00:02:33 <oerjan> <tswett> Someone please create a type theory that only describes functions whose output length is at most a polynomial function of the input length. <-- there are type theories capturing PTIME and (iirc) LOGSPACE, search for bellantoni-cook
00:02:42 <oerjan> pretty sure i've mentioned this before
00:03:00 -!- adu has joined.
00:03:44 <zzo38> I noticed that in MinGW, time_t is 32-bits and __time64_t is 64-bits. Is there a good way to get 64-bit UNIX timestamps regardless of the system?
00:06:02 <fizzie> elliott: Also also, our stuff (including those discussed com-puters) will be shipped probably somewhen in February, since we need to find some permanent place to live first.
00:06:10 -!- centrinia has joined.
00:06:58 <elliott> it's a good thing Turing was curious about finding a way to pute coms, or the world today would be so different.
00:07:13 <elliott> fizzie: oh, right, you did mention london though. thanks for ruining my joke.
00:07:29 <elliott> I suppose I could argue that hexham is just a really, really far out suburb of london.
00:08:59 <Sgeo> http://www.dose.com/lists/3339/19-iPhone-Tricks-And-Tips-Apple-Doesn-t-Want-You-To-Know-7-Just-Made-My-Life
00:09:22 <Sgeo> Isn't having 'hidden'/undiscoverable features exactly antithetical to good UI design?
00:11:09 <fizzie> elliott: We've been talking about visiting Hexham (just, you know, for general tourism), you've made it sound so exciting.
00:11:38 <elliott> fizzie: I'm... really not sure I'd recommend that.
00:11:52 <Sgeo> I've been thinking of visiting Norway. Or Alaska. Or someplace, far up north. Even though I hate the cold
00:12:23 <elliott> have you considered the Lake District or something instead? :p
00:12:58 <Sgeo> I could take a vacation there for a few years hoping to see a light show
00:13:15 <fizzie> We went to Lofoten (in Norway) one Easter (last? the one before?) and it was a nice trip.
00:13:22 -!- hjulle has joined.
00:13:26 <fizzie> It's also not especially cold there.
00:13:40 * Sgeo is looking to see the northern lights
00:14:06 <boily> Norway is nice. it has oerjans.
00:14:24 <boily> Sgeo: surprising UI features are evil.
00:14:52 <fizzie> Sgeo: http://www.dpreview.com/articles/0886868715/mountain-magic-shooting-in-the-lofoten-islands <- that's pretty much from where we were.
00:17:21 <Sgeo> "The Lofoten Islands are a stormy destination, to put it nicely. On average, I generally estimate there to be one clear day/night per week. So if you only have a single week in hopes of seeing the Northern Lights, you are taking a chance with the weather cooperating. In this short time period you’re much better off heading east to Sweden and a place like Abisko, which enjoys many more clear winter nights than Lofoten."
00:17:46 <fizzie> Yes, it might not be the best place for that particular goal.
00:17:51 <fizzie> Don't quite remember if we saw any auroras.
00:18:22 <oerjan> coastal climates do tend to have clouds.
00:18:31 <Sgeo> I should figure out if I can vote
00:19:21 * Sgeo is guessing not... what counts as 'moving' in election law?
00:19:51 <oerjan> i hear election laws in the US vary by state
00:20:46 <oerjan> and in chicago you don't even need to be alive to vote
00:23:21 <oerjan> and that there's a recent uproar because some places have started require actually showing id to vote, which it is claimed discriminates against african-americans.
00:23:42 <oerjan> (it made more sense in context)
00:24:01 <elliott> that kind of thing also tends to be a pain for married women, right
00:24:21 <oerjan> ...i don't think i get the reference
00:24:27 <elliott> since surname changes that people don't necessarily bother to propagate to all their legal documents.
00:24:28 <boily> we had interesting cases of cats and potted plants voting in some recent elections.
00:25:13 <elliott> (well, married people in general, but the most common case)
00:26:27 <oerjan> in norway it's really pretty simple, there's a national register of all citizens and when you vote you're crossed off on the list in your municipality. also you must show generally show id, i'm not sure if you're still allowed not to if the vote taker knows you personally.
00:27:21 <fizzie> AIUI, "honeymoon" trips are often complicated by passport-and-surname-related irregularities.
00:27:36 <oerjan> as in, you don't register to vote, the government sends you a note to remind you if you're allowed. it also simplifies the process if you bring that note, but that's not mandatory.
00:28:53 <oerjan> also if you're paranoid the local list of voters is displayed in public, so you can go downtown to check if you're in the list.
00:29:13 <oerjan> (i suppose if you're paranoid you may or may not think that's a good thing.)
00:29:15 <elliott> clearly we need mumble mumble PGP keys mumble mumble mumble secret voting mumble cryptography mumble
00:30:07 <boily> I'm all for rampant cryptography and privacy, but GPG is borken beyond all fungottable repair.
00:30:08 <fungot> boily: " i ca'n't see anything at present. all in good time." " all fnord tell lies."
00:30:26 <oerjan> also, when foreign observers observe our votes they complain that the process is too trusting of people, but conclude that it probably works well enough in practice
00:32:43 <boily> it's a Klingon word. (at least I think so.)
00:32:59 <Phantom_Hoover> i remember in the referendum, you just said who you were on the electoral register
00:33:21 <Phantom_Hoover> like there wasn't anything to keep me from going to a different polling station and voting again as someone else
00:33:38 <oerjan> Phantom_Hoover: did you get to vote? i thought you'd moved outside scotland.
00:35:18 <Phantom_Hoover> also it's pretty normal to keep your official residence the same because of how volatile university housing is
00:37:12 -!- Guest48077 has joined.
00:38:15 -!- Guest48077 has left.
00:38:49 <zzo38> Why did Wikipedia break?
00:40:11 <oerjan> darn we didn't get to ask how they found us
00:40:26 <oerjan> or to properly `bienvenido them
00:40:55 <oerjan> do any other channels get these canaima users?
00:41:03 <HackEgo> AndoDaan_: ¡Bienvenido al centro internacional para el diseño y despliegue de lenguajes de programación esotéricos! Por desgracia, la mayoría de nosotros no hablamos español. Para obtener más información, echa un vistazo a nuestro wiki: http://esolangs.org/. (Para el otro tipo de esoterismo, prueba #esoteric en irc.dal.net.)
00:41:26 <oerjan> AndoDaan_: we get so many bewildered venezuelans that we had to make that
00:41:44 <fizzie> oerjan: None of mine, at least.
00:41:49 <boily> colombians too, mind you.
00:42:09 <oerjan> AndoDaan_: that's what we want to ask them. well, they seem to use the canaima linux distribution.
00:42:23 <oerjan> so it's _probably_ somewhere in that.
00:43:34 <oerjan> canaima is _mainly_ used in venezuela where it was invented, but presumably also in some neighboring countries.
00:55:32 <HackEgo> [wiki] [[Subleq]] http://esolangs.org/w/index.php?diff=40779&oldid=36583 * 128.100.122.180 * (+288) /* Basic */
00:56:28 <elliott> Bewildered Venezuelans is a good band name
01:03:32 <Sgeo> Turns out I can vote
01:06:02 -!- boily has quit (Quit: CONSTANT CHICKEN).
01:10:34 -!- vyv_ has quit (Ping timeout: 264 seconds).
01:11:53 -!- vyv has joined.
01:14:43 <Sgeo> "McMahon said it is unwise to issue bonds to purchase high-tech equipment that will become obsolete with new technological advances, often within a period of a few years."
01:15:00 <Sgeo> Is it just me, or does even computing equipment remain usable for a few years
01:17:26 <AndoDaan_> I'm on a laptop half a decade old, surfing the internet via a tethered cellphone...
01:18:01 <elliott> obsolete doesn't mean doesn't work
01:18:27 <Sgeo> Heading out to the polls >.>
01:18:55 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds).
01:38:46 -!- AndoDaan_ has quit (Ping timeout: 272 seconds).
01:45:59 -!- shikhout has joined.
01:47:52 -!- AndoDaan has joined.
01:47:53 <yukko> I think things moved a bit faster in the 90's/early 2000's so "obsolete within a few years" isn't too far off for some eras
01:49:04 -!- shikhin has quit (Ping timeout: 250 seconds).
01:53:22 <elliott> I feel like computer obsolescence was accelerating up to the late 2000s-ish.
01:55:15 <Sgeo> Is what I did any different from voting a straight ticket? I left some blank if my hasty Wikipedia'ing didn't give any information, but the ones I voted for were all the same party
01:56:14 <HackEgo> yukko: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: <http://esolangs.org/>. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
01:56:50 -!- elliott has set channel mode: -o elliott.
