00:05:14 -!- MoALTz has joined.
00:14:27 -!- derdon has quit (Remote host closed the connection).
00:20:32 <Sgeo> kallisti, update ng tan
00:41:26 <zzo38> Make something else.
00:48:32 <Gregor> Please sear this image into your mind: http://codu.org/hats/NewTopHat-med.jpg
00:48:49 <Gregor> Whenever you're doing something that Gregor would not approve of, just remember that image.
00:48:55 <Gregor> Together, we can all be more like me.
00:58:19 -!- [-away-] has changed nick to PiRSquared17.
01:02:47 <zzo38> OK I viewed it but I think that's all. I do not even necessarily know what you approve of, and in some circumstances I will not even care.
01:05:49 <kallisti> can anyone help a brother out with a congruence proof?
01:06:10 <kallisti> I feel like it's REALLY obvious.. but I'm unsure how to begin.
01:06:17 <calamari> your cute sister, on the other hand...
01:08:41 <kallisti> Let a and b be any integers. Prove that if a^2 + 2b^2 = 0 (mod 3), then either a and b are both congruent to 0 modulo 3 or neither is congruent to 0 modulo 3.
01:08:48 <kallisti> hmm, actually. I guess I should work backwards
01:08:56 <kallisti> because I can see how it works going backward... so I just... reverse that.
01:15:20 <fizzie> oerjan: As a special case of your thing, if g : R^+ -> R^+ continuous bijection, f(x) = { 0 if x = 0; -g(x) if x > 0; g^-1(-x) if x < 0 }. Might be useful if you want to have a specific sort of wiggle: just draw that in the x>0,y<0 quadrant, and mirror it over y=x to the x<0,y>0 one.
01:16:42 -!- Chef_ has joined.
01:20:09 -!- oerjan has quit (Quit: Good night).
01:52:38 -!- Phantom__Hoover has quit (Read error: Connection reset by peer).
01:58:35 -!- PiRSquared17 has quit (Read error: Connection reset by peer).
02:10:32 <zzo38> I wrote more of some D&D prestige class I was writing, still not yet complete.
02:16:29 <zzo38> How many tsume shogi games can be encoded on a QR code?
02:17:13 <myndzi> 'qr code' isn't very specific
02:17:21 <myndzi> there are different physical sizes and different amounts of error correction
02:17:35 <myndzi> assuming you mean the largest physical size and least error correction, i'm still not sure what tsume means :)
02:18:48 <zzo38> Tsume shogi is a kind of checkmate problem for shogi games. There is a limit to the number of moves (counting moves on both sides; so it is always odd number), and the first player must give check every turn, the first player's king is not part of this game, as well as a few other rules such as futile interposition.
02:19:05 <zzo38> Which level of error correction is suitable for a fax machine?
02:19:26 <myndzi> depends on if there are actual problems with the fax on the other side
02:19:32 <myndzi> i mean, it is a digital signal these days, yes?
02:20:17 <myndzi> so it's possibly 20 pieces on one side and 19 on the other
02:20:20 <myndzi> but probably much less ya?
02:20:27 <myndzi> can you set a hard limit on the number of pieces?
02:20:42 <zzo38> No I mean using a normal analog telephone service.
02:21:06 <myndzi> well, just print it big enough that any messiness from the fax will be minor compared to the size of the qr code's pixels
02:21:17 <zzo38> Yes, tsume shogi is 20 pieces on one side and 19 on the other. However, most of the pieces start off the board (exactly how many depends on the individual problem).
02:21:17 <myndzi> when i said physical size i mean dimensions in "pixels"
02:21:33 -!- sebbu has quit (Ping timeout: 276 seconds).
02:23:03 <myndzi> assuming bitstream because of probable compression
02:23:21 <myndzi> since captured pieces in shogi can be played again, i suppose you have to always account for all pieces
02:23:38 <myndzi> which means you would want to represent the location as one of 82 possibilities, but they are mutually exclusive
02:23:50 <myndzi> so you may be able to take advantage of that, likely other properties of shogi will make it easier
02:24:00 <myndzi> reducing the area or something
02:24:04 <myndzi> actually you could do that implicitly i bet
02:24:51 <myndzi> er, but you might want rectangles and you'd have to know which slice of the board it is anyway
02:24:57 <zzo38> Yes you do account for all pieces. Still, pieces off board on the same side which are the same kind of piece are interchangeable.
02:25:27 <zzo38> And I do suppose, you could do things such as restrict the king to start on one of the first five columns, for example.
02:25:30 <myndzi> gotta decide whether it's better to store state for pieces or pieces for locations
02:26:28 <myndzi> anyway, i guess range encoding makes the most sense, it's essentially specifying which permutation "this" is out of all possible permutations
02:26:43 <myndzi> and can benefit from statistical imbalance
02:26:45 <zzo38> There are compressions you could possibly do. I have made up a compression system for encoding sokoban levels once. One thing possible to do in tsume shogi, is you must ensure that the initial position is not check.
02:26:50 <shachaf> myndzi isn't exclusively a bot?
02:26:51 <myndzi> (large number of blank points on the board, for example)
02:27:26 <myndzi> i expect that the number of non-check positions is miniscule in comparison to the total number of positions and probably not worth it
02:27:38 <myndzi> unless you encode each piece in sequence as, like, a permuted value
02:27:43 <myndzi> first piece can be 1-81
02:27:49 <myndzi> multiply by 81, next piece can be 1-80
02:28:02 <myndzi> where each value is the number from one corner counting only available spaces
02:28:09 <myndzi> then you could implicitly exclude check positions
02:28:17 <myndzi> by simply not counting them in your tally
02:28:20 <zzo38> myndzi: Yes I have thought of that kind of thing, the factorial encoding.
02:28:45 <zzo38> (Not only for tsume shogi but for various purposes)
02:28:47 <myndzi> i expect that would give similar results to range encoding though, and range encoding might be simpler to handle
02:29:21 <myndzi> what you want to try to do with entropy encoding is somehow maximize the imbalance in the input sequences
02:29:37 <zzo38> I do not know the speed and stuff for computer to automatically calculate the number of moves. I do know a tsume shogi game for GameBoy, from Athena, which automatically responds to any move you might make (which must be check, of course), up to the maximum number of moves allowed.
02:29:58 <myndzi> i'm not really saying calculate all the moves
02:30:03 <myndzi> i'm saying calculate all the places one piece could be
02:30:11 <myndzi> then from the remaining places, calculate all the places the next piece could be
02:30:14 <myndzi> and encode those values
02:30:20 <myndzi> range encode those values
02:30:25 <myndzi> then you benefit from a bunch of sideboarded pieces
02:30:30 <myndzi> since they'll all be the same value
02:30:45 <myndzi> everything else gets encoded as a value between 0 and X where X is the max number of moves
02:30:52 <myndzi> this keeps lower numbers more likely, kind of
02:31:03 <myndzi> maybe don't use a left-to-right top-to-bottom thing, or use some form of prediction
02:31:04 <zzo38> Yes that can do. So that I can see, on average how many bytes; and then encode an entire book in this way to see the total length of the encoded data.
02:31:22 <myndzi> i did something very similar for tetris fields
02:31:34 <myndzi> can get like ... 8 byte(?) encodings
02:31:39 <myndzi> i forget the exact numbers atm
02:31:45 <myndzi> maybe it was more like 16
02:32:06 -!- amca has joined.
02:32:29 <myndzi> anyway, if you are preencoding everything
02:32:36 <myndzi> you can choose the optimal values for best compression
02:33:06 <myndzi> if you are doing many over time, it might be simpler to share a set of probabilities calculated from a large sample set
02:33:09 <myndzi> then you don't have to send that data
02:33:24 <myndzi> or finally, you can do adaptive encoding where you base the proportions on the data you've decoded so far
02:33:32 <Jafet> You are planning to fax QR codes?
02:33:54 <myndzi> the question i'd like to attack is, how better to encode piece locations
02:33:58 <myndzi> to get more redundant data
02:35:16 <myndzi> you could get really tricky and do something like
02:35:47 <myndzi> order the locations by representing squares that are likely to have pieces (squares that protect pieces of your side)
02:36:02 <myndzi> (or squares that are not directly threatened by the other side)
02:36:15 <myndzi> that'd gain you some, but i'd like it if there was something better than that
02:36:29 <myndzi> perhaps order them within those two divisions by distance from the initial piece
02:36:46 <myndzi> or the previous piece, or some coordinates
02:36:58 <myndzi> probably coords since the piece order is fixed, then you can further reduce the values that encode locations
02:36:59 <Jafet> Strange positions tend to come up in puzzles as often as they don't appear in games
02:37:13 <myndzi> right, but given an arbitrary set of pieces
02:37:24 <myndzi> do you think it's likely or unlikely that many pieces are all threatened at once?
02:37:42 <Jafet> For a small set of pieces, rather likely
02:37:43 <myndzi> and given the nature of the problems, do you think it's likely or unlikely that the pieces will be spread out or more clustered?
02:38:05 <Jafet> I don't play shogi, but I'd assume they can be arranged either way.
02:38:09 <myndzi> ah, i sort-of tangled "threatened" with "undefended"
02:38:22 <myndzi> by threatened i meant in danger of capture, in a bad position
02:38:32 <myndzi> so i guess you could flag "protected" pieces and use that to identify danger
02:39:17 <myndzi> another possibility is to create an encoding in which intelligence in choosing the input values can reduce the size
02:39:29 <myndzi> that is, there are many orders for the input, but the program can choose the one that yields the best results
02:39:39 <myndzi> have to offer features for the encoder to take advantage of in that case
02:39:47 <myndzi> i'm not quite sure what those could be
02:40:10 <myndzi> you could possibly do something like ordering pieces along their threat areas
02:40:15 <myndzi> and creating a chain of pieces in sequence
02:40:21 <myndzi> that all can "attack" the next one
02:40:32 <myndzi> make the longest chains = best compression, since the range of values is much smaller
02:40:44 <myndzi> but you gain some weight by having to specify WHICH piece
02:41:40 <zzo38> Jafet: For something entirely different (not tsume shogi) I am planning to fax QR codes, or some kind of barcode anyways; it doesn't have to be QR code. The only thing is so that the sender can receive the same page possibly with marks filled in by pencil and understand (so, the codes will be twice faxed)
02:41:40 <myndzi> and it was a bit silly of me to say that the same-encodings of the out of play pieces will compress good for being the same value, since the out-of-play pieces can be derived implicitly, though i suppose you have to identify which side they are captured by
02:42:03 <myndzi> qr codes can actually be quite resistant to markings
02:42:11 <myndzi> it's quite possible you could mark up and only fax once
02:42:36 <myndzi> the format is pretty interesting
02:42:40 <zzo38> myndzi: In general, most of them belong to the second player (the player having the king).
