57 <kallisti> why would that not be acceptable.
17:31:06 <oklofok> "prone to off-by-one errors in every facet of daily life." who isn't
17:31:14 <kallisti> if you prove a statement is false, then its negation is necessarily true..
17:31:35 <fizzie> kallisti: See: http://en.wikipedia.org/wiki/Mathematical_intuitionism
17:31:39 <kallisti> oklofok: MAYBE PROGRAMMING HAS WARPED MY BRAIN
17:31:44 <fizzie> "A major force behind Intuitionism was L.E.J. Brouwer, who rejected the usefulness of formalized logic of any sort for mathematics. His student Arend Heyting postulated an intuitionistic logic, different from the classical Aristotelian logic; this logic does not contain the law of the excluded middle and therefore frowns upon proofs by contradiction."
17:32:04 <oklofok> seriously the hardest part about even things like writing proofs is to get the indices right, "err is this k or k-1... well whatever, you get the point".
17:32:10 * kallisti hasn't gotten to a stage in his math education where he ever thinks about intuitionistic logic.
17:32:57 <fizzie> oklofok: There's a maths professor at our place who starts with subscript indices, then they gradually rise up to be "middle indices", and by the end of the blackboard they're superscript. (Or vice versa.)
17:33:28 <fizzie> Middlescript is just like regular text except it's kinda smaller.
17:33:48 <kallisti> I think it would make sense to be vertically centered as well.
17:34:09 <fizzie> I suppose some of them are. They sort of range all over the VERTICAL RANGE.
17:34:19 <Phantom_Hoover> <fizzie> oklofok: There's a maths professor at our place who starts with subscript indices, then they gradually rise up to be "middle indices", and by the end of the blackboard they're superscript. (Or vice versa.)
17:34:21 <kallisti> so slightly above normal script, "in the middle"
17:34:58 <Phantom_Hoover> I had a chemistry teacher who wrote the stuff he was teaching up on a whiteboard, and it sloped down towards the right a lot.
17:35:43 <fizzie> Oh, and the indices immediately drop down to subscript if there's a case of superscript-denoting-exponentiation nearby.
17:36:03 <Phantom_Hoover> You wouldn't have to bother adjusting the height of your hand, for one thing.
17:36:06 <kallisti> fizzie: that sounds really confusing.
17:36:44 <calamari> argh.. I need to go back to magical C kindergarten.. can someone check whether I converted these arrays properly? http://pastebin.com/tbbBbZPN
17:36:50 <fizzie> Well, usually you can tell from the context, since normally it's the same "k" that keeps wandering up and down.
17:39:05 <fizzie> The ragged-array construction is not exactly equivalent, but I suppose it's the closest sensible thing. And there's a school of thought that would write the mallocs as malloc(num_iso_transfers * sizeof *tfr) and malloc(num_iso_transfers * sizeof *isobuf) and malloc((64 * 3072) * sizeof **isobuf) and/or malloc(64 * 3072) since sizeof(unsigned char) is and always will be exactly 1.
17:39:34 <kallisti> fizzie: all C code should strive to be self-documenting.
17:40:40 <fizzie> Also instead of "(64 * 3072) * sizeof X" it maybe a good idea to introduce the size_t-ness already in the multiplication for when you port it to a place with 16-bit ints, where (64 * 3072) will be 0.
17:40:51 <fizzie> (Okay, that's the picking of nits.)
17:41:30 <fizzie> (Sadly, there's no size_t literal suffix. Wouldn't 3072z look just peachy?)
17:42:58 <zzo38> You could modify CWEB to add such things; WEB added various things to Pascal which they didn't have at first so you could do with C as well
17:43:54 <calamari> fizzie: I was mainly worried whether I even was technically correct, but I appreciate your style tips as well!
17:44:44 <fizzie> I didn't find anything to complain about on the correctness front, so I sort of had to make do.
17:46:31 <fizzie> (I almost had to resort to the "malloc might return NULL" backup option.)
17:47:05 <calamari> fizzie: I am checking for NULL believe it or not :)
17:48:06 <fizzie> Since 64 * 3072 is constant, it *would* be possible to have unsigned char (*isobuf)[64*3072]; isobuf = malloc(num_iso_transfers * sizeof *isobuf); without the whole array-of-pointers deal, but then isobuf has a slightly unwieldy type, and maybe looks a bit arcane.
17:54:43 <fizzie> 19:51 <fizzie> ,cc unsigned char (*isobuf)[64*3072]; size_t s = sizeof *isobuf;
17:54:46 <fizzie> 19:52 <candide> fizzie: <no output: s = 196608>
17:54:49 <fizzie> That's one useful bot for checking that I'm not spouting pure bullcrap.
17:54:51 <fizzie> fungot: Why can't you be that useful ever?
17:54:51 <fungot> fizzie: wouldn't that force you to use fnord to read a text stream/ to read a text i just wrote
17:55:11 <fizzie> fungot: I... guess that's fair. Never mind, then.
17:55:11 <fungot> fizzie: i found some sql foo recently. it's often easier to be forced is not really
18:00:49 <Sgeo> kallisti, update if you didn't see it
18:03:16 <Gregor> !c unsigned char (*isobuf)[64*3072]; printf("%d\n", (int) sizeof(isobuf));
18:03:27 <Gregor> !c unsigned char (*isobuf)[64*3072]; printf("%d\n", (int) sizeof(*isobuf));
18:03:52 <Gregor> That is one severely weird thing to do to C.