01:57:05 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: selcome: not found
01:59:15 <elliott> (or should that be `selamatpagi)
02:01:17 -!- FreeFull has joined.
02:04:00 <elliott> (thank you, I'm here all week.)
02:04:50 <Sgeo> I don't get it at all
02:30:19 -!- oerjan has quit (Quit: Nite).
02:30:53 <elliott> Sgeo: bad nichijou joke. nichijoke
02:31:10 <Sgeo> Japanese comedy manga?
02:46:36 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: selamatjalan: not found
02:50:24 -!- Vorpal_ has joined.
02:52:53 <Bike> I think Malay?
02:54:23 -!- Vorpal has quit (Ping timeout: 265 seconds).
03:06:45 <zzo38> How to convert between Gregorian dates and UNIX timestamp (without using date/time library functions)?
03:25:08 <tswett> Start by dividing the timestamp by 86400; that's your day number.
03:26:13 <tswett> Converting from date to timestamp is easier, I think.
03:28:27 <tswett> You could just loop through the years starting from 1970, subtracting the appropriate number of days for each year you go through.
03:28:57 <tswett> But then doing the conversion could end up taking several microseconds.
03:32:03 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: rienvenido: not found
03:32:13 <tswett> `run cat `which relcome`
03:32:14 <HackEgo> #!/bin/sh \ welcome "$@" | rainwords
03:32:42 <zzo38> tswett: I already got division by 86400; that might I already did before I have asked
03:33:18 <tswett> `run echo -e '#!/bin/sh\nbienvenido "$@" | rainwords' > /hackenv/bin/rienvenido
03:33:26 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: /hackenv/bin/rienvenido: Permission denied \ /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: /hackenv/bin/rienvenido: cannot execute: Permission denied
03:33:36 <tswett> `run chmod +x /hackenv/bin/rienvenido
03:33:40 <HackEgo> HackEgo: ¡Bienvenido al centro internacional para el diseño y despliegue de lenguajes de programación esotéricos! Por desgracia, la mayoría de nosotros no hablamos español. Para obtener más información, echa un vistazo a nuestro wiki:
03:35:14 <zzo38> (I don't need any dates before 1980, though; I do need to be able to calculate dates much farther into the future, though)
03:35:45 <tswett> You want to get an accurate year number.
03:36:31 <tswett> A 400-year period always contains 146097 days.
03:37:27 <zzo38> Yes, and also month and day. I do know how leap years are working. Leap seconds are irrelevant for UNIX timestamps, although perhaps if the program uses the current time for anything, if the current time is a leap second then it should sleep until it is not a leap second anymore.
03:37:39 <tswett> So you want to take your day number and divide it by the average number of days in a year, which is to say, you want to multiply it by 400 and then divide it by 146097.
03:38:18 <tswett> Which will give you a year number which is accurate to within a couple of days at worst.
03:38:44 <zzo38> Within a couple of days, is not good enough.
03:38:49 <tswett> If the day is close to a year boundary, you can always use your year number to calculate the number of leap days that have happened.
03:39:15 <tswett> And then you can use that number to correct your year number.
03:39:30 <tswett> It seems like this problem ought to be easier than this.
03:39:58 <tswett> Okay, you'll probably need a function that takes a year and returns the number of leap years between 1970 and the end year.
03:41:29 <tswett> By taking the day number and multiplying and dividing, you can come up with an approximate year number. You can use the approximate year number and your leap year counting function to come up with an exact day offset from the beginning of the year.
03:42:07 <tswett> If the day offset is negative, subtract a year and add 365 or 366 to the day offset.
04:03:07 -!- tswett has quit (Ping timeout: 246 seconds).
04:06:34 <zzo38> Is y/4-y/100+y/400-477 correct (if y is an integer type)?
04:07:59 <AndoDaan> Do milleniums influence if it's a leapyear or not?
04:16:16 <zzo38> No, only 4, 100, and 400.
04:20:38 <Jafet> Don't forget to extrapolate leap seconds
04:21:49 <Jafet> That's ok, we're still converting Gregorian.
04:22:08 <Jafet> Maybe Gregor knows something about this.
04:23:33 <lifthrasiir> zzo38: look at http://en.wikipedia.org/wiki/Determination_of_the_day_of_the_week#Implementation-dependent_methods_of_Sakamoto.2C_Lachman.2C_Keith_and_Craver
04:24:06 <lifthrasiir> it is a day-of-the-week function, but without `%7` that can be used as a timestamp conversion (with a different epoch though)
04:28:51 <zzo38> Leap seconds are irrelevant to UNIX time, from what Wikipedia says; you can treat any dates as if there aren't any leap seconds.
04:29:30 <zzo38> lifthrasiir: O, it has the same "y/4-y/100+y/400" as I used.
04:30:57 -!- vanila has quit (Quit: Leaving).
04:31:00 <zzo38> That algorithm isn't actually quite right if you want the full timestamp; it misses many things.
04:31:22 <zzo38> But I think I figured out how to convert from Gregorian to UNIX; it is other way around I wanted, now.
04:31:58 -!- shikhout has quit (Ping timeout: 255 seconds).
04:35:42 <AndoDaan> zzo38: http://ptspts.blogspot.be/2009/11/how-to-convert-unix-timestamp-to-civil.html
04:36:12 <AndoDaan> tested it in lua, and with a little rounding, it looks good.
04:37:12 <AndoDaan> 2014.68364748072.99006568413932.2204460492503e-0154.610833333333336.6539
04:38:03 <AndoDaan> 2014.6836474807 2.9900656841393 2.2204460492503e-01 4.6108333333333 36.65 39
04:39:40 <zzo38> To convert into UNIX timestamps, this is the algorithm I used and it appears to be working: http://sprunge.us/ZTZX
04:43:55 <zzo38> It says that program you link to is only for Ruby, and won't work on other programming language due to rounding negative divisions.
04:44:50 <AndoDaan> ah... no well done for me then.
05:10:33 -!- adu has quit (Quit: adu).
05:31:58 <zzo38> Actually I think I figured it out now.
05:36:41 <Sgeo> Why does YouTube show me a survey asking what ads I've seen recently? Does it not track that information? Or do they want to know what I recall seeing, not what's been presented? (I guess that's rather likely)
05:52:05 -!- MDude has changed nick to MDream.
06:04:27 -!- password2 has quit (Ping timeout: 245 seconds).
06:09:43 <zzo38> AndoDaan: http://sprunge.us/cVGF (it isn't particularly good, though, but it is more than good enough)
06:12:56 <AndoDaan> doesn't return if it's Easter or not... 2/10
06:14:01 <AndoDaan> i kid. seem pretty straight forward.
06:14:48 <zzo38> Yes I know it doesn't say if it is Easter, but that is not relevant here. (I did write a program in TeX to calculate when is Easter, though.)
06:14:53 <AndoDaan> though, i didn't think about the need to index the days of the months.
06:16:28 <AndoDaan> I vaguely remember easter days forming a diamond pattern.
06:16:29 <int-e> https://en.wikipedia.org/wiki/Computus#Gauss_algorithm
06:18:17 <zzo38> The official algorithm used by the Church is a bit complicated but there are ones that are simpler.
06:20:04 <zzo38> Easter calculation in TeX: http://sprunge.us/bWhX
06:20:13 <AndoDaan> seems random to me. they should have just set a day "x-mas" style.
06:21:00 <int-e> the proposal has been made... https://en.wikipedia.org/wiki/Reform_of_the_date_of_Easter
06:21:49 <zzo38> TeX wasn't really designed to calculate when is Easter, but it also wasn't designed to play chess or to interpret BASIC (both of which have been done).
06:23:02 <zzo38> int-e: Another proposal is astronomical method, which is a method I have proposed once.
06:27:05 <zzo38> What I would be interested in is if in a program like Astrolog you could also input queries such as ">> Sun at 0 Aries >> Sun opposite Moon >> Sunday" and have it figure out exactly when that is (and plot a horoscope for that date/time if requested)!
06:28:41 <AndoDaan> hmm. with the earth wobbles and such.
06:29:32 <AndoDaan> apparently nobody has been born under their designated starsign for 3000 years. years
06:30:01 <zzo38> AndoDaan: That isn't quite right. People are confusing signs with constellations having the same (or similar) names.
06:30:47 <zzo38> Signs are all 30 degrees and there are twelve; they are a unit of measurement for ecliptic longitude. Constellations all have different sizes and aren't a measurement of ecliptic longitude or anything like that.
06:31:50 <zzo38> Do you know the "Tropic of Cancer" and "Tropic of Capricorn"? The "Cancer" and "Capricorn" mentioned are the signs, not the constellations. (Also, the constellation is actually called "Capricornus" and not "Capricorn"; the sign is "Capricorn".)
06:33:05 <AndoDaan> i know those lines on the globe. i can see how they relate to zodiac.