02:42:51 <myndzi> the data is ordered to present the least chance of damage in one physical location damaging the data
02:43:17 <zzo38> myndzi: It isn't the QR codes which will be marked up by pencil; it is other stuff similar to scantron or something like that
02:43:18 <myndzi> at the highest error correction iirc you can lose half the data and still be ok
02:44:03 <myndzi> zzo38: in that case the smallest encoding might be something like, a list of pieces captured by the attacking player
02:45:12 <zzo38> myndzi: Yes, it is what I was thinking. Even when written on paper, the pieces available for the second player to drop are not mentioned since they are implied
02:46:04 -!- sebbu has joined.
02:46:04 -!- sebbu has quit (Changing host).
02:46:05 -!- sebbu has joined.
02:51:05 <myndzi> grr this is getting annoying
02:51:21 <myndzi> any ideas on isolating an intermittent hardware problem (system freeze)?
02:52:42 -!- sebbu2 has joined.
02:52:42 -!- sebbu2 has quit (Changing host).
02:52:43 -!- sebbu2 has joined.
02:56:00 -!- sebbu has quit (Ping timeout: 240 seconds).
02:56:08 -!- sebbu2 has changed nick to sebbu.
02:59:50 <ais523> hmm, do you get anything useful if you point a barcode reader at a zebra?
03:00:08 <ais523> this sounds like a "silly source code" idea
03:00:21 <ais523> writing the program is easy, but you have to find an appropriate zebra to input it to the interpreter
03:23:46 -!- nooga has quit (Ping timeout: 260 seconds).
03:30:50 <ais523> zzo38: no, I don't have any zebras spare
03:37:26 <zzo38> ais523: Can you use some pictures of zebras from Wikimedia Commons or whatever?
03:37:46 <ais523> hmm, I'm not sure I have an appropriate barcode scanner eitehr
03:38:53 <zzo38> There are computer programs which will do it, such as zbar; possibly you can cut out part of it to make it like a proper barcode see if it scans
03:44:12 <ais523> hmm, the information content of a zebra barcode is quite low
03:44:23 <ais523> so presumably you'd make it an alternate syntax for GolfScript or a similar language
03:46:40 <zzo38> First you should try it to see if you can get anything at all
03:51:29 -!- calamari has quit (Quit: Leaving).
04:02:23 <Gregor> Part of any barcode is marks that are consistent across all barcodes of the same style, so that the scanner can compensate for size/distance.
04:02:29 <Gregor> So, no, you can't scan a zebra.
04:02:39 <Gregor> Not unless it happens to be of the rare barcode breed of zebra.
04:02:52 <pikhq_> Do I hear a breeding project?
04:03:42 -!- Gregor has set topic: Breeding zebras as an information storage medium | http://codu.org/logs/_esoteric/.
04:06:10 <zzo38> Then invent a new barcode format.
04:06:52 <Jafet> Or the better-sounding bracode
04:07:12 <zzo38> (I also think of a similarity to the chiromancy idea I have posted on esolang wiki list of ideas a while ago)
04:08:19 <pikhq_> Barcodes: silliest backup medium?
04:08:34 <Jafet> No, that goes to DVDs.
04:08:50 <quintopia> archival ink is known to last a long time
04:09:03 <pikhq_> quintopia: Well. There is that.
04:09:12 <pikhq_> Qr code is a form of barcode.
04:09:53 <pikhq_> "barcode" in modern use refers to any optical machine-readable data storage.
04:10:24 -!- Chef_ has quit (Remote host closed the connection).
04:10:37 <Jafet> DVDs rot, are cumbersome, and have no integrity monitoring
04:10:59 <zzo38> I first learned about 2D barcodes from an issue of 2600
04:11:09 <pikhq_> Jafet: They're not even the silliest quickly-rotting backup medium!
04:11:27 <Jafet> Well, they're the silliest that people seriously use
04:11:59 * Jafet brandishes an Iomega product
04:12:38 <pikhq_> Also, DVDs aren't *too* bad if you actually work with them. Like, I dunno. Write on multiple discs, replace yearly, verify checksums?
04:13:00 <pikhq_> Especially considering DVDs are tiny.
04:13:12 <Jafet> A jukebox mechanism renders any possible cost savings moot
04:13:21 <pikhq_> You'd be better off doing something like, say, a few external hard drives.
04:14:13 <pikhq_> This also has the advantage of not being insanely slow.
04:15:19 <pikhq_> Anyways. Silly backup media? Heck, let's go insane. Delay lines.
04:15:40 <pikhq_> "I've got a lightminute of data storage here!"
04:16:05 <Jafet> Mercury was cheaper than silicon wafers!
04:17:09 <pikhq_> And so was antimatter, your point? :P
04:17:32 <Jafet> This reminds me of a company touting their battery-backed SDRAM drive to be great for swap
04:18:00 <pikhq_> Well, yes, RAM is wonderful for swap. :P
04:28:43 <Gregor> Apparently a "type 40" QR code can store up to 2,953 characters. It's very dense. Assuming you could print, say, twelve of those at readable DPI per side of a sheet of paper, you've got about 69KB per sheet of paper. That means a floppy disk is 21 pages.
04:29:46 <pikhq_> That means that a complete, if minimal, Linux install would be ~21 pages.
04:30:49 <Gregor> The source code for the linux kernel, compressed, would be at least 750 pages though :)
04:30:49 <pikhq_> Also, because QR codes make this relevant: 2,953 *octets*.
04:31:08 <zzo38> Do you mean letter size paper?
04:31:38 <Gregor> zzo38: Since it all comes down to what DPI you can print and what DPI you can scan, I'm not really specifying a paper size, just estimating how many QR codes you could fit on some "reasonable" size.
04:31:53 <pikhq_> You can also store numeric, alphanumeric, and kana/kanji data, each with varying amounts of storage.
04:32:27 <pikhq_> But, it's just 2,953 octets, so. :)
04:32:35 <zzo38> I looked at some QR code programs they have many files needed to encode it properly. I once wanted to make a QR encoder in TeX. I do not even know how.
04:34:40 <zzo38> But I did write a POSTNET encoder in TeX.
04:39:43 -!- Chef_ has joined.
04:45:53 -!- Chef_ has quit (Remote host closed the connection).
04:48:19 -!- Chef_ has joined.
04:50:59 <pikhq_> http://gayhomophobe.com/ Best thing?
04:51:19 <pikhq_> (maintains a count of the days since a homophobe was caught in a gay sex scandal)
05:00:04 <ais523> pikhq_: you're not going to find a QR code on a zebra :)
05:00:35 <pikhq_> Not without breeding zebras as an information storage medium.
05:03:29 <itidus20> i think google is so busy advertising it's privacy policy because it's gonna screw everyone in the ass
05:03:45 <itidus20> if there was no downsides i don't think they would promote it so much
05:12:45 <ais523> itidus20: the advertising is useful, it let me know how to delete my Google account, so I did
05:13:50 <itidus20> its not that i don't like google.. just that i feel like all this effort to advertise some major changes to their privacy policy is probably really bad news for me
05:14:26 <Gregor> When they changed it, people started to rumble about how they were trying to sneak it under everyone's noses.
05:14:35 <Gregor> I'm pretty sure they started advertising it mainly to say "No! No! We're not!"
05:14:48 <itidus20> well whats the worst they can do i guess
05:14:49 <Gregor> They kinda went overboard though.
05:15:21 <itidus20> . o O ( /join #google-privacy-policy ) lol
05:16:29 <Gregor> My understanding is that previously their privacy policies were self-contained, which is to say that your information from one Google service usually couldn't be used in another (depending on the two services in question), and furthermore every service had its own privacy policy. They integrated it in both senses: There is only one Google privacy policy now, but it covers all Google services, so they're universally allowed to share amongst themselves.
05:17:33 <Gregor> It's the kind of thing that I just assumed they were doing in the first place, I was shocked by the suggestion that they might not have been X-D
05:18:44 <itidus20> if they do anything wrong we should just email them your discouragement photo
05:21:06 <ais523> Gregor: I like the theory that they're trying to discourage a lawsuit
05:21:57 <zzo38> Do you enable show-punycode on your computer?
05:22:02 <ais523> but it reminds me slightly of the VPSAD mousetrap I pulled off once, when you make absolutely sure that you've let someone know the potential consequences of someone, so that you can actually make said potential consequences happen later and they can't deny it
05:23:47 -!- Chef_ has quit (Remote host closed the connection).
05:24:59 -!- Chef_ has joined.
05:31:18 -!- Chef__ has joined.
05:34:01 -!- Chef_ has quit (Ping timeout: 240 seconds).
05:46:26 -!- PiRSquared17 has joined.
05:57:56 <zzo38> I wonder, if it is ever possible to homograph attack within the Dungeons & Dragons game
06:04:24 -!- pikhq has joined.
06:05:10 -!- pikhq_ has quit (Ping timeout: 276 seconds).
06:05:17 -!- augur has joined.
06:08:08 <pikhq> Apparently in the US up until the 70s there were life insurance vending machines.
06:08:20 <pikhq> In an airport you could buy life insurance for the duration of your flight.
06:08:42 <pikhq> From a vending machine.
06:09:04 <quintopia> but also life insurance used to be very different
06:09:20 <pikhq> It was apparently ridiculously profitable.
06:09:26 <pikhq> Which makes sense.
06:09:45 <pikhq> Given the safety record of flight, that's nearing on "license to print money".
06:10:30 <quintopia> its no better than the record of cars if you look at it closer
06:10:53 <quintopia> just because cars are a more likely form of death doesnt mean theyre less safe on average
06:11:24 <quintopia> i dont have numbers on me, but ive seen them
06:11:27 <pikhq> How, precisely, are you measuring their safety? Risk of death per time spent in the vehicle or some such?
06:11:58 <Jafet> Safety is usually measured in unpleasant events per unit distance per passenger
06:12:01 <pikhq> Course, I'd say that selling a life insurance policy for the duration of a car trip would *also* be ridiculously profitable.