18:04:54 <Gregor> Well, it's not like it's going to perform bounds-checking, and the amount of space will be documented by the allocation anyway.
18:05:17 <fizzie> !c unsigned char (*isobuf)[64*3072]; printf("%zu\n", sizeof *isobuf); /* it's like *2011*, man, it's okay to use C99 features */
18:05:45 <Gregor> I guess if you really absolutely needed sizeof(*isobuf) to be right for macros or something, there's that.
18:06:33 <Gregor> Also, I compile all my code with -Wall -Werror -ansi -pedantic :)
18:08:46 <fizzie> I'm still not sure what the weird thing was. I mean, the argument for "x = malloc(sizeof *x)" is that when you have a "struct funky *x" and you update it to "struct hunky *x" later, you don't have to remember to update a "x = malloc(sizeof(struct funky))" which, of course, wouldn't be detected at compile-time to be a problem.
18:09:07 <fizzie> (Also sizeof without parens looks cleaner.)
18:10:19 <Gregor> It's highly unlikely that you're going to be able to swap a struct for an array regardless, the malloc being the same is only a first convenience on a path of suffering ...
18:12:21 <fizzie> That was just a general-purpose justification for "malloc(sizeof *x)", where x might be one sort of a struct that gets changed to another, closely related sort.
18:17:12 <fizzie> (candide's funky "print out all locals at the end" thing is some sort of a gdb-driven thing.)
18:19:47 <fizzie> 20:19 <fizzie> ,cc char can_i_have_some_uninitialized_data_please[4096];
18:19:48 <fizzie> 20:19 <candide> fizzie: <no output: can_i_have_some_uninitialized_data_please = '\0' <repeats 1224 times>", .">
18:19:52 <fizzie> That was a bit boring.
18:20:23 <fizzie> Well, there's the ", ." part, I'm not sure what that's supposed to be.
18:27:44 <Gregor> So what does candide do for protection?
18:28:07 <fizzie> It's some sort of a VM thing.
18:28:22 <fizzie> I forget exactly, someone mentioned something about it.
18:28:33 <fizzie> One of the well-known ones, anyway.
18:31:23 <fizzie> 20:31 <fizzie> ,cc #include <stdlib.h>\n system("cat /proc/cpuinfo");
18:31:25 <fizzie> 20:31 <candide> fizzie: processor : 0 vendor_id : AuthenticAMD cpu family : 6 model : 2 model name : QEMU Virtual CPU version 0.12.5 stepping : 3 microcode : 0x1000065 cpu MHz : 800.000 ...
18:32:08 <ion> `run sh -c 'ping 8.8.8.8'
18:32:43 <fizzie> `run file `which ping`
18:32:46 <HackEgo> /hackenv/bin/ping: Bourne-Again shell script text executable
18:32:50 <ion> `run cat /hackenv/bin/ping
18:32:54 <fizzie> Sounds like the best sort of ping.
18:32:59 <HackEgo> /hackenv/bin/ping \ /bin/ping
18:33:15 <ion> `run file /bin/ping
18:33:18 <HackEgo> /bin/ping: setuid ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.18, stripped
18:33:23 <ion> `run /bin/ping 8.8.8.8
18:33:26 <HackEgo> connect: Network is unreachable
18:45:53 -!- augur has quit (Remote host closed the connection).
18:55:56 -!- azaq23 has joined.
18:56:37 -!- elliott has joined.
18:56:44 <lambdabot> elliott: You have 1 new message. '/msg lambdabot @messages' to read it.
18:58:39 <elliott> anyway: http://esoteric.voxelperfect.net/wiki/Esolang_talk:Community_Portal#Site_move_.2F_wiki_read-only to anyone who hasn't seen it
18:59:10 <elliott> 09:45:52: <oerjan> wow it's really happening
18:59:40 -!- NihilistDandy has joined.
19:01:34 <fizzie> elliott: Oh, so you're MAKING THIS HAPEN.
19:03:36 * elliott just assumes Gregor is whoaing at my GETTING THINGS DONE lyfestyle.
19:03:58 <Gregor> Whoaaaaaaaaaaaaaaaaaah
19:04:05 <Gregor> That whoah was at "lyfestyle"
19:04:23 <elliott> This marks the first thing I have ever accomplished without Gregor's mockery.
19:04:36 <Gregor> That's not quite true.
19:04:40 <Gregor> You've done things I haven't noticed.
19:05:01 <elliott> This is the first thing I have ever accomplished.
19:05:13 <Gregor> Ah, so that statement was vacuously true.
19:05:43 -!- NihilistDandy has quit.
19:06:25 * elliott makes about ten copies of esostuff.7z in different locations on his hard drive.
19:06:49 <elliott> $ sudo cp esostuff.7z / && sudo chown root:root /esostuff.7z && sudo chmod 500 /esostuff.7z
19:06:58 <Sgeo> Sounds useful were your HD to crash
19:07:08 <elliott> Yes, like say if I were to drop my laptop.
19:14:33 <HackEgo> folez rended sovrainte hed rig tfi ptyczner num izrasoagh jlr dihiw qua avissio inflicheim romentalnuri fubau sametaria pogent souall ress via comr autstara prae electorei
19:19:58 -!- derdon has quit (Remote host closed the connection).
19:22:40 -!- jix has quit (Quit: Lost terminal).