06:37:23 <zzo38> Many people don't know that they are related to signs and not constellations. Some astronomers wanted to rename them, but they are just confusing signs with constellations.
06:38:06 <AndoDaan> I certainly was confused - in error - about it.
06:38:24 <zzo38> Well, now you know!
06:38:35 <AndoDaan> I mean i never gave it much thought, but i just assumed.
06:40:32 <AndoDaan> Thanks. :) (i've said "nobody is born under their star sign" a number of times before, arguing about astrology.)
06:41:09 <zzo38> Well, now hopefully you can stop making bad arguments against astrology, and perhaps learn to make better arguments against astrology!!
06:42:26 <AndoDaan> or just quit arguing all together.
06:43:14 <AndoDaan> and realize that I'm as proned to be wrong as anyone else.
06:44:02 <AndoDaan> well, as some portion of anyone else.
06:48:47 <int-e> My favourite argument is that a nurse helping to deliver a baby exerts a greater gravitational force on the baby than those stars (except the Sun)
06:50:16 <AndoDaan> idk, int-e... a lot of people felt 'floaty' when last there was a Super Moon
06:50:32 <int-e> the moon is not a star.
06:50:59 <AndoDaan> so think how much bigger stars on than the Moon!
06:51:13 <int-e> and how much further away
06:51:32 <int-e> the inverse square law is a quite powerful force diminisher
06:51:39 <AndoDaan> bah, inverse square law is overrated
06:52:24 <AndoDaan> but seriously, astrology came before Newton, right?
06:52:38 <zzo38> Yes, and I think even Newton studied it.
06:53:01 <int-e> I don't even doubt that astrology works, among people who believe in it.
06:53:39 <int-e> But that's a matter of psychology.
06:54:01 <zzo38> I have heard (probably said of astrology) that the stars to not compel, but only incline. I would say they only suggest, and even then only because people make it to do so! But you mustn't underestimate the forces of suggestion.
06:54:04 <AndoDaan> But, Astrologer wouldn't attribute whatever influence celstial have to gravity.
06:54:58 <int-e> Yes, if you read your horoscope every day then that will have some effect on your life.
06:55:18 <AndoDaan> Yeah, the best Astrologers, like the best palm readers, are the best cold readers.
06:56:47 <AndoDaan> but then again, it's a certain type of person that would be inclind to read their horoscope every day.
06:56:47 <Sgeo> Is it possible to statically link rustc so I could use it as a library?
06:56:48 <Sgeo> Hmm, I could do something like that with Haskell and GHC, right?
06:56:57 <int-e> > let i=tail.scanl(flip(:))[] in i"abcd">>=i
06:56:59 <lambdabot> ["a","b","ab","c","bc","abc","d","cd","bcd","abcd"]
06:58:01 <AndoDaan> ...And here I am arguing Astrology again, when I just said, on record, that I was quiting it.
06:58:16 <zzo38> Yes, as you can see, we can all make mistakes!
07:02:30 <zzo38> The horoscopes in the newspapers aren't even quite correct. To be correct, the date would vary a bit each year (for a similar reason to why we have leap years), and also the time of day. The Sun's ecliptic longitude will be at exactly "0 Aries" when it is the spring equinox in the northern hemisphere. When this happens varies by year; also it happens at a certain time of day. (Wikipedia has a list.)
07:02:41 <zzo38> So, the actual dates listed in the newspaper horoscopes are only approximate.
07:04:59 <zzo38> This has nothing to do with the text of the horoscopes, however; only the dates.
07:14:43 <zzo38> About "My favourite argument is that a nurse helping to deliver a baby exerts a greater gravitational force on the baby than those stars (except the Sun)", well, the fixed stars normally are not even used in a horoscope. The planets in our solar system, as well as the sun and moon, are used. (The Earth is the center of measurement, so it has no spherical coordinates, and is therefore not included.) If you mean planets, you are right about that, tho
07:19:17 <zzo38> I rarely read the newspaper horoscope myself, but when I do I read all of it instead of just one. Reading just one is sort of like reading only the thirteenth chapter in a novel which contains twenty-five chapters in total.
07:21:03 <AndoDaan> They're mostly Barnam statemts. Vague stuff that can apply to nearly anyone.
07:21:48 <zzo38> Yes, I am aware of that.
07:22:29 <zzo38> Someone else I know who sometimes reads them, told me that the people who write those kind of vague stuff are the people that are good at it; it takes some skill to write in the way they do. (I don't know if it is true or not.)
07:24:52 <zzo38> The same person was confused about what his Sun sign was; he was merely curious about it. What I found, when I put it into the computer, is that I did not have enough information, since the Sun moved from one sign to the next on the day he was born, so I would need the time of his birth too in order to be accurate. So he generally just guesses, and we both agree that it doesn't really matter for that purpose.
07:30:20 <AndoDaan> I wonder if there's like, professional Astrology software out there.
07:31:06 <AndoDaan> I mean, made by believers for believers. And with all the care any business should have.
07:32:38 <AndoDaan> probably one of my dumber wonders. there's even a websit dedicated to reviewing astrology software.
07:32:40 <zzo38> There certainly is astrology software out there; I use one called Astrolog. They are generally very good at doing astronomical calculations; it doesn't matter if you believe in it or not. (Some astronomers, in fact, use it too; together with more proper astronomical software packages. Some people, including myself, would like to be able to combine the features, in order to make it more useful.)
07:33:00 <zzo38> (Note also that Astrolog does not cost anything to get.)
07:34:03 <zzo38> It is easy to see on there, what is the exact time of the full moon, because the aspect line crosses the center of the horoscope, which is clearly marked.
07:35:33 <zzo38> I have even been able to use the "ephemeris chart" mode in order to calcuate when is Chinese New Year.
07:36:27 <zzo38> The "ephemeris chart" mode in Astrolog is a plot of time on the vertical axis and ecliptic longitude on the horizontal axis.
07:37:03 <zzo38> Generally, it displays one month at a time. You can select which objects to plot and which to ignore. (For the purpose of Chinese New Year, only the Sun and Moon are important.)
07:37:54 <AndoDaan> I really should know all this already. (Would you believe I use to be president of my local youths astronomy club)
07:39:05 <int-e> . o O ( just means you have a big mouth )
07:40:12 <AndoDaan> You can come out. It was truer then than it is now.
07:41:08 <AndoDaan> Still pretty true sometimes though. :p
07:41:39 <zzo38> Astrolog has other features too, such as a "timed exposure" mode, which causes it to not clear the screen before redrawing it. There is also a setting for "harmonic factor", which multiplies all ecliptic longitudes by the number specified (an integer from 1 to 30000; it is 1 by default).
07:41:59 <zzo38> I have used both of these things, but haven't used any harmonic factors larger than 2.
07:42:11 <zzo38> (In fact, I have used these features together.)
07:43:12 <zzo38> Actually, I used these two features together with a third one; the option to change what colors are used for various purposes in the program.
07:45:26 <zzo38> I forget what I was calculating, but it involved the full moon, which was why this became useful.
07:48:02 <zzo38> If the harmonic factor is set to 2, then the lines for the Sun and Moon cross both for the full moon and for the new moon.
07:48:23 <zzo38> As far as I know, astrologers generally use harmonic factors for an entirely different purpose than I used them for, while astronomers don't generally use harmonic factors at all.
07:48:51 <zzo38> But this shows the flexibility of use of such computer software!
07:49:54 <AndoDaan> I'm reading "top 5 astrology softwar 1999"
07:50:25 <AndoDaan> though, i guess, there would be much need to update it.
07:50:59 <AndoDaan> physics doesn't change, and if the graphics are good enough, than the graphics is good enough.
07:52:26 -!- centrinia has quit (Ping timeout: 244 seconds).
07:58:57 <zzo38> You might need to download new ephemeris data (although this is unlikely), but even if you do, it is the JPL ephemeris, which is made by NASA and is independent of the software (a lot of software with astronomical calculations uses the JPL ephemeris).
08:01:06 <AndoDaan> See, even my "physics doesn't change" line demostrated a lack of understanding what's involved.
08:03:28 <zzo38> Nevertheless it is very unlikely that new ephemeris data will be needed unless you want to calculate positions for the far past or future. But it isn't completely impossible.
08:03:44 <zzo38> (Extremely unlikely, I should say, perhaps.)
08:12:01 <b_jonas> zzo38: I haven't used astronomical software on my own computer, but I have accessed online websites running simple astronomical software to compute sunrise and sunset times and moon phases.
08:13:21 <b_jonas> If I ever wanted to compute sunrise/sunset/noon times or moon phases automatically, I would get some software for it.
08:13:33 <b_jonas> I don't much care about astrology though.