06:12:15 <pikhq> People usually don't die from car accidents. :)
06:12:44 <zzo38> Use extremely convoluted plans to somehow get dead during the flight, also killing a single person who happens to be on that flight that you and many other people also really want dead
06:13:23 <pikhq> And be sure to be signed up for cryonics, and will your life insurance winnings to be held in trust for you.
06:13:44 <pikhq> It's a several-win scenario, if it works.
06:14:25 <pikhq> Course, if you really wanted everyone else on the plane to be dead, you're still winning.
06:15:07 <quintopia> if i sold life insurance, it would contain a reclamation clause. becoming undead means unclaiming the insurance money. you have to pay it back with interest for the time you spent dead.
06:16:17 <Jafet> What if a dead guy gets snap-cloned
06:16:55 <pikhq> quintopia: Yeah, but alas, most life insurance goes with legal death.
06:17:26 <quintopia> Jafet: including mental state at time of death?
06:18:23 <zzo38> I did not even mean necessarily everyone; if you could get some people surviving and others dead possibly if you can figure out a way to do so. While not allowing anyone to know how the plan was done because it is different from everything else.
06:18:23 <Jafet> The exact state is probably unimportant.
06:19:40 <quintopia> if its different enough to constitute a different person, that would be a new person without any policy
06:29:16 <zzo38> But can you define "a different person"? There is even philosophical, such as Ship of Theseus, and if you have iron in two things in two places and it immediately switches places, and if a person is different when they are grow up or not, and so on
06:29:55 <Jafet> You don't need to, just invent snap-cloning and leave the question to insurance company lawyers
06:30:07 <Jafet> Shotgun philosophy
06:30:55 <pikhq> The lawyers will find a way to define a "different person", and sanity be damned.
06:31:32 <Sgeo> kallisti, update
06:32:46 <zzo38> That might work in law, but philosophers will realize it is wrong and think of different ways
06:33:38 <zzo38> (And that applies to both sane and insane philosophers; so you are damned either way...)
06:34:17 <pikhq> Yeah, but the philosophers are liable to spend centuries arguing about it. :P
06:34:36 <zzo38> pikhq: Yes of course that is now philosophy works
06:35:05 <zzo38> And it is a good thing that philosophy works that way
06:36:51 <zzo38> At least in my opinion.
06:39:36 <kallisti> when you could type in google searches
06:39:45 <kallisti> and then click on a different kind of youtube search
06:39:55 <kallisti> GONE ARE THOSE DAYS, APPARENTLY?
06:40:23 <kallisti> it's when you click the search in the top panel
06:40:32 <kallisti> wow that's really fucking weird.
06:41:07 <pikhq> Remember the days when it was possible to search for an exact string in Google?
06:42:13 <pikhq> No, Google got "smart" and that doesn't necessarily work.
06:43:34 <ais523> pikhq: we actually found a trick for searching for an exact word without it being autocorrected
06:43:37 <ais523> put extraneous hyphens in it
06:43:46 <ais523> (yes, this is ridiculous)
06:44:59 <Jafet> You puny humans and your pathetic attempts to outsmart google
06:50:36 -!- Chef__ has quit (Remote host closed the connection).
06:51:42 -!- itidus20 has quit (Ping timeout: 245 seconds).
06:53:21 <zzo38> Google has been getting worse all the time since they first existed....................!!!!!!
06:55:21 <myndzi> no kidding @ exact searches
06:55:30 <myndzi> i ran into one the other day that blatantly ignored -"phrase"
06:55:39 <myndzi> turns out something superceded it, reversing the order worked
06:55:52 <myndzi> i found a search last week that returns only youtube blended results
06:56:01 <myndzi> no organic results at all on the entire first page
07:01:33 <zzo38> Do what was suggestion in 2600 once; make a program which occasionally sends bogus requests to Google
07:04:30 <myndzi> dunno what that'll do to help anything
07:11:29 -!- pikhq_ has joined.
07:14:17 -!- MoALTz has quit (Ping timeout: 260 seconds).
07:14:37 -!- pikhq has quit (Ping timeout: 272 seconds).
07:15:41 -!- MoALTz has joined.
07:21:34 <pikhq_> Latin transcribed into katakana is... So very strange.
07:21:42 <pikhq_> ステラ ノビス ノン コンチェスィト ノン コンチェスィト
07:22:50 <myndzi> wish i could still read that haha
07:23:08 <pikhq_> "stera nobisu non konchesito non konchesito" in wapuro romaji, "sutera nohìsu nonn konntiēsuīto nonn konntiēsuīto" in pikhq romasì, "Stella nobis non concessit non concessit" in more normal orthography.
07:24:05 <myndzi> i suppose it has to do with lots of consonant endings in latin
07:24:54 <pikhq_> And Japanese has this funny pattern of everything being "consonant-vowel".
07:25:03 <pikhq_> That produces some really strange transcriptions.
07:25:09 <myndzi> consonant-consonant will do it too
07:26:30 <myndzi> i've gotta learn kanji sometime soon
07:26:37 <myndzi> i've now bought at least 4 manga series that i can't read
07:26:59 <myndzi> *.kitoh, also gunnm 'cause it's out of print
07:27:22 <myndzi> i can't say i have great taste haha
07:27:29 <myndzi> but i wanted to own them
07:27:30 <pikhq_> Also, trust me, you want to learn kanji if you want to speak Japanese. Some people will tell you you can get away with it.
07:27:32 <myndzi> i do want to get kknj too
07:27:33 <zzo38> I have all the Akagi manga series so far, including the special editions.
07:27:47 <pikhq_> You can't read fucking *Youtube comments* without kanji.
07:27:51 <myndzi> oh i said i ought to learn *kanji*
07:28:03 <myndzi> though i would like to do that too at some point
07:28:14 <pikhq_> Japanese ain't that hard.
07:28:16 <myndzi> i suppose i'll need some words here and there for stuff that doesn't have kanji
07:28:25 <myndzi> yeah, it's more of commitment and time that's holding me up
07:28:43 <myndzi> i'm hoping that maybe duolingo will add japanese soon
07:28:59 <zzo38> I can understand some kanji, as well as knowing the hiragana and katakana entirely. Mixing kana with kanji does actually make it to read fast and better even though some people say it isn't. I know this because of my own experience
07:29:40 <myndzi> well, for a little while i was going through remembering the kanji, with the aid of that one website... has a derivative name, can't think of it at the moment
07:29:43 <myndzi> spaced repetition works good
07:29:52 <myndzi> i remember all the kanji i picked up during that, it's just not many
07:30:08 <pikhq_> 漢字とカタカナとひらがなで書いて本当に読むのが速いぜ！
07:30:12 <myndzi> reviewing the kanji = website
07:30:13 <pikhq_> myndzi: Reviewing the Kanji?
07:30:36 <myndzi> dunno how frequent compounds are, though, and i bet rtk doesn't teach that
07:30:40 <myndzi> at least they'll be somewhat obvious i think
07:30:47 <myndzi> i mean, obvious that it's a compound
07:31:04 <pikhq_> Compounds are highly frequent and highly obvious.
07:31:32 <pikhq_> And a significant amount of them are also fairly simple in construction.
07:32:27 <myndzi> well that's encouraging anyway
07:32:34 <pikhq_> 販売機 "trade sell machine" -> "vending machine"
07:32:52 <myndzi> but yeah, uh, my budding library isn't exactly easy reading
07:32:59 <myndzi> i should get yotsuba& :>
07:34:59 <pikhq_> Yeah, that's quite easy reading.
07:35:32 <myndzi> but it's written with an &!
07:35:44 <myndzi> i just can't write hiragana or whatever
07:35:54 <pikhq_> No, that's a translation artifact. :P
07:36:32 <myndzi> though it makes more sense
07:37:17 <myndzi> my computer is freezing every ~hour now
07:37:23 <myndzi> hopehopehope it's the video card
07:45:01 <zzo38> When writing Japanese texts by hand I will use katakana because I find hiragana more difficult to write
07:45:02 -!- amca has quit (Quit: Farewell).
07:45:22 <zzo38> (Although I can read hiragana OK)
07:45:26 <myndzi> i mean, i don't have an ime installed or know the syllables to write with
07:45:36 <myndzi> so i had to write it gringo-style
07:50:40 -!- monqy has quit (Quit: hello).
07:50:57 -!- cheater has joined.
08:15:51 <pikhq_> Funny, when writing English texts intended for myself by hand I will use kanji because I find it easier to write. :P
08:21:06 <zzo38> pikhq_: I will sometimes do that too; but I also use English abbreviations (often shorter than a standard abbreviation), mathematical notation, and various other things.
08:21:24 <pikhq_> Actually. Yeah, that's basically my note-taking.
08:21:40 <pikhq_> I get some insanely dense notes.
08:22:20 <ais523> hmm, surely there are quite a few things expressible in English that can't be losslessly expressed in either kanji or kana?
08:22:44 <zzo38> As well as things I just omit because the way I think, they are implied.
08:22:46 <ais523> if there's not a corresponding Japanese word, there won't be a kanji, and English and Japanese use different alphabets, in terms of which syllables are expressible
08:23:48 <zzo38> ais523: I think that is the case in any language where some things are impossible to translate exactly, both in meanings and in syllables.
08:23:51 <pikhq_> ais523: First, I end up using the kanji ad hoc to map to morphemes, second, it's used not entirely.
08:24:01 <pikhq_> Just like kanji are used in Japanese, actually.
08:24:16 <ais523> so it's a substitution cipher that uses kanji?
08:25:02 <pikhq_> I should note that this is literally how Japanese is written.
08:25:34 <zzo38> Sometimes I write my name using the kanji for black
08:29:21 <Sgeo> I'm going to cry at the number of times I see languages contrasted against Java.
08:29:40 <Sgeo> I'm not comparing you to Java, I already think that Java's crap, how do you stack up against a _good_ language?
08:30:29 -!- zzo38 has quit (Quit: It's...).
08:33:28 <ais523> it'd make sense to compare on performance
08:33:39 <ais523> apart from startup time, Java's pretty fast, but often pretty memory-hungry
09:11:35 -!- lambdabot has quit (Read error: Operation timed out).
09:12:49 -!- monqy has joined.
09:15:20 -!- lambdabot has joined.
09:53:15 -!- nooga has joined.
10:00:47 -!- Phantom_Hoover has joined.
10:43:53 -!- azaq23 has quit (Quit: Leaving.).