19:23:09 -!- jix has joined.
19:26:00 <fizzie> That doesn't look like 50 to me.
19:26:05 <HackEgo> laisiiparvitsevassanovitra raisimmistamiin irtävältä vällänsäänsä hauksi havaksemmältä hallenne koukisimpan törmääremme sovesitoskin sisemmiksemme poimpienne kokoosissa karityksi tullasiksi epäivänansa yllyttä tuottauksi lisemiselviyöstöiksesi hyvänne
19:27:11 <elliott> 213.222.12.125 - - [17/Feb/2012:11:46:49 +0000] "GET //phpMyAdmin//scripts/setup.php HTTP/1.1" 404 169 "-" "Plesk"
19:29:23 -!- pikhq has quit (Ping timeout: 245 seconds).
19:29:31 -!- pikhq has joined.
19:33:17 -!- boily has quit (Ping timeout: 272 seconds).
19:36:22 -!- Taneb has joined.
19:36:48 <Taneb> Thue-Morse sequence as a Luigi program
19:41:06 <elliott> @ask ais523 Do we want the RC patrol stuff?
19:41:40 <fizzie> Heh, someone's hit my QBasic help file translation with the Google search ""QBASIC PATH NOT FOUND" WIN 7".
19:41:53 <fizzie> (Also you made me look at server logs.)
19:42:29 <elliott> that "-" is the referer, right?
19:42:42 <fizzie> Probably, in the default log-config.
19:43:14 <fizzie> Well, it looks like it's trying to be Apache-combatible, for log-file-lysis purposes.
19:44:03 <fizzie> Oo, a good old "GET /w00tw00t.at.ISC.SANS.DFind:)" too. (It's one of those vulnubbarity scanners.)
19:45:13 <fizzie> /admin/index.php, /admin/pma/index.php, /admin/phpmyadmin/index.php, /db/index.php, /dbadmin/index.php, ..., /phpMyAdmin-2.5.6-rc1/index.php, ... -- it's like "how many ways you can spell it" contest.
19:45:33 <elliott> fizzie: What's a good thing to look through a several-hundred-megabyte text file?
19:45:55 <fizzie> I would think that depends on what you need to look for.
19:45:55 <Sgeo> If I were to write, say, Maybe stuff in Smalltalk, would it be acceptable to use gather: as bind and with: on the class side as return?
19:46:11 <Sgeo> Because Collections pretty much do that,
19:47:03 <elliott> fizzie: I just want to browse it.
19:47:31 <elliott> Sgeo: Monads are pointless in any dynamic language.
19:47:55 <elliott> You can't write generic code at all without explicitly plumbing around the dictionary (in this case the class) everywhere, which makes it incredibly painful.
19:48:05 <Sgeo> Is there a nicer way to do list comprehensions in Smalltalk than a bunch of gathers?
19:48:14 <Sgeo> {1. 2} gather: [:a | {1. 3} gather: [:b | Array with: (a+b)]]
19:48:35 <Taneb> Nobody cares about my implementation of the Thue-Morse sequence in Luigi
19:49:00 <fizzie> Oh, right. I don't really know. I've used less with rather big files too; if you do something that requires it to do a line count (like "seek to 50%") there's a long pause, but afterwards it caches, for that one run. Maybe haven't done *several*-hundred-megabyte files, though.
19:49:00 <Sgeo> elliott, well, couldn't I make a trait or something, put the generic code in the trait?
19:49:32 <Sgeo> Having some trouble working out how to browse and use traits though
19:50:11 <fizzie> Neither of Vim and Emacs is, according to my own experiences, all that good about hueg files, at least out-of-the-box just-like-that.
19:50:31 <elliott> fizzie: Emacs seems to be working OK enough.
19:50:41 <elliott> The file also has incredibly long lines, sigh
19:50:45 <Gregor> vim I've found to be particularly awful with huge files.
19:51:07 <elliott> like, almost 1,000,000 col lines
19:51:35 <fizzie> It is that, yes. But Emacs didn't deal all too well with a "single multi-megabyte line" file either. Movement was sort-of laggy.
19:51:41 <Taneb> Oh god oh god oh god
19:51:48 <fizzie> Not sure if it was in some non-optimal mode, though.
19:51:49 <Sgeo> Maybe I should try CL with SLIME?
19:52:11 <Taneb> I've found the Ook!++ spec
19:52:12 <fizzie> Sgeo: Did you ever answer whether your login-watching was about stalking someone or something completely different?
19:52:14 <elliott> fizzie: Yah, I'm just resigned to a fairly painful experience.
19:52:23 <elliott> But Ctrl+down to skip huge lines seems to work OK.
19:52:30 <Sgeo> fizzie, oh, I don't remember if that was asked
19:52:45 <Sgeo> fizzie, professor wants us to stalk him?
19:53:28 <elliott> You know what I hate more than MediaWiki?
19:55:44 <elliott> Anyway, OK, I have everything I need. I'll probably start installamating things later today or whenever ais is on.
19:55:51 <fizzie> "less" does this 400M Apache log quite fine, though it doesn't have excessively long lines, and I'm not sure what would be the correct option to skip one of those. (Though "less -S" would show always one line on one line -- and then of course force you to scroll vertically to see what's beyond the edges.)
19:56:40 <elliott> I use less -S for looking at access.log.
19:56:47 <elliott> Otherwise the alignment of the fields and dates and such gets mucked up.