08:15:48 <b_jonas> (technically, if nethack counts then I have ran astronomical software of course.)
08:17:22 <b_jonas> But of course, sunrises and moon phases can be approximated by very simple approximations during the surrounding few millenia, so it's nothing really complicated.
08:18:04 <b_jonas> Eclipses and phases of Venus might also be relevant without astrology.
08:21:22 <b_jonas> Also the positions of the Sun and Moon on the sky too.
08:22:18 -!- centrinia has joined.
08:25:57 <zzo38> But I was thinking if a software like Astrolog could include a search function that you can type in such queries like: locate Jerusalem; January 1, [Year?] >> Sun at 0 Aries >> Sun opposite Moon >> Sunday; and then if you run this query, it ask you what year you want and tell you the astronomical reform for when is Easter.
08:28:53 <zzo38> Do you like these kind of format?
08:36:26 <mroman> make it an EDSL in Haskell and you got a deal.
08:37:35 <zzo38> I suppose something like that could be made as an EDSL in Haskell, but I haven't been able to find any ephemeris software for Haskell! (I have found messages about people wanting to port Swiss Ephemeris, which is the same software I was thinking of too.)
08:43:16 <mroman> hm. blsq doesn't have any date routines :)
08:44:03 <b_jonas> zzo38: yeah, we need an irc bot that can tell where the Sun and Moon are for people typing on smartphones in windowless cubicles
08:46:03 <zzo38> You could try to instead install ephemeris software onto the smartphones?
08:47:32 <zzo38> And then it work without needing the internet connection.
08:47:50 <b_jonas> nah... do smartphones these days even have accurate clocks?
08:49:12 <mroman> Isn't that required for GPS anyway?
08:50:23 <b_jonas> mroman: oh, right, I guess if you have GPS signal and a GPS reciever you can get the tiem from that
08:50:29 <b_jonas> shows how I'm not a smartphone user
08:54:33 <blsqbot> I've heard they're really big now.
08:56:06 <b_jonas> nah, no sorry, I won't yet.
08:59:31 <mroman> Smartphones are pretty cheap fwaiw.
08:59:53 <mroman> 130 CHF and you can get an android smartphone :)
09:00:04 <mroman> (I just went into the shop and bought the cheapest phone I could find)
09:00:55 -!- drdanmaku has quit (Quit: Connection closed for inactivity).
09:01:10 -!- skarn has quit (Ping timeout: 244 seconds).
09:10:05 <zzo38> Is there a portable way to make a C program to sleep for one second including on Windows?
09:10:15 <zzo38> (Other than making a loop)
09:13:31 <mroman> but that's probably not very portable :)
09:14:47 <mroman> and I don't think you intend on using GLib
09:16:37 <zzo38> I am using MinGW, but I intend to work in Linux and other systems too.
09:21:21 <zzo38> I found that unistd.h in MinGW declares usleep.
09:22:45 <mroman> yeah but unistd isn't that portable either
09:23:04 <zzo38> But, it isn't listed in The Open Group specifications.
09:23:44 <mroman> which takes an unsigned int seconds
09:23:48 <HackEgo> [wiki] [[User:Pietu1998]] N http://esolangs.org/w/index.php?oldid=40780 * Pietu1998 * (+115) Created page with "Random hobby programmer. Created [[Lenguage]]. More [http://codegolf.stackexchange.com/users/30164/pietu1998 here]."
09:29:14 -!- skarn has joined.
09:29:47 <zzo38> MinGW doesn't seem to have it, though.
09:33:41 <mroman> http://pubs.opengroup.org/onlinepubs/7908799/xsh/sleep.html
09:46:38 <HackEgo> [wiki] [[Lenguage]] M http://esolangs.org/w/index.php?diff=40781&oldid=40703 * Pietu1998 * (+169) added command table from [[Binaryfuck]]
09:51:17 <mroman> brainfuckiest brainfuck fuckiest fuck fuck
09:51:20 <mroman> or what was that called again
09:51:54 <mroman> "Most ever Brainfuckiest Fuck you Brain fucker Fuck"
09:53:16 <zzo38> Nevertheless I did not find sleep() in MinGW header files.
09:55:16 -!- impomatic_ has quit (Ping timeout: 250 seconds).
10:03:39 <mroman> Then, said length is converted to binary, left-padded by zeroes to a multiple of 3
10:03:47 <mroman> since 1b isn't a multiple of 3
10:03:56 <mroman> I left pad it with zeroes until it is a multiple of 3?
10:04:35 <mroman> That's going to take a while.
10:06:18 <mroman> Taneb: Can't talk. Need to prepend zeroes.
10:14:15 <HackEgo> [wiki] [[Gray Snail]] http://esolangs.org/w/index.php?diff=40782&oldid=30483 * AndoDaan * (-6) fixed link
10:20:24 -!- vifino has joined.
10:20:39 -!- vifino has quit (Remote host closed the connection).
10:46:50 -!- MickNartin has joined.
10:53:55 <mroman> I ran out of memory :(
10:57:17 -!- centrinia has quit (Quit: Leaving).
11:08:04 -!- ais523 has joined.
11:12:13 -!- bart1 has joined.
11:17:07 -!- bart1 has quit (Ping timeout: 265 seconds).
11:29:34 -!- boily has joined.
11:54:35 <mroman> the todo list for blsq is getting bigger and bigger :(
12:22:36 -!- boily has quit (Quit: UN CHICKEN ANDALOU).
12:30:29 -!- AndoDaan has quit (Ping timeout: 272 seconds).
12:46:29 -!- oerjan has joined.
12:50:18 <oerjan> darn int-e must have guessed what i did, although not entirely
12:50:40 <oerjan> (he also changed to a slow solution)
12:51:14 -!- ais523 has quit (Quit: lunch).
12:55:03 <oerjan> `` rm bin/rienvenido #It's too long for extra color codes
13:02:36 <fizzie> Lenguage seems to be one of those unary-brainfucks where you can't start a program with +s.
13:02:47 -!- AndoDaan has joined.
13:14:05 -!- CADD has joined.
13:14:09 <oerjan> no, it refers to binaryfuck, which uses an initial 1
13:15:12 <oerjan> i suppose the implementation may or not be consistent with that.
13:17:36 <oerjan> incidentally, this means that for both binaryfuck and unary, there are programs not corresponding to a legal brainfuck program...
13:19:21 <oerjan> binary = bin(total)[2:]
13:19:28 <oerjan> looks like it excludes something
13:19:37 <HackEgo> python: can't open file 'print "test"': [Errno 2] No such file or directory
13:19:53 <oerjan> `` python -c 'print "test"'
13:20:00 <oerjan> `` python -c 'print "test"[2:]'
13:20:18 <oerjan> `` python -c 'print bin(15)'
13:20:33 <oerjan> oh it only skips the 0b
13:23:28 -!- CADD has quit (Quit: leaving).
13:26:51 -!- CADD has joined.
13:28:00 -!- ais523 has joined.
13:28:47 <fizzie> oerjan: Well. It says "evaluated like binaryfuck", but I took that to mean just the way the groups of 3 are evaluated. I don't know why, since it could as easily have meant the result needs to include the initial 1.
13:29:00 <fizzie> (And I didn't look at the reference implementation.)
13:30:08 -!- CADD has quit (Client Quit).
13:30:22 <fizzie> Probably because "left-padded by zeroes to a multiple of 3, and then evaluated like Binaryfuck" sounds like it could never result in a legal Binaryfuck program.
13:30:24 -!- CADD has joined.
13:33:01 -!- AndoDaan_ has joined.
13:33:05 <fizzie> I seem to recall some discussion about a practical enumeration of brainfuck programs that would only include balanced []s.
13:33:31 <fizzie> Don't remember if that ended up in anything.
13:34:00 <oerjan> we tried to construct a bijection between naturals and well-formed brainfuck, yeah
13:34:45 -!- AndoDaan has quit (Ping timeout: 255 seconds).
13:35:10 <oerjan> we certainly got to a conceptual solution, if not several
13:35:34 <oerjan> there were complications if we didn't want exponential blowup due to nesting
13:35:37 <Lymia> fizzie, oerjan: What are you talking about?
13:36:11 <oerjan> it started with http://esolangs.org/wiki/Lenguage
13:36:41 <oerjan> although we seem to have passed on to reminiscing about number-encoding brainfuck in general
13:37:12 <fizzie> Yes, I don't remember if there was something particularly "practical" (if such a word can ever be used) -- reasonably small values, yet "easy" to compute the brainfuck program given the natural number. That was at least someone's goal.
13:37:28 <oerjan> also anyway, it doesn't _really_ matter much if there's an initial 1 or not in an encoding, since you can just prepend -+ or - if necessary
13:37:43 <fizzie> I would have prepended >< for some reason.