10:52:04 -!- tzxn3 has joined.
11:00:34 -!- oerjan has joined.
11:02:12 <oerjan> i suspect i should have replaced AVG a long time ago.
11:03:08 <oerjan> (an antivirus that become gradually more annoying over the years)
11:03:41 <ais523> Microsoft Security Essentials is reasonably non-obnoxious, and apparently also quite good at catching malware (although not the best)
11:04:06 <oerjan> then they made the mistake of sneaking their toolbar past me, and when i discovered that it looked easier to replace the whole program than uninstalling just the toolbar...
11:04:58 <oerjan> and yes, i went to MSE. that "non-annoying" on their page was _so_ alluring.
11:08:44 -!- monqy has quit (Quit: hello).
11:13:06 <oerjan> <zzo38> There are computer programs which will do it, such as zbar; possibly you can cut out part of it to make it like a proper barcode see if it scans
11:13:28 <oerjan> ...clearly that program name is missing an e somewhere.
11:13:53 <oerjan> serious missed pun opportunity.
11:16:57 <oerjan> oh, there is one actually called zebra.
11:17:26 <ion> zebra would be so easy to google.
11:18:15 <oerjan> ion: i was thinking more like ezbar or so
11:18:39 <ion> Not zbear?
11:18:59 <oerjan> > permutations "zebra"
11:19:00 <lambdabot> ["zebra","ezbra","bezra","ebzra","bzera","zbera","rbeza","breza","berza","r...
11:19:34 <ion> > (permutations <=< subsequences) "zebr"
11:19:35 <lambdabot> ["","z","e","ze","ez","b","zb","bz","eb","be","zeb","ezb","bez","ebz","bze"...
11:20:36 <lambdabot> Data.List subsequences :: [a] -> [[a]]
11:21:01 <oerjan> WHAT WAS WRONG WITH filteM (const [True, False]) ?
11:21:47 <lambdabot> [[],[1],[2],[1,2],[3],[1,3],[2,3],[1,2,3],[4],[1,4],[2,4],[1,2,4],[3,4],[1,...
11:22:03 <oerjan> i guess that doesn't work on infinite lists.
11:23:50 -!- Taneb has joined.
11:24:54 <Taneb> Why the sudden 'stral'ness, mate
11:25:24 <oerjan> crikey, can't a mate speak a bit strine?
11:26:48 <fizzie> Something something barbie something.
11:26:59 <oerjan> btw is http://www.mezzacotta.net/garfield/?comic=1004 broken for everyone else?
11:27:44 <oerjan> i get a broken picture :(
11:28:59 <oerjan> even directly at http://www.mezzacotta.net/garfield/comics/1004.png
11:29:30 <oerjan> AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
11:31:04 <fizzie> oerjan: How's http://zem.fi/~fis/1004.png for you?
11:31:50 <fizzie> Wuird. Maybe there's something STRANGE about the PNG.
11:32:27 <fizzie> $ convert 1004.png 1004.gif
11:32:27 <fizzie> convert: Incorrect bKGD chunk length `1004.png' @ warning/png.c/PNGWarningHandler/1457.
11:32:28 <oerjan> i noticed it the day it went up, but thought it would surely be fixed by now, but if only i see it...
11:32:57 <fizzie> Your PNG-grokking thing must be pickier than others'.
11:33:23 <fizzie> Anyway http://β.zem.fi/~fis/1004.gif
11:34:42 -!- Jafet1 has joined.
11:34:54 <oerjan> hm it's also picky about showing that β
11:35:09 <oerjan> at least it allowed it as input.
11:35:18 -!- Jafet has quit (Ping timeout: 252 seconds).
11:35:37 <fizzie> I'm sort-of regretting that β. It seemed like a good idea at the time, but seeing it written "xn--nxa" all the time in Chrome/Chromium is being the annoy.
11:35:54 <fizzie> And apparently in IE too.
11:36:19 <oerjan> well it _is_ suggesting i change the language settings to allow it.
11:36:57 <fizzie> Chrome's rule too is "show characters in the language the user can read", approximated somehow.
11:38:05 <fizzie> While Firefoxy things whitelist all .fi TLD addresses, which doesn't really make sense for weird things not in the second-level domain. (They whitelist all TLDs that have a policy for misleading special characters in place.)
11:40:11 -!- Phantom_Hoover has quit (Remote host closed the connection).
11:40:52 <fizzie> Personally I think they could just allow everything below the second level; even if I were to write "finnishbankinginstitution" with some homographs, a) "finnishbankinginstruction.zem.fi" wouldn't do me much good, and b) I could do it with the real characters just as easily.
11:41:42 -!- Phantom_Hoover has joined.
11:41:43 <fizzie> But I suppose they're trying to fix "looks weird", not "can be used for sneakishness".
11:43:14 <fizzie> I wonder if IE would allow the beta-like 'ss', which is in latin-1. Chromium didn't, for me. (On the other hand, I don't think it allowed ä either, I must have it set in some English mode.)
11:48:06 <oerjan> <ais523> but it reminds me slightly of the VPSAD mousetrap I pulled off once, when you make absolutely sure that you've let someone know the potential consequences of someone, so that you can actually make said potential consequences happen later and they can't deny it
11:49:36 -!- Jafet1 has changed nick to Jafet.
11:51:47 <fizzie> Vice President of sadness, maybe?
11:52:51 <fizzie> (Vice President of Student Activities and Development, suggests G.)
12:16:22 -!- ineiros_ has quit (Ping timeout: 252 seconds).
12:41:44 <ais523> only nomic-style scam I've done in real life
12:41:53 <ais523> I didn't /do/ anything with it, just wanted to see if it was possible
12:42:02 <ais523> (also, it only lasted one year, probably)
12:42:18 <ais523> (due to real life's rule 101 equivalent)
13:04:11 -!- oerjan has quit (Quit: leaving).
13:27:36 <ais523> ooh, Reddit's reflecting #esoteric, in a bit
13:27:44 <ais523> we were discussing typed Underload recently
13:28:04 <ais523> and someone went and implemented a simple typed concatenative language in Haskell
14:15:27 -!- Jafet has quit (Read error: Connection reset by peer).
14:15:27 -!- Phantom__Hoover has joined.
14:15:27 -!- tzxn4 has joined.
14:15:27 -!- Phantom__Hoover has quit (Changing host).
14:15:27 -!- Phantom__Hoover has joined.
14:15:49 -!- Jafet has joined.
14:18:01 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds).
14:18:01 -!- Taneb has quit (Ping timeout: 240 seconds).
14:18:01 -!- tzxn3 has quit (Ping timeout: 240 seconds).
14:18:03 -!- Phantom___Hoover has joined.
14:18:13 -!- user2 has joined.
14:18:21 -!- kmc has quit (Ping timeout: 240 seconds).
14:18:45 -!- kmc has joined.
14:20:41 -!- tzxn4 has quit (Ping timeout: 240 seconds).
14:20:41 -!- Phantom__Hoover has quit (Ping timeout: 240 seconds).
14:20:41 -!- ais523 has quit (Ping timeout: 240 seconds).
14:20:58 -!- ais523 has joined.
14:22:06 -!- Sgeo has quit (Ping timeout: 240 seconds).
14:37:28 -!- Sgeo has joined.
15:44:19 -!- Sgeo has quit (Ping timeout: 276 seconds).
15:46:30 -!- Sgeo has joined.
16:12:09 -!- MoALTz has quit (Quit: brb).
16:15:52 -!- MoALTz has joined.
16:18:40 -!- sebbu has quit (Ping timeout: 272 seconds).
16:25:22 -!- Taneb has joined.
16:32:02 <ais523> heh, it seems that Pokémon's fallen guilty to the clbuttic situation, but targeted at themselves; it's impossible to trade an English-language Cofagrigus via the internet trade thing, unless you change its name from the default, because the default name trips the profanity filters
16:33:27 <ais523> it's not the only one, either :)
16:33:57 <Gregor> SMBC has been having a really good run recently: http://www.smbc-comics.com/?id=2525
16:35:12 <ais523> <Missingno. Master> A small update on the GTS fail swear filter situation. For some reason, someone on Serebii is having trouble trying to trade his Archeops, pretty much exactly how we have trouble trading Cofagrigus, Skuntank, and Marshtomp. I cannot for the life of me see the possible profanity in "Archeops", though.
16:35:19 <ais523> that's quite the profanity filter
16:36:25 -!- elliott_ has joined.
16:36:31 <Taneb> Maybe it's to do with the che?
16:37:22 <ais523> several people don't have internet access at weekends
16:37:31 <ais523> (people who access only from school or work)
16:38:47 <elliott_> * elliott :Nick/channel is temporarily unavailable
16:39:04 <elliott_> ais523: My estimate of the chances of that applying here are 0.
16:39:14 -!- elliott_ has changed nick to elliott.
16:39:16 <lambdabot> elliott: You have 2 new messages. '/msg lambdabot @messages' to read them.
16:39:36 <ais523> is one of them a reply from i?
16:39:47 <ais523> wait, it was oerjan who screwed that up, not you
16:40:30 <Gregor> @tell elliott_ Some years from now you will log in with an underscore in your nick, and before you ghost the original, you'll be reminded of this message I sent years prior. For no reason.
16:41:29 <ais523> elliott: he forgot the name in an @tell
16:41:40 -!- elliott has changed nick to elliott_.
16:41:42 <lambdabot> Gregor said 1m 11s ago: Some years from now you will log in with an underscore in your nick, and before you ghost the original, you'll be reminded of this message I sent years prior. For no reason.
16:41:44 -!- elliott_ has changed nick to elliott.
16:42:31 <elliott> <lambdabot> oerjan said 18h 16m 27s ago: i just looked at the [[wikipedia:complex number]] you added to Excela and it _does_ show identical to a local link. i hope that has changed in recent versions then.
16:42:47 <elliott> @tell oerjan It displays purple when visited in MW 1.18, unlike wikilinks.
16:43:04 <elliott> @tell oerjan And the same blue as external links when not.
16:43:58 <ais523> it should have some sort of externalness mark on it, really
16:44:32 <Gregor> Ideally a tiny Wikipedia logo.
16:45:29 <ais523> but that's copyrighted!