19:59:39 -!- H3LLB0Y has joined.
20:02:40 <Sgeo> "scroll vertically" for that?
20:06:40 <Sgeo> Trying to make a million Processes was a bad idea, I think
20:06:54 <elliott> Hmm, the migration plan might be slightly better adjusted slightly, to be a little out of sync with the two servers or with a brief period of inaccessibility.
20:07:19 <Sgeo> Hmm, Pharo's style of clarifying which ] goes with which [ is not so great for the colorblind
20:08:23 <coppro> are you the colorblind?
20:09:05 <Taneb> What colours are used?
20:10:05 <Sgeo> green and purple I think, for starters
20:10:19 <Sgeo> Can't really tell right now, image frozen with my line selected
20:10:32 <Sgeo> The thing can't even do 100 processes comfortably.
20:10:40 <fizzie> Sgeo: Horizontally is probably what I was supposed to say. Vertically you need to scroll anyways.
20:12:53 <Sgeo> The thing doesn't seem to be able to handle even 100 processes
20:12:59 <fizzie> I always scroll consensually only.
20:13:18 <Sgeo> 1 to: 100 do: [:x | [[true] whileTrue] fork]
20:13:19 <fizzie> Consensually, apparently.
20:13:40 <fizzie> Shouldn't have second-guessed.
20:15:33 <Sgeo> Tried it with 10
20:15:44 <Sgeo> But there was a difference: I didn't have Process Manager with autoupdate on
20:16:14 <Sgeo> Things slowed down a bit. Things froze when I opened Process Manager and turned autoupdate on
20:24:17 -!- ais523 has joined.
20:25:16 <itidus20> so if you have a function which relates the number of turns taken in tictactoe with the number of x's on the board, then the derivative of that function would be about 1/2 eh
20:26:37 <Jafet> No, the derivative is 0 almost everywhere
20:26:49 <elliott> I take it you've seen the esowiki
20:27:55 <ais523> no, I haven't, I've only just got online
20:27:55 <lambdabot> ais523: You have 1 new message. '/msg lambdabot @messages' to read it.
20:28:00 <lambdabot> elliott asked 46m 54s ago: Do we want the RC patrol stuff?
20:28:08 <elliott> ais523: ah, see the link at the bottom of the main page then
20:28:24 <ais523> elliott: I don't think Esolang's the right size for RC patrolling, it's best on medium-sized wikis
20:28:34 <ais523> nethackwiki's just the right size, but esolang's too small and wikipedia's too large
20:29:01 <Jafet> Why doesn't RC patrol work for small wikis?
20:29:22 <ais523> Jafet: because you want admins to check each change once each, rather than once between them
20:29:48 <elliott> ais523: more importantly than DB lock, I contacted Graue and the process has begun :P
20:29:53 <elliott> I have an esostuff.7z here
20:29:58 <elliott> actually, several copies of it, because I'm paranoid
20:30:24 <Sgeo> But not paranoid enough about suddenly turning into me?
20:30:40 <ais523> Sgeo: I've never considered that I might suddenly turn into you
20:30:56 <ais523> but if I /did/, I'd have your memories not mine, so I probably wouldn't realise I was previously ais523
20:31:04 <Jafet> I suppose that's true for a wiki dedicated to weird things.
20:31:38 * Sgeo in particular is referring to a tendency of breaking hardware
20:31:39 <calamari> elliott: you taking over the wiki?
20:31:49 <fizzie> Jafet: I would assume, if the function is defined like that, it has a discrete domain, and not that it's a step-function-like thing from (a part of) R.
20:33:16 <fizzie> I, for one, welcome our new wiki overlords. (Isn't that what you're supposed to say in situations like this?)
20:33:22 <Jafet> fizzie: there is no derivative on a discrete domain
20:33:30 <fizzie> Jafet: Right, so it's not 0.
20:34:04 <calamari> meet the new boss, same as the old boss?... err wait maybe that's not the one :P
20:34:55 <itidus20> yeah i guess a tic tac toe game does not really become a continuous curve just because it's convenient
20:35:12 <Jafet> Now you are obliged to invent it
20:35:17 <Jafet> Analytic tic-tac-toe
20:35:28 <fizzie> You can do thing with differences that are rather similar to derivatives, though.
20:36:03 <itidus20> well, incase you can't guess, i am reading up on derivatives as part of a quest to understand functions
20:36:30 <Jafet> You can't understand functions by reading about derivatives
20:36:30 <elliott> calamari: does anything actually use that EsoShell stuff? I'd rather not set up another namespace
20:36:36 <elliott> I think you're responsible for it
20:36:41 <Jafet> Not that there is much to understand about functions in general
20:36:59 <itidus20> but i see that it works if i warp "a function which the number of turns taken in tictactoe with the number of x's on the board" into y = 0.5 * x
20:37:20 <itidus20> which obviously isn't really true about tictactoe
20:37:37 <elliott> http://esolangs.org/w/index.php?title=Special%3AAllpages&from=&namespace=100
20:37:38 <elliott> http://esolangs.org/w/index.php?title=Special%3AAllpages&from=&namespace=101
20:38:07 <fizzie> More to the point, even if you restrict x to integers, there aren't 0.5 crosses at turn 1.