13:38:25 <oerjan> fizzie: i think the fibonacci base thing i thought of would have reasonable growth, since it encoded a whole list with only golden ratio blowup or thereabouts
13:38:29 <ais523> what about this: we encode the 8 commands as octal digits; then we use the numbers which correspond to malformed programs to encode the programs which would have leading zeroes
13:39:04 <mroman> what about this: no more "let's encode brainfuck differently" derivations of brainfuck?
13:39:49 <oerjan> mroman: well the old discussion was interesting in so far as it tried to construct a proper _bijection_ while all usual such derivatives leave out either some numbers or some programs
13:39:56 <CADD> oerjan: very cool
13:39:58 <int-e> oerjan: I do know Wilson's theorem. And I'd never thought that I would see the day that I would prefer (!!) :: [a] -> Integer -> a over (!!) :: [a] -> Int -> a...
13:40:34 <elliott> yeah, my brainfuck numbering thing was awful
13:40:53 <int-e> oerjan: I rather suspect that you have a more clever way of doing the substring part
13:41:49 <Lymia> http://codegolf.stackexchange.com/questions/40562/asciis-95-characters-95-movie-quotes < oh god, using esolangs to remove important characters fo non-esoteric languages.
13:41:51 <elliott> a better way to think of the problem may be "construct an enumeration of all brainfuck programs with random access"
13:41:57 <Lymia> Why are eople so cruel
13:42:09 <elliott> (okay, that doesn't guarantee the program -> natural direction, but...)
13:45:20 <int-e> oerjan: oh another thing I can't use: 0^n+n is shorter than max 1n
13:45:57 <int-e> (well, I can, but it's unlikely to help)
13:45:58 <HackEgo> [wiki] [[Talk:Checkout]] http://esolangs.org/w/index.php?diff=40783&oldid=40702 * Ais523 * (+345) /* Impossible to checkout to level 6? */ whoops
13:46:35 <int-e> oerjan: and since I'm spilling tricks: The (n+k) pattern used to enable loops of this kind: main=readLn>>=f;f(n+1)=f n
13:47:07 <HackEgo> [wiki] [[Checkout]] http://esolangs.org/w/index.php?diff=40784&oldid=21322 * Ais523 * (+114) /* Checkouts */ fix an omission that was pointed out in talk
13:47:25 <b_jonas> int-e: ah, quitting by running into an undefined case. nice.
13:47:39 <oerjan> int-e: yeah it's really (0^n+n) vs. max 1n
13:47:41 -!- Lymia has quit (Ping timeout: 272 seconds).
13:47:55 <int-e> b_jonas: yes, that is common practice in the Haskell entries
13:48:07 <ais523> int-e: is the n+k pattern actually official Haskell?
13:48:13 <int-e> b_jonas: a lot of text processing is done by m@main=getLine>>=f>>m
13:48:25 <int-e> ais523: not anymore, it was removed in Haskell2010
13:48:38 <ais523> as being overly specific?
13:48:49 <ais523> you could still do a match against 0, 1, n
13:49:18 <oerjan> ais523: there are some old anagolf haskell solutions that seem impossible to tie, int-e theorizes the loss of n+k patterns is why
13:49:43 <int-e> the other contender is that import List not longer works.
13:49:53 <int-e> but I have no example for that
13:50:51 <mroman> with some ghc version you couldn't do import List anymore
13:51:01 <mroman> and I'm pretty sure I have submitted solutions that do import List
13:51:07 <mroman> (instead of import Data.List)
13:51:12 <mroman> but it doesn't work anymore on anagol
13:51:26 <b_jonas> well, these sound more likely than an entry that you can't tie because you can no longer define an instance of Monad without defining an Applicative instance for that type
13:51:38 <int-e> ais523: (n+k) patterns are a bit onerous on the compiler and gain little. f(n+k) is something like f n'|n'>=k = let n=n'-k in.... (or, in Haskell2010, f n|n<-n-k,n>=0)... using all of Num, Ord and Eq.
13:52:10 <int-e> ais523: One compiler author argued against plain n-patterns (f 0 etc) as well, but they were deemed too useful to drop.
13:52:36 <CADD> int-e: what was the argument?
13:52:49 <CADD> int-e: or even better, link to the article?
13:53:21 <int-e> CADD: lengthy discussion in 2010, I'm not sure how to feed google.
13:53:39 <CADD> int-e: kk, no worries
13:53:46 <CADD> int-e: mayne a tl;dr then?
13:54:08 <int-e> @google "remove npluskpatterns"
13:54:30 -!- Lymia has joined.
13:54:30 <int-e> there must have been a haskell' proposal for that.
13:54:54 <int-e> here: https://ghc.haskell.org/trac/haskell-prime/wiki/RemoveNPlusK
13:56:22 <int-e> CADD: The main reason was that the feature was hardly used and no fans stepped up to defend the feature, while enough people expressed their dislike.
13:56:33 <CADD> int-e: yup, ive seen that. what interested me is why n patterns are bad
13:57:11 <CADD> b_jonas: very few do
13:57:13 <int-e> CADD: oh from a compiler writing perspective they are about as bad as n+k patterns: to desugar f 0 = ..., you need to have something like f x | x==0 = ...
13:57:39 <int-e> CADD: from a programmer's perspective, they are just too useful to drop
13:57:52 <b_jonas> they, like, try to pretend that Integers are represented as a lazy unary data structure, when I actually want to use the fact thta they aren't, they're strict and implemented with machine-level arithmetic as binary
13:58:48 <int-e> > let f :: Double; f 0 = 1; f x = x in map f [-0.1,0,0.1]
13:58:50 <lambdabot> Couldn't match expected type ‘a0 -> a0’
13:58:50 <lambdabot> with actual type ‘GHC.Types.Double’Couldn't match expected type ...
13:58:51 <lambdabot> with actual type ‘GHC.Types.Double’
13:59:02 <int-e> > let f :: Double -> Double; f 0 = 1; f x = x in map f [-0.1,0,0.1]
13:59:07 <CADD> int-e: although im sure like n+k patterns, you could use viewPatterns as well, or at least that is my intuition
13:59:31 <int-e> (it was proposed to restrict n patterns to just Integral, but that also didn't gather a big fanbase)
14:00:00 <b_jonas> int-e: they should be restricted to a _unary_ natural type
14:00:01 <Lymia> elliott, I... just had a terrible idea for that cold golf
14:00:09 <Lymia> x86 using only printable characters.
14:00:14 <b_jonas> a unary lazy natural type (thus one that can store infinity too)
14:00:22 <ais523> Lymia: printable x86 is fun
14:00:49 <Lymia> Somebody already did it with 80386
14:00:51 <int-e> b_jonas: that was about n patterns. I hope you're not suggesting to abolish f (-1) = 42
14:00:55 <b_jonas> Lymia: people are doing that already
14:00:57 <Lymia> x86 counts as distinct, right?
14:01:28 <b_jonas> int-e: oh, n patterns. I dunno about those. I don't much like them, but they're not as bad as n+k patterns, so I think they can stay.
14:01:43 <ais523> XP_W^VH%35%DCPYXPH%=5%=CP[]UM#(UX%??t&* * * * ZR 1() !GFF=\ouU0_0<0^3L1L0^292L1^1Q1L2Y1D1\3R3P0A3B2D0<1p3p3o11131p3>2D0<3:0<18253<2:170D021p3D0>0A0D0<183<3:1:1D0432041p24143o031p2D0p331o0A2D0B2A3I1J2I1J2D321124310o13031D0=0p3o0302113A220=2I1J2I1J22112o0D011412112B02112o0D042A0D0432041p2B2B2D0o2B1I1J2n3I1J2p231o0o1p212D0B042=0?0B1B2B042B242A142B1B2B042A1@2B1=342B3@1m0m032p0o1o0p3B0D2<2@090@061@1;0@15050@382@380@0D2A0B0o2132130B2B1@263o251B1A042A1@
14:01:44 <ais523> 3=342B3@3=3I1J2B0B2B02112o0B2B1B2o2=1323331o043B0B0o2132130B2B1@163o251B1B1A1;193;2=311412112B011412112B0A1@3A0o2B1A0o2B1>2@1>2A132p0p22112o032B0B0o2132130B2B1@263o251A0I1J2D2C143C143D2A0C380C3B3C3=0C3A1B2B02112o0B2B1B2o2A3?0A1B0B2B02112o0B2B1B2o2=0C3p0C3B1B2C3<0C3A0C3@0C3B3B2B02112o0B2B1B2o2C1?0B1>2D1B0B0@163o251=1B0o2132130I1J2B2B13213p03112B0B0o2132130B2B1@163o251A0D2p242D2B1B1C2C2B0@263o251=1B0o2132130B2B13213p03112B0B0o2132130B2B1@
14:01:46 <ais523> 263o251A0D2o3B342D2B1B1C2C2@2@2B332p0o1o0p3B0I1J2D2608053735272:0C1@3?1C1;0<380616080:0C1=1?1C1=1<38063517191D3B07160C1=350=3o0C2B2D0B2D0B2D0B2D06282D0@1B0B1D0D1;3;2;2=1505013o171@053@0=0@052@390@1D2I1J2D290@052@2>1@290@152@1D2A0B0o2132130B2B1@263o251B1C2C211412112B0A1@1A0o2B1C2C232p0o1o0p3B0D2<0<0<1?1D2A0B0o2132130B2B1@263o251B1C2C2B0C3m2C3B332p3313123B0I1J2B0o2132130B2B1@263o251A0>0@1A0@0B1A0@
14:01:46 <int-e> Oh I used to have a DOS game of life program, text only, as a signature...