16:45:47 <ais523> (seriously, it is, and the WMF go and enforce it too; an interesting side effect of this is that you can't use the Wikipedia logo in a Wikipedia article)
16:45:59 <elliott> a tiny wikipedia logo with all the letters omitted
16:46:02 <ais523> (except occasionally due to fair use)
16:46:24 <elliott> Debian actually replaced Firefox's icon with a custom one really quickly
16:46:30 <elliott> it worked for Ubuntu for like one release :P
16:46:56 <Gregor> I was actually thinking of the stylized 'W', not the puzzle globe.
16:47:00 <ais523> I actually /like/ the name Iceweasel…
16:47:27 <elliott> Gregor: oh, that's the favicon, right?
16:47:53 <elliott> 04:28:43: <Gregor> Apparently a "type 40" QR code can store up to 2,953 characters. It's very dense. Assuming you could print, say, twelve of those at readable DPI per side of a sheet of paper, you've got about 69KB per sheet of paper. That means a floppy disk is 21 pages.
16:47:53 <elliott> 04:29:17: <Gregor> Sorry, /version/ 40.
16:47:57 <elliott> Gregor: Stop talking in the past now and read http://ollydbg.de/Paperbak/.
16:50:44 <elliott> Waaaaaaaaaaaaaah, now stfu and ignore the program parts
16:54:37 <Gregor> Doesn't download properly.
16:55:43 <Gregor> wtf, that was mind-bending.
16:55:51 <Gregor> Firefox goes all nutty trying to download it, wget had no problem.
16:56:23 <elliott> Gregor: wget uses paper-backed downloads.
16:56:28 <elliott> That's why it's more reliable.
16:58:39 -!- derdon has joined.
17:01:38 -!- fnaptagur has joined.
17:01:49 <HackEgo> fnaptagur: 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
17:02:48 -!- fnaptagur has left.
17:05:25 <ais523> yet another example of WKOE which has plagued #esoteric for a long time
17:05:33 <ais523> hmm, what does the WKOE expand to?
17:06:27 <elliott> ais523: wrong kind of esoteric
17:07:32 <Gregor> We have no evidence that it was a WKOE.
17:07:39 <Gregor> Could have just been an imbecile.
17:08:20 <elliott> Gregor: I have a sense for WKOEs.
17:09:33 <elliott> this just in: Java gets unsigned arithmetic
17:10:11 <ais523> elliott: not really, it's just adding more library functions for doing it
17:10:19 <ais523> major gain from this is unsigned < and >
17:10:29 <ais523> >>> is more arithmetic than < is
17:11:32 -!- MoALTz has quit (Ping timeout: 260 seconds).
17:13:47 <Phantom___Hoover> fizzie, if that was the case we'd have to jump ship immediately before Wolfram bought the place up.
17:16:20 <elliott> wait, the lament impersonator wasn't iament
17:17:10 <elliott> @tell oerjan "removing fowl language" -- you, removing the word "cocks" from [[Esolang:Sandbox]], 2007
17:17:32 <elliott> @tell oerjan I don't think anyone can ever forgive your puns.
17:19:11 <elliott> ais523: is it bad to edit the database to fix a typo in a log message?
17:19:33 <ais523> yes, that's ridiculous :)
17:22:02 -!- MoALTz has joined.
17:22:40 -!- Taneb has quit (Quit: stuff).
17:23:51 -!- Chef_ has joined.
17:25:54 <elliott> ais523: stop using commas like that
17:27:28 <elliott> "amusingly, to wrap around the OCaml REPL; I didn't realise that ledit itself was written in OCaml, but trying to determine its version number to answer your question, it became obvious, because nothing else has an argument parser that works quite like that"
17:27:30 -!- Taneb has joined.
17:27:39 <elliott> also, stop using BBCode on reddit; this is the second time you've done it without noticing it doesn't work
17:29:32 <ais523> it's its fault, it doesn't have a preview button
17:29:37 <ais523> I'm used to previewing before posting
17:29:42 <ais523> this means I tend not to preview afterwards
17:29:50 <ais523> also, Markdown is incredibly bad for use on fora, wikis, etc.
17:30:05 <ais523> it does seem reasonably good for its intended purpose, which is writing documentation
17:30:08 <ais523> but not for sandboxing HTML
17:30:10 <elliott> it has a "help" link right next to submit
17:30:12 <ais523> it's designed to be intermixed with HTML
17:30:19 <ais523> elliott: that doesn't show preview, though! that just explains markdown
17:30:31 <elliott> which is a sufficient explanation that BBCode /won't/ work
17:31:07 <fizzie> Did you know: "In order to program a computer, you must be able to think like the computer."
17:31:40 <ais523> bbcode fixed, at least
17:31:46 <ais523> elliott: the problem is not that I don't know that bbcode doesn't work
17:31:55 <ais523> the problem is that there's no real indication that I typed it, but carefully scanning the post for bbcode
17:31:59 <Taneb> Are we still having the Alan Dipert is asleep problem?
17:32:01 <ais523> or looking at it /after it's submitted/
17:32:11 <ais523> because there isn't a preview
17:33:11 <olsner> look at it after submitting, then do edit if it was wrong
17:34:26 <elliott> ais523: reddit also has an edit link
17:34:41 <Taneb> elliott, the whois info gives a phone number. Is that an option?
17:34:52 <ais523> elliott: right, I just edited the BBcode out
17:35:13 <ais523> Taneb: it's rare for that to go to the person who owns the domain, normally it goes to a domain registrar switchboard
17:35:39 <elliott> ais523: http://www.reddit.com/r/programming/comments/p31g9/the_bintrue_command_and_copyright/c3m7ndz :P
17:35:43 <elliott> also, I'm not going to phone someone
17:36:12 <Taneb> He hasn't tweeted in 4 days
17:36:37 <ais523> elliott: that was ages ago! and nobody else commented on the bbcode
17:36:42 <ais523> thus, I'll assume that people are just parsing it mentally
17:36:43 <elliott> Taneb: he seems to tweet only infrequently
17:36:54 <elliott> (I tried to use that to figure out if he'd be awake or not when I sent it, heh)
17:36:58 <ais523> you can tell I use webforums much more than I use reddit, right?
17:37:01 <elliott> ais523: But I saw it and went ":(".
17:37:06 <elliott> ais523: You don't want me to go ":(".
17:37:08 <ais523> oh, and bbcode is much better than markdown
17:37:30 <elliott> ah, somebody put ais523 on repeat again
17:37:31 <Taneb> He normally has a few tweets every couple of days
17:38:18 <fizzie> Taneb: Maybe he's using an offline Tweeter and connects every couple of days? (Has anyone made one? I's sure *someone* has.)
17:38:57 <ais523> elliott: did you see the comment where I found a contradiction in the WTFPL?
17:39:14 <ais523> http://www.reddit.com/r/programming/comments/pql5k/simplified_tetris_in_less_than_140_bytes_of/c3rmuw9?context=3
17:42:26 <elliott> ais523: it appears to be a mistake; contact Sam Hocevar, maybe he'll publish WTFPLv3
17:42:38 <elliott> and we can have flamewars over which version is better
17:43:32 <ais523> personally, I think the WTFPL is a very bad license
17:43:37 -!- MoALTz has quit (Ping timeout: 260 seconds).
17:43:40 <ais523> especially because there's no warranty disclaimer
17:44:15 <ais523> in fact, it's possible it has no legal force at all, and someone who licenses work over the WTFPL can subsequently sue for copyright infringement if someone else distributes it
17:44:19 <elliott> Why is there no “no warranty” clause?
17:44:19 <elliott> The WTFPL is an all-purpose license and does not cover only computer programs; it can be used for artwork, documentation and so on. As such, it only covers copying, distribution and modification. If you want to add a no warranty clause for a program, you may use the following wording in your source code:
17:44:19 <elliott> * the extent permitted by applicable law. You can redistribute it
17:44:21 <elliott> * and/or modify it under the terms of the Do What The Fuck You Want
17:44:23 <elliott> * To Public License, Version 2, as published by Sam Hocevar. See
17:44:25 <elliott> * http://sam.zoy.org/wtfpl/COPYING for more details. */
17:44:42 <ais523> (I say this due to the lawsuits that upheld the validity of the GPL and Artistic License; they were both on bases that don't apply to the WTFPL)
17:44:56 <elliott> ais523: the FSF's lawyers accept WTFPL
17:45:03 <elliott> and the OSI have implied that they consider it effective
17:45:14 <ais523> I guess estoppel might work
17:45:16 -!- MoALTz has joined.
17:45:28 <elliott> "Title: WTFPL Submission: http://crynwr.com/cgi-bin/ezmlm-cgi?17:mss:634:200902:aglgcgbhmfcheffmdgon License: http://sam.zoy.org/wtfpl/ Comments: It's no different from dedication to the public domain. Author has submitted license approval request -- author is free to make public domain dedication. Although he agrees with the recommendation, Mr. Michlmayr notes that public domain doesn't exist in Europe. Recommend: Reject"
17:45:52 <ais523> (you have to be careful about such things; the GPL was broken in Germany for a couple of years until the government specifically passed an amendment to a law to fix it)
17:45:59 <elliott> (cite for FSF: http://lists.debian.org/debian-legal/2002/09/msg00032.html)
17:46:26 <elliott> ais523: anyway, that's a silly objection since it applies to even ubiquitous licenses like BSD3/MIt
17:46:36 <ais523> elliott: right, indeed
17:47:03 <ais523> but it's easy to envisage a different case that doesn't apply to the WTFPL due to lack of economic gain from the licensing
17:47:13 <elliott> Gregor: what's stupid about it?
17:47:24 <ais523> the Artistic License was upheld in court solely on the attribution requirement (the other parts may have been relevant too but weren't tested)
17:47:27 <Gregor> The no warranty clause bit.
17:47:37 <elliott> Why is there no “no warranty” clause?
17:47:37 <elliott> The WTFPL is an all-purpose license and does not cover only computer programs; it can be used for artwork, documentation and so on. As such, it only covers copying, distribution and modification. If you want to add a no warranty clause for a program, you may use the following wording in your source code:
17:47:39 <elliott> * the extent permitted by applicable law. You can redistribute it
17:47:41 <elliott> * and/or modify it under the terms of the Do What The Fuck You Want
17:47:43 <elliott> * To Public License, Version 2, as published by Sam Hocevar. See
17:47:45 <elliott> * http://sam.zoy.org/wtfpl/COPYING for more details. */
17:47:49 * elliott prepares to set up a keyboard macro for the inevitable third paste
17:47:53 <ais523> and the GPL on the basis that the author got an economic benefit out of the requirement for distributed modifications to be backcontributed
17:47:58 <Gregor> But stupid people will happily use it without the no-warranty clause.