20:38:20 <Jafet> You've (pointlessly) generalised the discrete function [1..5] -> [1..5] to the real function R -> R
20:38:30 <calamari> damn that's awesome I completely forgot about that
20:38:40 <Jafet> The latter happens to have a derivative, but it also happens to have nothing to do with tic-tac-toe
20:38:44 <calamari> I think I was using the wiki as a filesystem
20:39:10 <calamari> you can go ahead and ditch that
20:39:25 <Jafet> calamari: it's better than git!
20:41:19 <Sgeo> I take it there's no chance I'll be given a PSOX namespace
20:44:06 <elliott> I can assure you it is policy to ensure you are not given a PSOX namespace.
20:47:35 <Taneb> Does anyone want to see the spec for Ook!++?
20:48:29 <itidus20> ok what i can sort of imagine is both players having sort of squirt guns which constantly fill in part of the tictactoe board at a constant rate but player 1 begins squirting 1 turn early
20:50:17 <Taneb> I'm almost as old as you can get while still being younger than the submition of the final manuscript of Andrew Wiles' proof of Fermat's Last Theorem
20:50:34 <Taneb> (it was on October 25, me on November 3)
20:50:53 <ais523> Taneb: that sounds suspiciously like a BF derivative
20:51:28 <Taneb> ais523: I'm sure it doesn't count if I had never really paid any attention to BF
20:51:41 <Taneb> But yeah, it was when I was a little esolangling.
20:51:44 <ais523> well, Ook! is a BF derivative
20:51:50 <ais523> so it sounds like a BF derivative derivative
20:52:01 <fizzie> A BF second derivative.
20:52:21 <fizzie> Sometimes called a BF curvature.
20:52:25 <Sgeo> Maybe a BF hundredth derivative might be so different from BF that it's actually good.
20:53:13 <Taneb> Oh damn, it's not even backwards compatible.
20:55:01 <Taneb> Ook!++ is however still true to the spirit of Ook!, in that it is usable by orang-utans
20:55:16 <Taneb> (it introduces the "Eek" keyword.)
20:56:27 <Jafet> So, real tic-tac-toe. Given two functions x(t) and o(t) on the unit square, 0 ≤ t ≤ 1 and x(t), o(t) ∈ ([0,1] × [0,1]). Each function "claims" the points it first passes through. If x(t) = o(t), neither function claims the point. The first function to claim all points on a line of length 1 wins.
20:56:42 <Jafet> I'm still not sure if the functions should be continuous.
20:58:08 <Sgeo> Can both achieve the victory condition?
20:58:10 <Jafet> I'm also not sure how many quantifiers are needed to express this game in first order logic
20:58:12 <elliott> Jafet: How is this better than continuous chess?
20:58:30 <Sgeo> If one goes along one diagonal and another goes along the other diagonal but neither are exactly on the diagonal
20:58:37 <itidus20> i'm indirectly responsible for both i think
20:58:38 <Sgeo> How much room for maneuvering is there
20:58:54 <Jafet> elliott: it's easier to write an AI!
20:59:48 -!- nooga has quit (Ping timeout: 245 seconds).
20:59:56 <Jafet> Also, I haven't restricted the functions to be continuous
21:00:05 <Jafet> Though for the sake of analysis, I should
21:00:15 <fizzie> Jafet: How do you take turns in this thing? Do you just keep altering your functions?
21:00:33 <Sgeo> fizzie, I imagine that it's just one function vs the other
21:00:46 <Sgeo> On turn 0? Jafet may have different ideas ofc
21:00:48 <Jafet> Well, I haven't even yet restricted the functions to be computable
21:01:26 <Jafet> Actually, they can't be computable because they need to pass through a dense set
21:01:45 <Jafet> Of course, particular instances of the functions are definable
21:01:57 <Jafet> And by definition, we can only play using those
21:03:53 -!- nooga has joined.
21:04:22 -!- nys has quit (Quit: quit).
21:04:53 <itidus20> Sgeo: by my understanding, 2 horizontal functions would result in a draw
21:05:18 <Sgeo> ...I didn't even think of two horizontal functions
21:05:23 <itidus20> unless they weer at the same horizontal
21:05:33 <Taneb> It would still be a draw?
21:05:39 <itidus20> in which case... they would draw but shamefully
21:05:49 <fizzie> If you put in a space-filling curve that fills the whole square, can you lose? I'm still not quite sure of the whole "first function" thing, is that like in terms of t or what? As in, the smallest t_e for which x(t) or o(t) with t \in [0,t_e] claims a line? But couldn't you just make a space-filling curve that's be arbitrarily "fast"?
21:06:11 <Jafet> fizzie: sure, but the other function can be equally fast
21:06:22 <Jafet> I guess it's a draw by symmetry
21:06:26 <Taneb> I think it's "claims a line of length 1 with minimal t"
21:06:44 <fizzie> So that was the correct meaning of "first"? Hokay.
21:07:11 <Jafet> I wonder if you can inject any asymmetry into this game
21:07:18 <itidus20> it sounds a lot like tron as a function
21:07:36 <Jafet> You can't make x(t) start a finite amount earlier, since it can draw a line in any finite amount of time
21:08:01 <Jafet> You can make it start infinitesimally earlier, as long as you tell me what the hell that even means
21:08:17 <Jafet> Maybe x(t) is allowed to claim an arbitrary point first
21:08:44 <Jafet> Or (as in go?) a countable set of points
21:09:15 <Taneb> You know what would be fun, less theoretically daunting, and easier to implement?
21:09:22 <Taneb> Simultaneous tic-tac-toe
21:09:39 -!- oerjan has joined.