14:01:47 <ais523> 1B1C2C2B0B0B2B02112o0B2B1B2o2=132o0311010B0B0o2132130B2B1@163o251B1C1@2?1?2B1A011412112B0A1@3A0o2B1A0B0B2B02112o0B2B1I1J2B2o2=132o0311010B0B0o2132130B2B1@163o251B1A3?0A3?2?0B1A011412112B0A1@3A0o2B1A032431013p331B0B0o2132130B2B1@163o251B1A032431013p331B0B0o2132130B2B1@263o251I1J2B1B2@0B1B1>2@1>3I1J2m1I1J2@@A5
14:02:05 <int-e> ...it was even robust to some amount of whitespace change
14:02:33 <ais523> this program is my solution to "you want to send a file to a Windows system over a link that corrupts non-ASCII characters, there is no useful software at the other end"
14:02:36 <b_jonas> there was some recent article about x86 shellcode that looks like English words,
14:02:42 <ais523> actually, that's the output of running the program on its own source code
14:03:02 <ais523> (there's no whitespace past the start of the first line)
14:03:03 <b_jonas> as a research on how heuristic filters can be tricked or something.
14:03:24 <ais523> the main problem I found with ASCII 8086 is that you have no flow control instructions
14:03:35 <ais523> meaning that you need to self-modify the code in order to write them into your program
14:03:45 <Lymia> <ais523> this program is my solution to "you want to send a file to a Windows system over a link that corrupts non-ASCII characters, there is no useful software at the other end"
14:03:48 <ais523> which makes it quite easy to detect
14:03:49 <Lymia> Bootstrap with a .bat file?
14:03:58 <Lymia> AFAIK you could encode arbitary binary text through that.
14:03:58 <int-e> b_jonas: not so recent, perhaps :)
14:04:10 <ais523> Lymia: I don't know of a way to encode binary as a batch file
14:04:21 <Lymia> certutil -decode encodedInputFileName decodedOutputFileName
14:04:29 <ais523> actually I had to do exactly that once, except the target system was running Linux
14:04:33 <ais523> so I used busybox printf
14:04:47 <Lymia> certutil should be standard
14:04:48 <ais523> Lymia: hmm, now you're relying on external useful software
14:05:02 <b_jonas> I encoded a file with perl once
14:05:04 <ais523> I doubt certutil exists on DOS, and possibly not on Windows 95 either (which is about the age of this code)
14:05:22 <ais523> anyway, a proper encoder would bootstrap into proper uucode
14:05:37 <ais523> rather than the rather awkward "two bytes per character" of that encoding above
14:05:46 <b_jonas> in fact I think shar assumes the destination has uudecode
14:05:49 <ais523> (I think it alternates between 6 and 2 bytes)
14:05:55 <ais523> b_jonas: it does, if you encode binaries
14:06:01 <ais523> uuencode is actually in sharutils, packagewise
14:06:14 <Lymia> Isn't base64 standard on Linux though?
14:06:33 <oerjan> mroman: i think what happened with List was, at some point the current Prelude diverged so much from H98 that it could no longer be shared with the H98 compatibility layer, so all the old modules were put in the haskell98 package and hidden by default.
14:06:36 <b_jonas> Lymia: perl is standard on Linux and you can use it for this pretty well
14:06:41 <elliott> I mean, that's what the x means :p
14:07:10 <elliott> i386 is just an abbreviation.
14:07:24 <Lymia> This is probably going to end terribly, but
14:07:33 <Lymia> What if I tried to use Malborge in that challenge
14:08:55 <b_jonas> and m68k is a cpu arch entirely unrelated to x86, despite that it has "8" and "6" in it
14:09:04 <oerjan> b_jonas: you still can define Monad without Applicative in all officially released GHC versions, although that will change with 7.10; there's just a warning in the latest one. anagolf's still on 7.4, though.
14:09:50 <b_jonas> oerjan: oh... and can you even use that instance after you define it that way?
14:14:07 -!- AndoDaan has joined.
14:15:33 -!- AndoDaan_ has quit (Ping timeout: 272 seconds).
14:18:00 -!- ais523 has changed nick to ais523\unfoog.
14:20:10 <mroman> since x64 means 64bits
14:20:17 <mroman> x86 probably means it's got 86bits
14:20:19 <b_jonas> also x86 is derived from x80
14:20:36 <mroman> which had just 80 bits
14:21:32 <mroman> I a wouldn't know anything about these things
14:22:56 <int-e> ah I should've included a random number generator ... a black screen is all I get in dosbox
14:24:04 <int-e> anyway http://sprunge.us/cUIB is the code, executable as a .COM file.
14:24:34 <int-e> and one can change whitespace after the 64th character.
14:30:43 <b_jonas> mroman: about "bits", http://www.scottaaronson.com/blog/?p=1558
14:33:54 <ais523\unfoog> most subsequent revisions put digits in the middle
14:34:11 <ais523\unfoog> I think it goes up to 80686? but after a bit, people started using names rather than numbers, such as "Pentium"
14:34:53 <int-e> a good thing, too, it avoided the 801086 mess
14:38:40 <mroman> then with Pentium they started using numbers again
14:38:57 <mroman> then suddenly they started using letters
14:39:09 -!- FreeFull has quit (Ping timeout: 265 seconds).
14:39:20 <int-e> pentium, pentium pro, pentium mmx, pentium 3, pentium 4,
14:39:29 <int-e> I don't recall a pentium 2
14:39:57 <oerjan> b_jonas: Monad isn't a subclass of Applicative at all until next release, so yes.
14:40:25 <mroman> http://en.wikipedia.org/wiki/Pentium_II
14:40:40 <oerjan> b_jonas: of course the probability that a good golf solution involves defining a monad instance is rather low.
14:40:59 <int-e> mroman: thanks. the roman numerals look more familiar
14:47:55 <HackEgo> Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: <http://esolangs.org/>. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
14:48:26 <ais523\unfoog> oerjan: if the task was complicated enough, it might be worth it because of do-notation?
14:48:29 <mroman> dark blue on my black background isn't the best thing :)
14:53:19 <int-e> ais523\unfoog: still unlikely, since one can define 3 single character infix operators
14:56:59 <int-e> so in addition to do notation, you need code that requires lots of bindings, do a<-n;b<-o;c<-p;...;m<-z
14:57:56 <ais523\unfoog> a Monad instance doesn't actually have to follow the monad laws, right? Haskell won't try to prove it
14:58:05 <ais523\unfoog> so you could just define something sufficiently monad-ish that it parsed
14:58:07 <int-e> ais523\unfoog: right.
15:00:41 <mroman> I broke the monad-law and I won.
15:01:29 <mroman> Note to myself: Check quotes before saying them.
15:02:40 -!- AndoDaan_ has joined.
15:02:57 -!- AndoDaan has quit (Ping timeout: 255 seconds).
15:03:03 -!- adu has joined.
15:07:55 -!- Phantom_Hoover has joined.
15:10:35 <oerjan> ais523\unfoog: the thing is, you can get a lot of mileage in golfing for the cost of the characters "instance Monad where", and you'd probably need to define a data type as well...
15:12:26 <oerjan> do notation doesn't always beat explicit >>=, and only by a few characters. so, unlikely.
15:14:31 -!- `^_^v has joined.
15:18:34 -!- bb010g has quit (Ping timeout: 256 seconds).
15:22:25 -!- nyuszika7h_ has joined.
15:22:42 -!- bb010g_ has joined.
15:23:30 -!- FreeFull has joined.
15:24:21 -!- oerjan_ has joined.
15:24:30 -!- nyuszika7h has quit (Ping timeout: 256 seconds).
15:24:31 -!- oerjan has quit (Ping timeout: 256 seconds).
15:27:14 -!- nyuszika7h_ has changed nick to nyuszika7h.
15:29:19 -!- oerjan_ has changed nick to oerjan.