17:48:08 <Gregor> And since only stupid people use it, that means that nobody will ever read that.
17:48:23 <elliott> The cutting your own body parts off bit.
17:48:26 <ais523> hmm, I wonder if you can take some WTFPL software and imply an indemnification clause into it, on the basis that it does anything
17:48:30 <elliott> [safety instruction manual gets quoted]
17:48:54 <ais523> if I can do anything with the code, can I take that as permission to have the original author responsible if anything goes wrong with the code?
17:49:04 <ais523> (an indemnification's a bit stronger than a warranty)
17:49:14 <Gregor> elliott: It would be stupid for the chainsaw manufacturer to sell it without the safety instructions. It would be stupid to publish code without a warranty disclaimer.
17:50:00 <elliott> Gregor: The WTFPL manufacturer sells it with the safety instructions, and does not publish code without a warranty disclaimer.
17:50:09 <elliott> That quote is from the WTFPL page.
17:50:29 <Gregor> elliott: I'm not claiming that the WTFPL author is being stupid, I'm claiming that people who use the WTFPL tend to be stupid. Since they only use it because "waaah I hate legalese"
17:50:44 <Gregor> But the people who use the WTFPL are also distributors, distributors of code.
17:50:56 <elliott> OK, so "WTFPL. So stupid." == "[Most] WTFPL [users]. So stupid."
17:51:28 <Gregor> But the reason why most WTFPL users are so stupid is because the license only exists so you can go "waaah I hate legalese"
17:51:45 <elliott> "In fact, both license types have unacceptable obnoxious clauses (such as reproducing a huge disclaimer that is written in all caps) that severely restrain our freedoms. The WTFPL can solve this problem.
17:51:45 <elliott> When analysing whether a license is free or not, you usually check that it allows free usage, modification and redistribution. Then you check that the additional restrictions do not impair fundamental freedoms. The WTFPL renders this task trivial: it allows everything and has no additional restrictions."
17:52:04 <elliott> Gregor: Wrong; WTFPLv2 exists to allow everything, at least.
17:52:10 <elliott> Dunno about the motivations of WTFPLv1 (by a different author).
17:52:53 <elliott> He does a lot of things. He was the leader of Debian for a while, I think.
17:53:38 <elliott> Right, yes: http://en.wikipedia.org/wiki/Debian#Project_leaders
17:55:30 <elliott> the original WTFPL is nicer
17:55:50 <fizzie> The King of Debian. That would be a nice title.
17:55:52 <elliott> do What The Fuck you want to Public License
17:55:52 <elliott> Copyright (C) 2000 Banlu Kemiyatorn (]d).
17:55:52 <elliott> 136 Nives 7 Jangwattana 14 Laksi Bangkok
17:55:53 <elliott> Everyone is permitted to copy and distribute verbatim copies
17:55:55 <elliott> of this license document, but changing it is not allowed.
17:55:59 <elliott> Ok, the purpose of this license is simple
17:56:03 <elliott> DO WHAT THE FUCK YOU WANT TO.
17:56:07 <elliott> fizzie: /kickban fizzie disrupting the sanctity of [[]]s
17:57:08 <elliott> hmm, CC0 is only like half the size of the GPL
17:59:52 <elliott> that was meant as a comment that CC0 is long
18:01:49 <elliott> ais523: oh, I have a better cite for the FSF: http://www.gnu.org/licenses/license-list.html#WTFPL
18:05:47 -!- sebbu2 has joined.
18:05:47 -!- sebbu2 has quit (Changing host).
18:05:48 -!- sebbu2 has joined.
18:21:12 -!- zzo38 has joined.
18:28:26 -!- zzo38 has quit (Remote host closed the connection).
18:28:40 -!- sebbu2 has changed nick to sebbu.
18:33:35 -!- Chef_ has quit (Remote host closed the connection).
18:41:18 -!- nys has joined.
18:41:54 -!- user2 has changed nick to tzxn3.
18:43:48 -!- hagb4rd has joined.
18:44:04 <nys> bootstrappables
18:44:22 <nys> i like that word
18:44:25 <nys> bootstrappables
18:44:37 <elliott> bootstrappable is an adjective
18:45:33 <nys> i like bootstrappable languages
18:47:01 -!- Taneb has quit (Ping timeout: 240 seconds).
18:50:04 <elliott> bootstrappability isn't a property of a language
18:50:23 <elliott> "can implement itself" works, but is a rather weak properly
18:50:31 <elliott> every TC language meets it, so almost every language people talk about meets it
18:50:36 <elliott> and tons of non-TC languages meet it too
18:51:03 <nys> i mean the way forth is mostly written in forth
18:51:30 <Gregor> So, not a similar property at all.
18:52:50 -!- MoALTz_ has joined.
18:55:38 -!- MoALTz has quit (Ping timeout: 244 seconds).
18:56:00 <elliott> nys: that's not really what bootstrapping is used to mean
18:57:16 <nys> a self-sustaining process that proceeds without external help
18:57:35 <elliott> Yes, which does not apply to a Forth written in a non-Forth language, even if most of the library is written in Forth.
18:57:44 <elliott> Also, that definition excludes everything.
18:57:52 <elliott> If you don't depend on an OS, you depend on the CPU's instruction set.
18:57:57 <elliott> Eventually you depend on the universe.
18:58:22 <nys> well, once you write the few basic words, you can write the rest without any external help
18:59:39 <elliott> That's just a minimal core. The same applies to e.g. the lambda calculus.
19:00:12 <Gregor> Having a minimal core allowing for broad abstraction is a nice property, but not related to bootstrappability.
19:01:19 <nys> okay, it has a minimal core as well as bootrappability
19:01:28 -!- MoALTz__ has joined.
19:01:46 <nys> what is your definition of bootstrappability?
19:01:59 <elliott> A bootstrapping compiler is one that can compile itself.
19:02:07 <elliott> So, a Forth compiler in Forth is bootstrapping.
19:02:20 <elliott> Or bootstrappable. Whatever.
19:02:38 <elliott> gcc, CMUCL, SBCL, GHC, etc. are all bootstrapping compilers.
19:02:42 <elliott> Most Forths too, but not all.
19:03:54 -!- MoALTz_ has quit (Ping timeout: 244 seconds).
19:04:02 <elliott> nys: A small core aids bootstrappability, of course.
19:04:03 -!- ais523 has quit (Remote host closed the connection).
19:04:06 <elliott> Since you have to implement less.
19:05:25 <Gregor> To define it in terms of the language instead of the compiler, a language in which a compiler or interpreter for the language can be implemented is a bootstrappable language.
19:06:06 <nys> <elliott> bootstrappability isn't a property of a language
19:06:27 <Gregor> <elliott> well <elliott> "can implement itself" works, but is a rather weak properly
19:06:31 <Gregor> I choose to use that definition :)
19:07:30 <elliott> Technically all languages are bootstrappable if you allow compilers.
19:07:44 <elliott> The identity function in language L is a bootstrapping compiler for L.
19:07:52 <elliott> So any language that can implement cat qualifies.
19:08:02 <elliott> (Okay, not "all languages", but all languages with basic IO support.)
19:08:18 <Sgeo> My first thought was Gravity :/
19:09:13 <elliott> I kind of want to get esolangs.org transferred to me with this latency >_>
19:10:19 <fizzie> elliott: It expires in 24-May-2013, maybe you can claim it then if Whoever Whoeveritwas doesn't wake up before that.
19:11:29 -!- Taneb has joined.
19:12:20 -!- Taneb has quit (Remote host closed the connection).
19:13:51 <Sgeo> What happened?
19:18:30 <elliott> Sgeo: Nothing, that's the problem.
19:18:47 <elliott> fizzie: I rather expect renewal or squatting will take it before I could.
19:19:47 <fizzie> Well, maybe, but presumably e'd have to be awake to renew it.
19:20:04 <fizzie> Anyway it's winter, all kinds of critters are hibernating.
19:20:05 <elliott> fizzie: You last registered a domain in 1995, I take it?
19:20:08 <elliott> Auto-renewal is standard now.
19:21:37 <fizzie> Blah-dard; none of my (two is plural!) domains "auto-renew". I mean, how'd that even work, automatically charge a credit card, or send an invoice, or something?
19:22:08 <elliott> Yes, of course it automatically charges your card... which the registrar already has the details of...
19:22:22 <elliott> I would say manual renewal is most definitely the exception this century. I will go to... Gregor to back this up.
19:23:01 <Gregor> All my dozens of domain names autorenew.
19:23:58 <elliott> Sponsoring Registrar:GoDaddy.com, Inc. (R91-LROR)
19:24:09 <elliott> Oh, no. I really must get esolangs.org transferred to me.
19:25:28 <elliott> http://s3.buysellads.com/1242768/35493-1283434214.jpg
19:25:28 <fizzie> I don't think my .org place has my credit card info in file; definitely not the government-run .fi place, since I've paid those with the national (net-)bank transfer thing-thing.
19:25:42 <elliott> This is the funniest ad ever.
19:25:48 <olsner> I get an invoice for my domain
19:31:41 <elliott> Oh, hmm, apparently they resell to a bajillion companies.
19:32:22 <fizzie> It's that register4less place. I haven't really bothered to change.
19:35:04 <fizzie> I should put in some IPv6 glue for selene.gehennom.org, and/or just not have it be inside the domain like that, since it's got a proper name in zem.fi anyway.
19:35:49 <elliott> That sentence makes absolutely no sense to me.
19:36:03 <elliott> (Also if I got esolangs.org transferred to me I could make it IPv6-ready!(TM)(C)(R))
19:38:41 <elliott> "To avoid dealing with the overhead of boxed values and to allow reuse of the built-in arithmetic operators, the unsigned API support does not introduce new types like UnsignedInt with instance methods to perform addition, subtraction, etc. However, that lack of separate Java-level unsigned types does mean a programmer can accidentally improperly mix signed and unsigned values. However, new unsigned types aren't the only way to mitigate this haza
19:38:41 <elliott> rd. For example, a naming convention of adding a trailing "U" or "_U" to variables holding unsigned values could be adopted."