21:09:42 -!- augur has joined.
21:10:10 <Sgeo> What is with the turns thing?
21:10:15 <Sgeo> Who is thinking in terms of terns?
21:10:41 <fizzie> Sgeo: Anyone who plays regular tic-tac-toe?
21:11:26 <itidus20> one way of imagining this that i got was jousting functions
21:13:35 <tswett> So, this is where everyone goes for names, right?
21:13:43 <tswett> "CASE Bus System". Come up with a backronym.
21:14:08 <fizzie> Computer-Aided Sex Education.
21:14:11 <itidus20> i should stop with these really bad topics...
21:14:21 <Taneb> Crazy Alternate Singing Enterprise
21:14:22 <itidus20> noone plays tictactoe and for good reason
21:14:26 <tswett> Consider something related to bus systems?
21:15:01 <fizzie> I'm not even sure what a bus system is. :/
21:15:01 <elliott> carbonated ale superbly eked
21:15:14 <tswett> Computer-Aided Sex Education is a nice backronym, but it doesn't sound like the sort of thing a bus would be called.
21:15:20 <elliott> casebussystem asebussystem sebussystem ebussystem
21:16:31 <fizzie> Cockburningly, Amazingly, Seriously Expensive Bus System.
21:16:46 <tswett> (Pronounced "Coburningly".)
21:17:06 <tswett> I'm starting to think that "CASE" isn't a great acronym for a bus system.
21:17:31 <oerjan> can acronyms stabilize ever?
21:18:07 <fizzie> What *is* a bus system? I mean, there's the system bus, and all kinds of different buses in different contexts. Or... is it, like, with wheels? A bus?
21:18:30 <fizzie> I was thinking of, you know, buses.
21:18:31 <tswett> A giant box with wheels that people get into and wait until it rolls into the right spot.
21:18:35 <Taneb> The wheels on the bus go round and round
21:18:46 <fizzie> Desert bus, right, I know about those.
21:18:47 <tswett> Carriers for relatively large amounts of data?
21:18:51 <Taneb> The wheels on the bus go round and round
21:19:12 <oerjan> ceteribus antabus sillybus exibus
21:19:34 <Taneb> Cars Are Sorta Expensive
21:19:53 <tswett> Taneb: ...yes! I'm using that.
21:20:41 <fizzie> I was going to go with "Clean And Sober Experience", by way of cheating via acronymfinder.
21:21:33 <fizzie> (It was the first I noticed on the list with positive bus-related qualities. But the car thing is better.)
21:22:34 <oerjan> cars and shuttles explode
21:23:21 -!- MoALTz has quit (Ping timeout: 252 seconds).
21:24:32 <elliott> capes always see exoskeletons
21:28:34 <tswett> elliott: çomeone already superiorated eu.
21:28:38 -!- oklofok has quit (Ping timeout: 240 seconds).
21:30:28 -!- fenriswoolf has joined.
21:31:47 <HackEgo> Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page
21:32:09 <ais523> we're busy moving the wiki to another server atm
21:32:44 <elliott> fenriswoolf: hi! how did you find out about this place?
21:32:47 -!- MoALTz has joined.
21:32:54 <zzo38> elliott: Good question.
21:33:15 <fenriswoolf> searched for esoteric ... have a tooth for the mysterious ...
21:33:26 <ais523> ah, you're probably talking about the other meaning of "esoteric"
21:33:31 <HackEgo> This channel is about programming -- for the other kind of esoterica, try #esoteric on irc.dal.net.
21:33:33 <zzo38> fenriswoolf: This channel we discuss a lot of various thing but mostly computer programming stuff.
21:33:41 <HackEgo> This channel is about programming -- for the other kind of esoterica, try #esoteric on irc.dal.net.
21:33:42 <zzo38> fenriswoolf: However, if you do have a question, please ask
21:34:07 <zzo38> (We are rarely on topic but who knows)
21:34:17 <fenriswoolf> i think i am not in your league ... i just started looking into python
21:34:37 <fenriswoolf> just found your channel name intriguing ....
21:35:11 <itidus20> fenriswoolf: what zzo38 is trying to point out is that this isn't the place for study of necronomicon or the like (although it could be potentially)
21:35:20 <fenriswoolf> but if you do not mind if i drop in when i run into a python road block
21:35:31 <zzo38> fenriswoolf: Try the Python channel.
21:35:51 <fenriswoolf> ah .. not everything esoteric is necromancy ...
21:35:58 <zzo38> If you have a question about anything at all (on topic or not) please ask; we discuss a large number of things on this channel which are not on topic and can be confusing so ask whatever.
21:36:05 <fenriswoolf> hey thanks ... have a great weekend ... theo
21:36:15 <itidus20> this channel is all about "weird" programming languages
21:36:19 <zzo38> itidus20: Exactly. It isn't, but can be potentially if anyone has anything to say
21:36:40 <Taneb> Hmm... I know a number of people called Theo
21:36:41 -!- fenriswoolf has left.
21:36:42 <zzo38> (But still isn't the main topic regardless. The main topic is esoteric computer programming, but we are often not on topic)
21:37:03 <zzo38> Taneb: I didn't know that.
21:37:13 <itidus20> like theres nothing to stop someone making an occult esolang
21:38:24 <zzo38> Yes, they do mention necromancy in that one... good point!!!
21:38:57 <elliott> Well, OK, there's no actual necromancy.