15:43:42 -!- CADD has quit (Remote host closed the connection).
15:43:47 -!- adu has quit (Quit: adu).
15:45:34 -!- AndoDaan_ has quit (Ping timeout: 258 seconds).
15:46:17 -!- Sprocklem has joined.
15:52:00 <Bike> http://33.media.tumblr.com/41aa1d635c8367c07bd174824fb09cc0/tumblr_nejfhwQqHS1qzcv7no1_1280.jpg Astrology is fun.
15:52:53 <HackEgo> [wiki] [[Musical notes]] http://esolangs.org/w/index.php?diff=40785&oldid=40698 * TomPN * (-57) /* Line functions */
15:54:37 -!- drdanmaku has joined.
15:57:09 -!- AndoDaan has joined.
15:58:20 <blsqbot> ERROR: Burlesque: (rm) Invalid arguments!
15:58:23 <blsqbot> ERROR: Burlesque: (rm) Invalid arguments!
15:58:29 <blsqbot> "9.00000000000000000000000000000000000000000000000000000000000000000000000000000
15:58:50 <blsqbot> Ain't nobody got time fo' dat!
15:58:52 <blsqbot> "23.1406926327792630000000000000000000000000000000000000000000000000000000000000
15:59:20 <blsqbot> ERROR: Burlesque: (P[) Invalid arguments!
16:00:09 <blsqbot> Ain't nobody got time fo' dat!
16:00:10 <elliott> Bike: I like how they circled two when all of them are equally absurd
16:00:34 <Bike> elliott: j-law will be wearing it next year
16:00:53 <mroman> fizzie: Is bfjoust fixed, btw?
16:02:44 <fizzie> Though EgoBot is back.
16:03:02 <fizzie> So you can (or, in fact, have to) submit to two hills in one command.
16:04:48 <mroman> !bfjoust cupnoodles (>)*9([(-)*8[+].]+>)*4(<)*4(<--<++)*4<--(>(+)*17>(-)*17)*4(>)*6(>[(+)*8[-].]>[(-)*8[+].])*-1
16:04:50 <zemhill> mroman.cupnoodles: points 7.67, score 29.82, rank 6/47
16:04:57 <EgoBot> Score for mroman_cupnoodles: 22.7
16:05:17 <mroman> where's the egobot hill located at?
16:06:01 <zemhill> ais523\unfoog: "!bfjoust progname code". See http://zem.fi/bfjoust/ for documentation.
16:06:02 <EgoBot> Use: !bfjoust <program name> <program> . Scoreboard, programs, and a description of score calculation are at http://codu.org/eso/bfjoust/
16:06:38 <mroman> only scores rank 40 on egojoust
16:12:00 -!- zzo38 has quit (Remote host closed the connection).
16:14:41 <fizzie> Could be something screwy with the zemhill scoring.
16:14:48 <fizzie> And, of course, they're quite different algorithms.
16:15:03 <fizzie> Points-wise it doesn't seem all that different.
16:15:44 -!- MDream has changed nick to MDude.
16:16:01 -!- mihow has joined.
16:19:55 -!- AndoDaan_ has joined.
16:21:59 -!- AndoDaan has quit (Ping timeout: 258 seconds).
16:22:56 -!- oerjan has set topic: See the fabulous redundant twins EgoBot and zemhill | BF Joust scoring poll: http://goo.gl/02KE0Y | https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/.
16:54:30 <newsham> tiny C compiler. https://github.com/rswier/c4/blob/master/c4.c
16:55:16 <Taneb> Interpreter, strictly speaking
16:55:40 <oerjan> @tell tswett The trick I discovered (and probably am not the first) for converting dates is that a multiplication and a division can give exact answers if you choose the right offset. in particular, for finding the year you want to treat january and february as belonging to the previous year, and putting the epoch at mar 1 in a 400-year like 1600 or 2000. there's a similar offset you can use for months within a year.
16:55:56 <oerjan> hum did that all get included
16:56:45 <newsham> ends with "within a year."
16:57:22 <oerjan> (i don't actually remember what the offset was, except that it was _not_ starting with march)
16:59:06 <oerjan> although i could rederive it, starting from march the day lengths are 31 30 31 30 31 31 30 31 30 31 31 2*
16:59:38 <oerjan> the february doesn't matter because we'll have cut the year there in a previous step
17:00:33 <oerjan> we see then there's a period there with july - december, 5 months
17:01:46 <oerjan> i think september = 0 will give the right offset
17:04:23 <oerjan> > let f m = (m-9)*(3*31+2*30)`div`5 in zipWith(-)`ap`tail$map f[3..14]
17:04:25 <lambdabot> [-31,-30,-31,-30,-31,-31,-30,-31,-30,-31,-31]
17:04:54 <oerjan> that's the essence of it.
17:06:12 <oerjan> @tell tswett it seems the offset for months is sep 1, then *(3*31+2*30)`div`5
17:09:27 <oerjan> i implemented this in schemenomic, way back.
17:10:28 -!- oerjan has quit (Quit: Later).
17:17:48 <int-e> eww... http://golf.shinh.org/p.rb?Wow
17:28:28 <Lymia> Doing the SO challenge in Malborge
17:28:32 <Lymia> Is proving surprisingly painful.
17:28:45 <Lymia> Because half the instructions are unusable at many points due to the restrictions on characters.
17:29:16 <Lymia> "#&'()*+,-./037;<=>@[\]^_ deopsvxXyz{|}~ being forbidden is generating huge streaks of unusuable characters.
17:33:25 <Lymia> http://codegolf.stackexchange.com/questions/40562/asciis-95-characters-95-movie-quotes
17:33:28 <ais523\unfoog> also, being surprisingly painful is the entire point of Malbolge
17:33:31 <Lymia> I'm writing a Malborge solution
17:33:33 <Lymia> <ais523\unfoog> also, being surprisingly painful is the entire point of Malbolge
17:33:47 <Lymia> Malborge would not be painful for writing a fixed printing program.
17:34:06 <int-e> In this case it seems malbolge's position-dependent decoding of instructions would be a blessing
17:34:19 <Lymia> int-e, it is, sort of.
17:34:23 <int-e> doesn't mean it's easy
17:34:38 <Lymia> Problem is, their selection of special characters means there's long runs where one instruction or another is just plain unavailable.
17:34:47 <ais523\unfoog> Lymia: I remind you that it took several months for the first hello world program in Malbolge, which was mixed-case and generated by an evolutionary algorithm
17:35:14 <Lymia> I have a position where the only available insturctions are / and u
17:35:24 <Lymia> Meaning I have to execute a p there
17:35:46 <int-e> I thought there were nops
17:36:02 <Lymia> Since I can only input "valid" instructios.
17:36:55 <ais523\unfoog> also, you can't use the non-ASCII hole because it's an ASCII tournament
17:37:25 <Lymia> http://pastebin.com/5fKv9cgS
17:37:32 <Lymia> To give an idea of what I actually have available at each position
17:38:26 <Lymia> I can never execute / since that reads stdin
17:38:29 <int-e> lazy k still works... `SK
17:38:31 <ais523\unfoog> 83 is by someone called "Snack", who knows of Esolang
17:38:41 -!- password2 has joined.
17:38:48 <Lymia> int-e, better do that now. :)
17:38:54 <Lymia> I intend to knock out ` with Malborge.
17:39:47 <int-e> But I can't be bothered.
17:40:29 <Lymia> http://pastebin.com/7X89ehSc
17:40:41 <Lymia> The top four instructions are the only ones I can really safely use whenever I like
17:41:14 <Lymia> 3 terrible positions.
17:41:23 <ais523\unfoog> oh, people are language-sniping, boring (apart from sniping the BF derivatives, I can get behind that)
17:41:23 <Lymia> Each of where I have to either jump (urgh, bad bad bad) or print (eeep)
17:41:36 <ais523\unfoog> Lymia: you could just try jumping and seeing where it goes
17:41:50 <ais523\unfoog> you can just pad from the source to the destinatino
17:42:01 <Lymia> ais523\unfoog, my current plan is to get d far into the padding
17:42:42 <Lymia> It seems that the padding usually looks something like this: 83 29443 83 29444 82 29444 83 29443 83 29444 82 29444 83 29443 83 29444 82 29444 83 29443 83 29444 82 29444
17:42:52 <FireFly> ais523\unfoog: plenty of people on PPCG knows about Esolang
17:43:03 <Lymia> So, if I get d to be around 29444, I can use j, *, and p to permute the values there.
17:43:08 <Lymia> Until I get the ASCII values I want to print.
17:43:20 <ais523\unfoog> PPCG is basically the Stack Overflow version of #esoteric
17:43:46 <Lymia> Since the large value would point into the part of the memory I'm using for the permutation, I could easily use j to loop back over the state I'm using once more
17:43:52 <ais523\unfoog> good luck, I hope you get there in time to not be sniped
17:44:04 <Lymia> Only way someone can snipe me
17:44:07 <Lymia> Is to get Malborge before me.