19:38:57 <fizzie> When you have a nameserver for X.Y inside it (i.e. foo.X.Y), it needs "glue" IPv4/IPv6 address records at the nameserves of Y; otherwise Y's nameservers will just tell everyone that "to find out about X.Y, ask foo.X.Y", which isn't quite enough. (With the glue, it's "..., ask foo.X.Y at 1.2.3.4 or 1:2:3:4:5:6:7:8".)
19:39:01 <elliott> Oracle are actually, honest-to-god proposing Hungarian notation as a bandaid for their inability to overload operators
19:39:10 <elliott> (Making defining a new type infeasible.)
19:39:37 <elliott> fizzie: You run your own nameservers?
19:40:05 * elliott wonders what GABRIELI.VPS.KOTISIVUT.COM is.
19:40:06 <fizzie> That's the way Things Were Done.
19:40:23 <fizzie> kotisivut.com is a Finnish hosting/VPS place.
19:40:30 <fizzie> It's literally "homepages.com".
19:40:51 <fizzie> (It's a friend's box.)
19:41:14 <fizzie> I... see there's a webserver there.
19:41:58 <elliott> fizzie: Things Were Done as in common practice at one time?
19:42:20 <elliott> Also, tell your friend their website is shitty.
19:43:24 <fizzie> Well, I don't know about common; at least it used to be the case that domain registration cost less $s than domain registration + DNS services; and free primary-DNS places were awfully awful.
19:43:52 <fizzie> There was this one granitecanyon thing.
19:43:59 <fizzie> I suppose it's dead now.
19:44:14 <elliott> Gregor mentioned them I think.
19:44:48 <elliott> GraniteCanyon.com - Google Public DNS
19:44:49 <elliott> whois.domaintools.com/granitecanyon.com
19:44:52 <elliott> --Google search for granitecanyon
19:44:58 <elliott> "Google Public DNS" appears nowhere on the page.
19:45:04 <elliott> The Google Public DNS site is also the third result.
19:45:05 <Gregor> granitecanyon is dead, yes.
19:45:17 <Gregor> But also most registrars throw in a free DNS server now.
19:45:18 <elliott> fizzie: Dead since at least 2001, it seems.
19:45:26 <elliott> But yeah, I've never seen a registrar not offer DNS.
19:45:31 <elliott> Also VPS companies tend to.
19:45:35 <elliott> (Slicehost and Linode both do.)
19:47:21 <fizzie> Anyway, I didn't have a VPS account back then, I just had a computer.
19:48:17 <fizzie> I see gehennom.org was created in Aug 2002; I think I had just moved to the place where I had a beehive in my curtains.
19:48:23 <fizzie> (It's the most memorable thing about that place.)
19:48:54 <fizzie> Yeah. I kept the windows open the whole summer, it was pretty warm there. Then when I was taking the curtains down when moving out, a beehive dropped down.
19:49:20 <fizzie> I spent a couple of hours elsewhere to wait for them to abandon it, as I recall.
19:49:38 <fizzie> It wasn't a *big* one, but still.
19:49:54 <kallisti> I need a non-#esoteric person to answer my questions
19:50:01 <fizzie> Actually I mean a wasp nest.
19:50:06 <fizzie> Sorry for the confusion there.
19:50:26 <olsner> #esoteric, your premier source for non-#esoteric people
19:50:34 <fizzie> They all fly and buzz and so on, anyway.
19:51:03 <kallisti> Nisstyre: why on earth do you frequent #python?
19:51:23 <fizzie> Like a MODERN-DAY JESUS, e is.
19:51:36 <kallisti> Nisstyre: are there no intelligent #python regulars that can do that already?
19:51:45 <Nisstyre> kallisti: they're in short supply
19:52:00 <elliott> fizzie: OK, a wasp nest is less impressive than a beehive.
19:52:04 <Nisstyre> kallisti: tbh I am on a lot of channels on freenode
19:52:11 <Nisstyre> and I do not chat on a lot of them much
19:52:22 <kallisti> asking #python about TCO is actually not a good way to troll. because, they apparently don't know what it is or why it's not in python.
19:52:32 <fizzie> elliott: Yes, I think so too. Still, it was memorable.
19:53:06 -!- monqy has joined.
19:53:31 <Nisstyre> kallisti: I was able to implement TCO on my calculator, since it lets you manipulate the stack in the language
19:53:54 <kallisti> Nisstyre: ah. so you manually performed the TCO I guess?
19:53:59 <kallisti> or was there a way to automate it?
19:54:03 <Nisstyre> with some functions like fold and unfold
19:54:11 <Nisstyre> and then I use those functions
19:54:55 -!- oerjan has joined.
19:55:39 <elliott> Nisstyre: If RPL has loops, you could just implement foldl as one.
19:55:47 <Nisstyre> elliott: I eventually did do that
19:55:50 <elliott> foldr too, though it'd be trickier.
19:55:53 <Nisstyre> the first way was just for fun
19:56:03 <lambdabot> elliott said 3h 13m 15s ago: It displays purple when visited in MW 1.18, unlike wikilinks.
19:56:03 <lambdabot> elliott said 3h 12m 57s ago: And the same blue as external links when not.
19:56:04 <lambdabot> elliott said 2h 38m 51s ago: "removing fowl language" -- you, removing the word "cocks" from [[Esolang:Sandbox]], 2007
19:56:04 <lambdabot> elliott said 2h 38m 29s ago: I don't think anyone can ever forgive your puns.
19:56:18 <Nisstyre> it has for loops, and do until statements
19:58:06 <oerjan> elliott: i punned already back then? ok.
19:58:43 <fizzie> It hasn't actually be... what, it's been *five years* since 2007?
19:59:21 <fizzie> That's half a decagon.
19:59:26 <elliott> oerjan: It was in December IIRC.
19:59:27 <oerjan> it's eerie how much faster my laptop got when ditching avg.
19:59:31 <elliott> So, quite many months since I first came to #esoteric.
19:59:45 <elliott> oerjan: Now ditch IE and marvel at how even faster it gets.
20:00:06 <elliott> fizzie: The wiki being 7 years surprised me greatly.
20:01:08 <oerjan> (technically i also ditched spybot, although i'm not sure if that was even running, since i never noticed it.)
20:01:35 <elliott> I believe Spybot is a batch tool.
20:01:38 * kallisti googles "functional programming python" and seethes with rage.
20:01:51 <elliott> Anyway, upwards of 90% of Windows "security" software is junk.
20:02:19 <kallisti> does oerjan secretly have the computer savvy of my parents?
20:02:24 <oerjan> avg was nice in the beginning, but they had to add more and more things.
20:02:31 <fizzie> A "Spybot" sounds like a bad thing, not like a good thing.
20:02:49 <elliott> oerjan: Well, "junk" as in "useless at best".
20:02:52 <kallisti> fizzie: dunno, they've worked well for my purposes.
20:03:00 <oerjan> fizzie: "Spybot Search & Destroy", i have been assuming it was anti-spy software.
20:03:02 <elliott> I'm not convinced the 9% is worth using, either, unless you're using some ancient unpatched version of Windows or the like.
20:03:13 <oerjan> i think my cousin may have added it.
20:03:14 <elliott> oerjan: It's anti-malware. Why do you install software you don't know what it does...?
20:03:25 <fizzie> oerjan: Oh, the name continues like that.
20:03:29 <elliott> Well, anti-(spyware|malware).
20:03:38 <elliott> Well, anti-(spyware|adware).
20:03:50 <olsner> oerjan: how did your cousin manage to install crap on your computer?
20:03:55 <fizzie> anti-(spy|ad)ware LET'S OPTIMIZE THOSE REGEX.
20:04:12 <elliott> "SpySheriff, also known as Brave Sentry, Pest Trap, SpyTrooper,[1] and SpywareNo,[2] is malware that disguises itself as an anti-spyware program." -- these ones are the best.
20:04:22 <elliott> MALWARE IS SUPER ILLEGAL AND MAY BE ON YOUR COMPUTER NOW!!!
20:04:30 <oerjan> <kallisti> does oerjan secretly have the computer savvy of my parents? <-- quite possibly, i tend not to learn the parts of computer maintenance which don't interest me.
20:04:32 <elliott> REMEMBER: ALWAYS STAY SAFE ON THE INTERNET BY NOT DOWNLOADING TRUSTED SOFTWARE
20:04:47 <kallisti> oerjan: I'm the same way actually, with linux.
20:04:57 * elliott wonders why oerjan is even running Windows.
20:05:01 <kallisti> the Windows stuff I just picked up as a kid.
20:05:02 <olsner> elliott: is this from the site of an anti-spyware tool?
20:05:27 <Nisstyre> kallisti: it's possible to do some very functional stuff with python
20:05:33 <Nisstyre> but you have to implement some things yourself
20:05:55 <oerjan> elliott: it came with the computer and i'm too lazy too change, and to scared of breaking it completely in the process.
20:06:19 <kallisti> Nisstyre: wouldn't the pythonic way be to use generators instead?
20:06:28 <elliott> oerjan: I would give anyone who managed to break a computer to the point where sticking in a Windows CD didn't work to reinstall Windows a medal.
20:06:44 <olsner> a pythonic way to do functional programming? derp
20:06:47 <elliott> oerjan: *break a computer to the point [...] by trying to install an OS a medal.
20:07:04 <Nisstyre> kallisti: http://codepad.org/4irStQot
20:07:33 <kallisti> Nisstyre: yes I'm aware that python has itertools and higher-order functions.
20:07:59 -!- Chef_ has joined.
20:08:00 <Nisstyre> kallisti: that would be better written with linked lists though, is my point
20:08:15 <Nisstyre> and not generators or iterators
20:08:28 <Nisstyre> of course it wouldn't be tail recursive
20:08:37 -!- oerjan has quit (Quit: Lost terminal).
20:08:50 <fizzie> elliott: He tried, and now it's broken.
20:09:24 <elliott> @tell oerjan HERE'S YOUR MEDAL!
20:09:25 -!- oerjan has joined.
20:09:35 <fizzie> Oh, maybe I "jumped the gun" there.
20:09:36 <monqy> speak of the devil
20:10:11 <oerjan> *sigh* that network error seems to have returned in recent days
20:10:11 <lambdabot> oerjan: You have 1 new message. '/msg lambdabot @messages' to read it.
20:10:16 <lambdabot> elliott said 52s ago: HERE'S YOUR MEDAL!
20:11:08 <oerjan> darn now my laptop wants a reboot
20:11:13 -!- oerjan has quit (Client Quit).
20:11:24 <elliott> "oerjan... reboot me... PLEASE."