21:46:11 <itidus20> delete p; Necromancer n; n.resurrect(p); http://delipit.ro/magazin/wp-content/themes/shopperpress/thumbs/inglip.jpg
21:48:58 <oerjan> is inglip still around
21:50:54 <oerjan> ...that's not really relevant to my question
21:51:07 <itidus20> your question is very difficult
21:51:52 <itidus20> i guess i would have to check inglip's homebase
21:54:22 <elliott> Sgeo: please look at your user talk pages /binbf and /ffbimp
21:54:32 <elliott> tswett: please look at your user talk page /EsoS
21:55:18 <Sgeo> elliott, where, on esolangs.org or new wiki?
21:55:52 <Sgeo> Done, what about them?
21:56:03 <Sgeo> (Although I wasn't logged in)
21:56:31 <elliott> I want to not bother copying this table:
21:56:32 <elliott> INSERT DELAYED INTO `mw_user_newtalk` (`user_id`, `user_ip`) VALUES (5,''),(0,'Sgeo/binbf'),(0,'Ihope127/EsoS'),(0,'202.139.118.19'),(0,'Sgeo/ffbimp'),(0,'MediaWiki default'),(489,''),(214,''),(0,'75.3.255.119'),(0,'88.226.134.138'),(0,'89.245.108.252'),(0,'18.85.1.11'),(255,''),(0,'203.223.152.82'),(374,''),(368,''),(11,''),(0,'76.235.211.209'),(305,''),(295,''),(529,''),(294,''),(545,''),(3,''),(12,''),(0,'96.32.244.172'),(369,''),(34,''),(0,'
21:56:32 <elliott> 122.213.250.14'),(0,'60.217.232.70'),(0,'202.176.202.137'),(0,'92.86.78.203'),(0,'72.200.73.175'),(0,'95.169.184.138'),(0,'200.29.96.75'),(580,''),(22,''),(27,''),(357,''),(21,''),(2,''),(669,''),(26,''),(468,''),(831,''),(592,''),(0,'188.121.63.154'),(562,''),(557,''),(0,'88.22.228.4'),(0,'71.238.223.31'),(0,'82.195.156.186'),(0,'63.105.26.46'),(0,'216.83.145.130'),(0,'68.226.23.83'),(0,'195.229.242.57'),(29,''),(0,'88.23.182.79'),(475,''),(0,'1
21:56:34 <elliott> 24.6.181.166'),(228,''),(0,'66.188.73.213'),(650,''),(154,''),(0,'64.75.72.4'),(0,'Ling111216L;'),(623,''),(0,'69.72.75.139'),(586,''),(661,''),(71,''),(0,'68.95.248.65'),(789,''),(771,''),(837,''),(838,''),(0,'Ling111216G'),(0,'Ling111216N'),(2712,'');
21:57:11 <Sgeo> You don't want to copy over our pages?
21:57:23 <Sgeo> Oh, as in, things we didn't check?
21:57:27 <tswett> So, my user_id is 0 and my IP address is 'Ihope127/EsoS'?
21:59:56 -!- Taneb has quit (Quit: TTFN).
22:10:12 -!- zzo38 has quit (Read error: Connection reset by peer).
22:12:37 <elliott> I am learning far more about MediaWiki than I ever intended to.
22:21:09 <ais523> elliott: that looks like the "you have new messages" banner table
22:23:01 <elliott> it's also had its schema changed since, so dropping it = one fewer thing that can gow rong
22:24:30 <elliott> wow, I just remembered that Wikipedia Review exists
22:27:23 <elliott> In some common configurations of MySQL 4.1 and later, mysqldump can corrupt MediaWiki's stored text. If your database's character set is set to "latin1" rather than "UTF-8", mysqldump in 4.1+ will apply a character set conversion step which can corrupt text containing non-English characters as well as punctuation like "smart quotes" and long dashes used in English text.
22:27:34 <elliott> ais523: remind me to check that
22:28:36 <ais523> I'm not even sure what that's a bug in
22:28:45 <ais523> as it's a set of locally reasonable configurations that make no sense globally
22:30:55 <elliott> `ar_title` varchar(255) character set latin1 collate latin1_bin NOT NULL default '',
22:30:58 <elliott> this thing is latin-1 up the wazoo :(
22:31:33 * elliott decides to just remove those
22:32:23 <elliott> ais523: what's the Perl for going to the next iteration of a while?
22:32:29 <elliott> "next" seems to switch to the next element without breaking control
22:32:44 <ais523> did you put it inside a bare block?
22:33:06 <ais523> you can say which loop you mean by labelling the loop, LOOPNAME: while(...) {}
22:33:09 <ais523> and then do next LOOPNAME;
22:33:12 -!- MoALTz has quit (Ping timeout: 240 seconds).
22:33:45 <elliott> but there's only one loop!
22:34:09 <elliott> `blob_index` varchar(255) character set latin1 collate latin1_bin NOT NULL default '',
22:34:09 <elliott> `blob_data` longblob NOT NULL,
22:34:09 <elliott> UNIQUE KEY `blob_index` (`blob_index`)
22:34:09 <elliott> ) ENGINE=MyISAM DEFAULT CHARSET=latin1;
22:34:21 -!- MoALTz has joined.
22:34:22 <elliott> and so the "next" should trigger
22:34:33 <elliott> wtfly, it /doesn't/ output the CREATE TABLE line itself
22:36:22 <elliott> this is why you don't write quick scripts in languages you don't really know
22:37:23 <ais523> don't you mean "last;"?