17:44:18 <ais523\unfoog> Lymia: or ban one of the characters you use, forcing you to rewrite the program
17:44:40 <Lymia> I'll probably want to hand-write initialization to get myself into the padding
17:44:49 <Lymia> But, a generator would be nice after that
17:45:32 <ais523\unfoog> what if all eight legal characters at any given position get banned?
17:45:45 <int-e> what other ideas do I have ... dc is going to work still, until all digits or P are gone.
17:46:21 <Lymia> ais523\unfoog, I'm screwed
17:46:25 <ais523\unfoog> don't you need at least some arithmetic? admittedly, dc has a lot
17:46:34 <ais523\unfoog> Lymia: then you'd better get this done quickly :-(
17:46:47 <Lymia> Every position has at least two characters open right now
17:46:53 <Lymia> So it'll take two bans to kill Malborge
17:47:10 <Lymia> There's malborge variants out there
17:49:54 -!- password2 has quit (Ping timeout: 255 seconds).
17:50:05 <AndoDaan_> There's also a BF variant that can be written with any character, it takes the length of your code and converts it to binary, and then to one of the 8 BF commands.
17:51:37 -!- AndoDaan_ has changed nick to AndoDaan.
17:51:49 <ais523\unfoog> AndoDaan: I was thinking of Unary, but that mandates 0s
17:51:57 <int-e> ok, two dc approaches: <n>d<m>%P (with some precalculation) <n1>ZP<n2>ZP... (where ni have the appropriate number of digits)
17:51:59 <Lymia> There's a character that'll force me to use jumps
17:52:05 <Lymia> Since it eliminates all safe instructions in a positino
17:52:09 <int-e> I couldn't do it without P.
17:52:32 <Lymia> int-e, I'll snipe P for you
17:52:50 <ais523\unfoog> Lymia: int-e: do it in reverse, submit a dc entry that makes most languages hard but Malbolge no harder than before
17:53:24 <Lymia> A number would do nicely
17:53:44 <Lymia> I'm not sure if I should go after ! or $
17:53:52 <int-e> how does one figure out which quotes are still available?
17:56:32 <Lymia> int-e, how about you get a number or something
17:59:46 <fizzie> int-e: But there's a P workaround right there in the man page!
18:00:05 <int-e> fizzie: [] are gone
18:00:22 <fizzie> The workaround also contains a P, which is something I hadn't noticed.
18:01:11 <fizzie> I guess that's reasonable, since you can't print anything else but numbers with anything else.
18:01:35 -!- Phantom_Hoover has quit (Ping timeout: 264 seconds).
18:02:25 <fizzie> ais523\unfoog: One (or more?) of the Unary clones allows any character.
18:02:38 <ais523\unfoog> so save that for #1 unless someone else snipes it first
18:02:58 <ais523\unfoog> ooh, what about 1L? that allows any two printables, IIRC
18:10:09 -!- AndoDaan has quit (Ping timeout: 264 seconds).
18:10:49 -!- AndoDaan has joined.
18:13:56 <int-e> Lymia: http://sprunge.us/GLDj is dc for quote 60, feel free to submit (I won't, my intellectual curiosity is satisfied). having 9 available simplifies matters but is not essential.
18:15:29 <int-e> Lymia: But I can do another if you want the "mess" one for "Malbolge".
18:17:40 <Lymia> Could you post it and strike Z?
18:17:45 <Lymia> I'm going to strike P next
18:18:56 <int-e> I will not post it.
18:20:06 -!- MoALTz has joined.
18:24:17 <int-e> oh I could use a newline instead of Ir. that makes me feel better.
18:26:34 <int-e> now to find a language that is turing complete with just tabs and newlines...
18:28:04 <int-e> ah. "but there may not be more that 64 tabs or 64 newlines in your program."
18:28:12 <int-e> somebody thought of that possibility.
18:33:41 <int-e> though technically that solution is invalid because it contains spaces. but those are not required.
18:40:50 <int-e> `WeLcOmE quintopia
18:40:50 <HackEgo> QuInToPiA: wElCoMe tO ThE InTeRnAtIoNaL HuB FoR EsOtErIc pRoGrAmMiNg lAnGuAgE DeSiGn aNd dEpLoYmEnT! fOr mOrE InFoRmAtIoN, cHeCk oUt oUr wIkI: <HtTp://eSoLaNgS.OrG/>. (FoR ThE OtHeR KiNd oF EsOtErIcA, tRy #EsOtErIc oN IrC.DaL.NeT.)
18:43:16 <int-e> Lymia: Oh I couldn't submit if I wanted to: dc has already been used.
18:52:34 <Lymia> My initializer works.
18:57:14 <int-e> oh fun. 64 bytes from 64.15.113.24: icmp_seq=293 ttl=55 time=27203 ms
18:57:51 -!- ais523\unfoog has quit (Read error: Connection reset by peer).
18:58:01 -!- ais523\unfoog has joined.
18:59:52 -!- ais523\unfoog has quit (Read error: Connection reset by peer).
19:00:03 -!- ais523\unfoog has joined.
19:00:55 -!- ais523\unfoog has quit (Client Quit).
19:12:12 -!- AndoDaan has quit (Ping timeout: 250 seconds).
19:16:13 -!- AndoDaan has joined.
19:21:11 -!- Sprocklem has quit (Read error: Connection reset by peer).
20:16:53 -!- Vorpal_ has changed nick to Vorpal.
20:22:19 -!- nycs has joined.
20:24:33 -!- `^_^v has quit (Ping timeout: 264 seconds).
20:25:46 -!- Sprocklem has joined.
20:32:36 -!- MoALTz has quit (Quit: Leaving).
20:35:49 -!- Sprocklem has quit (Ping timeout: 265 seconds).
20:52:54 <HackEgo> [wiki] [[Special:Log/newusers]] create * Notjohnconway * New user account
20:54:35 -!- atehwa has quit (Remote host closed the connection).
20:56:25 -!- Bicyclidine has joined.
20:57:25 <fizzie> Vorpal: For a short while.
21:00:34 <Vorpal> fizzie, How do you generate your site? Write raw html or some sort of site generator?
21:01:50 -!- Sprocklem has joined.
21:04:27 <fizzie> It's a bit of custom Python code, reading mostly raw HTML, with a rather simple include-with-parameter-replacement facility and some other such things.
21:04:39 <fizzie> (Also takes care of making the Atom feed.)
21:05:43 <fizzie> The bfjoust page I built with nanoc, which is a Ruby-based static-site generator, though it seems a bit overengineered.
21:20:30 -!- icedvariables has joined.
21:21:05 -!- AndoDaan_ has joined.
21:23:12 -!- AndoDaan has quit (Ping timeout: 264 seconds).
21:24:44 -!- icedvariables has quit (Ping timeout: 246 seconds).
21:26:01 -!- hjulle has quit (Remote host closed the connection).
21:27:31 -!- hjulle has joined.
21:37:01 -!- hjulle has quit (Quit: ZNC - http://znc.in).
21:38:18 -!- hjulle has joined.
21:49:04 -!- impomatic_ has joined.
21:56:35 -!- MickNartin has quit (Quit: Page closed).
22:03:05 -!- Bicyclidine has quit (Ping timeout: 260 seconds).
22:03:43 -!- hjulle has quit (Quit: ZNC - http://znc.in).
22:09:01 -!- oerjan has joined.
22:14:29 -!- hjulle has joined.
22:19:28 -!- Sprocklem has quit (Ping timeout: 245 seconds).
22:21:13 <oerjan> Lymia: i am pondering banning you until you learn to spell malbolge hth
22:21:58 <oerjan> (note: distance between pondering and actually doing is generally large)
22:25:00 <b_jonas> oh, the Musical Notes guy has returned
22:25:50 <b_jonas> and changed the definition so that the [ and ] instructions now owrk in Brainfuck except it still says "Loops cannot be nested."
22:36:16 -!- hjulle has quit (Quit: ZNC - http://znc.in).
22:40:31 -!- Phantom_Hoover has joined.
22:51:30 <nycs> http://www.bls.gov/OES/current/oes151131.htm
22:51:48 <nycs> apparently animal slaughtering is the top paying industry for computer programmers
23:10:56 -!- nycs has quit (Quit: This computer has gone to sleep).
23:39:01 -!- vanila has joined.
23:42:47 -!- Sprocklem has joined.
23:51:14 -!- Bicyclidine has joined.
23:56:03 -!- Sgeo_ has joined.
23:58:33 -!- Sgeo has quit (Ping timeout: 258 seconds).