22:37:32 <ais523> I forget what break does in Perl, but it's not the same as in C
22:37:46 <ais523> it breaks out of the enclosing given() block
22:38:32 <ais523> but without a use feature, that'll be interpreted as "break";, i.e. a no-op
22:38:36 <ais523> I recommend turning on strict and warnings
22:38:45 <elliott> oh, yes, I forgot those aren't on by default
22:39:01 <elliott> -rw-r--r-- 1 elliott users 198M Feb 17 09:47 esostuff/fulldump.sql
22:39:01 <elliott> -rw-r--r-- 1 elliott users 174M Feb 17 22:38 minidump.sql
22:42:49 <elliott> this will be... interesting
22:43:22 <elliott> ais523: hey, what's a better hostname, solidity or solidus?
22:43:39 <ais523> elliott: for Esolang, the first
22:43:50 <ais523> I prefer the more direct reference
22:44:03 <elliott> I'll leave it as it is, then
22:44:29 <elliott> oh, duh, my connection is slow because I'm uploading esostuff.7z
22:44:37 -!- zzo38 has joined.
22:46:17 <elliott> note to self: 17-Feb-2012 22:26
22:47:18 <shachaf> You could've said "note to self: 20 minutes ago".
22:48:05 <Sgeo> How do you expect to find the note to self?
22:49:14 <shachaf> That's why we need to spam the logs when elliott is gone.
22:49:34 <shachaf> Then they'll develop a better notetoselfing system.
22:50:42 <Sgeo> Oh, I thought your note to self was pointing to something you said
22:51:13 <Sgeo> Maybe you were
22:56:26 -!- oerjan has quit (Quit: Good night).
22:59:15 <elliott> ais523: get the fire extinguisher
22:59:33 <ais523> oh dear, are we about to have a flamewar?
22:59:48 <elliott> I'm about to deal with MediaWiki; that's basically the same, right?
22:59:49 <ais523> or should I turn my computer or network connection off? no way you can make it catch fire remotely if you don't have a connection
22:59:59 -!- tzxn3 has quit (Quit: Leaving).
23:01:16 <elliott> hmm, I bet the EsoShell pages cause problems
23:09:27 <zzo38> In Soviet Russia, esoteric programming language invents YOU!!!
23:10:05 -!- monqy has joined.
23:10:51 <Gregor> elliott: A very short exclamation en español.
23:12:19 <elliott> Hi monqy, take a look at the wiki.
23:12:44 <Gregor> But beware, looking at it will SHATTER YOUR FEEBLE MIND
23:14:06 <zzo38> elliott: Did you fix the wiki yet?
23:14:24 <elliott> zzo38: Well, I'm working on restoring the SQL dump now.
23:14:32 <elliott> It'll probably be read-only for the next two to three days.
23:15:31 <zzo38> Can you interpret the lines on your hand as a computer program?
23:17:24 <elliott> const (putStrLn "hello, world")
23:18:04 <zzo38> That isn't what I meant.
23:18:21 <zzo38> I mean making each one different
23:18:21 <monqy> thats eliots hands
23:18:43 <monqy> my hands are const (putStrLn "hi")
23:18:43 <zzo38> And I also don't mean you personally.
23:18:50 <itidus20> what does the const mean? as in uhh.. is that related to the K in SKI?
23:18:59 <zzo38> itidus20: Yes it is same thing
23:21:06 <zzo38> This is a jukebox described in one of Hofstadter's book. It is a very large one, with only one record, which seems ordinary. There are many record players which will move on rails and touch the record. Each record player plays different music from the same record. Can we make up a esolang like this?
23:23:41 <Gregor> zzo38: Arguably, if the set of all record players is taken to be the set of all languages for which all input strings are syntactically valid, then we already have that, although we don't have very many record players.
23:25:37 <itidus20> so, can someone help explain const (putStrLn "hi") to me?
23:26:19 <kallisti> itidus20: (const x) evaluates to a function which, when applied, ignores its argument and returns x.
23:26:26 <elliott> itidus20: it takes any argument and returns a program with prints "hi"
23:27:54 <itidus20> ah right, so x is (putStrLn "hi")
23:28:06 <itidus20> i forgot about the parentheses doing that
23:29:06 <itidus20> ok it makes sense in a kind of intuitive way
23:35:10 <itidus20> i wonder how complex a sprite could be
23:35:49 <itidus20> so this idea starts to quickly get weird
23:36:07 <itidus20> like first a sprite is defined by it's width and height in pixels
23:37:16 <itidus20> and then with an alpha channel, each new pixel is a function of the previous pixel
23:38:01 <itidus20> but each new pixel could be a function of the neighborhood of the previous pixel
23:40:31 <elliott> good news: the SQL import is going well
23:41:07 -!- Vorpal has quit (Ping timeout: 265 seconds).
23:41:17 <itidus20> but it could be that each new pixel is a function of the neighborhoods of the last n previous pixels
23:52:03 -!- variable has quit (Excess Flood).
23:52:04 <Jafet> zzo: a polyglot is a rather ironic thing to call a language.
23:55:19 -!- variable has joined.
23:55:50 -!- H3LLB0Y has quit (Quit: Leaving.).
23:57:18 <Sgeo> What is xcbzone? Google is unhelpful, and from context I'm guessing a game