20:11:25 <monqy> laptops are people too
20:12:08 <Nisstyre> elliott: actually a program like that might not be a bad idea
20:12:17 <Nisstyre> it could also tell you when it's low on memory, etc...
20:12:28 <monqy> oerjan... i'm low on memory... PLEASE
20:12:45 <elliott> Nisstyre: None of those things really demand a reboot.
20:12:53 <Nisstyre> elliott: no, I mean in general
20:12:59 <Nisstyre> in addition to telling you it needs a reboot
20:14:22 <kallisti> when doing TCO in the case of a tail recursive call, is it more common to simply remove the stack frame and replace it with the tail call's frame, or do some implementations actually translate the code into a loop?
20:14:26 <elliott> When does a computer ever need a reboot; things like Windows installers arleady tell you you need to reboot, and apart from that it's just kernel upgrades and the like :P
20:14:57 <elliott> kallisti: Tail recursion optimisation usually turns the call into assignments of arguments and a goto.
20:15:10 <elliott> kallisti: More general tail-call optimisation usually just replaces the stack frame.
20:15:16 -!- Chef__ has joined.
20:15:34 <kallisti> elliott: do implementations attempt to spot mutual recursion and such?
20:16:34 <olsner> you should get that for free if you do tail-call optimisation well enough
20:16:40 <olsner> at least if you mutually recurse by tail calls
20:16:47 <elliott> kallisti: TCO implies mutual recursion optimisation...
20:17:06 <elliott> That's what it mostly buys you over TRO.
20:17:18 <kallisti> I mean does it attempt to translate it to assignment in jumps in the same way as a tail recursive call
20:17:21 <elliott> (Apart from eliminating a bunch of additional constant space factors)
20:17:22 <kallisti> I assumed the normal TCO applied.
20:17:35 <elliott> kallisti: If you're doing general TCO optimisation, you don't need to do TRE.
20:17:46 <kallisti> would it not be faster in that case?
20:17:59 <kallisti> I assumed replacing stack frames was expensive.
20:18:03 <kallisti> or more so than assignments and gotos
20:18:06 -!- zzo38 has joined.
20:18:17 -!- Chef_ has quit (Ping timeout: 260 seconds).
20:18:28 <elliott> You assume a lot of things.
20:18:59 <kallisti> so deleting a stack frame and then replacing it with a stack frame of the same size is not a waste of time?
20:19:07 <elliott> I don't think you know how the stack works.
20:19:32 -!- oerjan has joined.
20:19:33 <monqy> what do you mean deleting, kallisti
20:19:35 -!- MoALTz has joined.
20:19:52 <kallisti> so it just overwrites the frame..
20:20:15 <elliott> The stack is a contiguous block of memory.
20:20:19 <elliott> It's whatever the stack pointer points to.
20:20:34 <elliott> The pointer gets incremented and decremented as stuff gets pushed and popped of it.
20:21:24 -!- MoALTz__ has quit (Ping timeout: 244 seconds).
20:22:45 <Nisstyre> kallisti: you might like this: http://www.amazon.com/review/R18C8QYFR1NU6J/ref=cm_aya_cmt?ie=UTF8&ASIN=0738202967#wasThisHelpful
20:22:48 <oerjan> <Gregor> LAME <-- you need to do it in private, duh
20:23:04 <elliott> oerjan: What a wonderfully context-free quote.
20:23:11 <oerjan> elliott: isolated mountain cottage, i said!
20:23:24 <elliott> Nisstyre: Without the #wasThisHelpful, one presumes.
20:24:28 <elliott> oerjan: <oerjan> <Gregor> LAME <-- you need to do it in private, duh <elliott> oerjan: What a wonderfully context-free quote.
20:24:32 <elliott> It wasn't in response to that?
20:24:45 <monqy> is this real guido
20:24:47 <oerjan> although feel free to pretend it was.
20:25:07 <elliott> It doesn't seem that unreasonable a review to me.
20:25:20 <elliott> Guido has said many stupider things.
20:25:24 <olsner> I'm a bit surprised he recognizes turing machines and basic computer science though
20:25:28 <Nisstyre> elliott: he basically gave it a bad review because he didn't understand what it was about
20:25:52 <elliott> Well, that's appropriate, no?
20:26:14 <elliott> You're meant to say what you think of a book in a review; he didn't find it that interesting and couldn't follow the last part.
20:26:21 <Nisstyre> well, no, not really, especially when it's supposed to be about QM, and you aren't educated in QM
20:26:36 <zzo38> Some people said computer science is not science and has nothing to do with computers; saying computer science is about computers is like saying astronomy is about telescopes.
20:27:02 <monqy> http://www.amazon.com/review/R2AVZ1CXQ025A9/ref=cm_aya_cmt?ie=UTF8&ASIN=B0014JLC7G guido reviews Spongebob 1.3MP Digital Camera
20:27:11 <kallisti> I'm not sure why a return address needs to be stored on a stack frame. I guess because the calling code is not always the frame directly above?
20:27:13 <Nisstyre> olsner: why would you be surprised that he knows about basic CS?
20:27:19 <elliott> Nisstyre: Okay, but the page for the book itself and the use of the term "last part" to refer to the quantum material implies to me that only part of the book is about QM.
20:27:21 <kallisti> ah, I guess when you add multiple processes to the mix...
20:27:26 * kallisti was thinking in the single-process case.
20:27:37 <elliott> zzo38: for values of some people equal to Dijkstra
20:27:52 <elliott> kallisti: If you don't store the return address on the stack, how do you return?
20:28:02 <olsner> Nisstyre: well, I've always assumed he didn't
20:28:26 <Nisstyre> olsner: he would have to in order to have implemented such a successful language
20:28:37 <kallisti> elliott: well would it be possible, in the case where there's only one thread of execution, to just increment the frame pointer by sizeof(stackframe)
20:28:42 <Nisstyre> (plus I'm pretty sure he has a degree in CS)
20:29:15 <Nisstyre> he does say a lot of stupid stuff though
20:29:17 <elliott> kallisti: how does that pass control back to the code that called you?
20:29:34 <oerjan> kallisti: note that afaik a frame doesn't contain the code or ip for its _own_ code
20:29:50 -!- Chef_ has joined.
20:30:09 <olsner> kallisti: if you have multiple threads, each thread has its own stack
20:30:24 <olsner> switching stacks (and some other state) is how multiprocessing is implemented
20:30:46 <elliott> Not multiprocessing, multithreading, no?
20:30:54 <elliott> Multiprocessing = multiple CPUs, to my understanding.
20:31:11 <olsner> right, I mean multitasking or something
20:31:33 <kallisti> so a multi-core system does each CPU have a seperate block of memory for stack?
20:32:14 <oerjan> kallisti: mind you iirc the Icon programming language has a stack regime so complicated that what you said does apply to it.
20:32:51 <olsner> kallisti: each cpu has its stack pointer pointing to the stack of whatever thread is running on it
20:33:07 <elliott> kallisti: Each thread has its own block of memory to use as a stack.
20:33:15 <elliott> oerjan: cactus/spaghetti stack, right?
20:33:25 -!- Chef__ has quit (Ping timeout: 245 seconds).
20:33:26 <elliott> But that's linked-list style, so it's not really comparable at all.
20:33:43 <kallisti> that makes way more sense that physically swapping out a large block of memory into specially designated "stack" block.
20:33:59 <olsner> indeed, that's partially why it's done that way
20:34:45 <oerjan> elliott: i'm not sure what it was called but i think it was an actual stack in order, not heap frames
20:35:15 <oerjan> it's just that frames sometimes linger even if you return beyond them
20:35:44 <oerjan> and it allows for some interesting control despite not needing heap
20:35:46 <elliott> kallisti: As I said, "it's whatever the stack poniter points to".
20:36:17 <elliott> oerjan: icon is awesome because you can write x == (y | z) :P
20:36:19 <kallisti> does the CPU have any information about the size of the stack frame?
20:36:53 <kallisti> in, frames, or, whatever unit you want.
20:37:19 <olsner> IIRC, Ada has two stacks
20:37:21 <kallisti> as in, the maximum size of the stack.
20:37:27 <elliott> kallisti: Why would it have a maximum?
20:37:59 <olsner> well, it does have a maximum limited by the amount of virtual address space
20:38:03 <kallisti> as far as I know, it's a thing in many languages to specify a max stack limit. I'm basically wonder how this is implemented.
20:38:25 <olsner> or, it wraps after that... that could be fine if you want an infinite circular stack
20:38:26 <elliott> If you mean Python: Manually.
20:38:37 <elliott> If you mean pthreads: you can control how much memory the OS allocates for you to use as stack.
20:38:48 <elliott> If you mean generally: question is too vague to answer and has incorrect underlying assumptions.
20:39:07 <Nisstyre> my calculator's stack is only limited by the amount of memory available
20:39:24 <elliott> Hmm, what do you call a 2D torus? :P
20:39:44 <oerjan> elliott: er, what does that mean?
20:39:54 <oerjan> a torus is 2d, in a sense
20:40:16 <elliott> oerjan: A wrapping 2D field is a torus.
20:40:19 <elliott> A wrapping 1D field... oh, right.
20:40:30 <elliott> Strange things, these "circles".
20:40:46 <oerjan> the nD torus is simply the topological product of n circles.
20:41:15 <elliott> What's the topological product identity?
20:41:40 <oerjan> to the point that i've seen blackboard bold T used to denote a circle
20:41:57 <oerjan> elliott: single point set
20:42:34 <elliott> Even if I actually meant 2D torus.
20:43:27 <kallisti> elliott: is my underlying assumption that there is a general way to specify stack size?
20:43:44 <oerjan> so that $\mathbb{T}^n$ denotes an n torus, even though $S^n$ customarily denotes an n-sphere.
20:43:56 <elliott> kallisti: One of your underlying assumptions is the word "size".
20:44:03 <elliott> Perhaps you assume the CPU allocates memory.
20:44:47 <kallisti> I'm wondering how a "stack overflow exception" occurs
20:45:00 <elliott> Most languages check it theirselves.
20:45:01 <kallisti> if the CPU has no knowledge of the bounds of the memory its accessing.
20:45:17 <kallisti> ah, okay. What about, say, C? is that just a normal memory access exception of some kind?
20:45:33 <elliott> oerjan: Maybe they're thinking of \mathbb{T} as a function with weird syntax.
20:46:07 <kallisti> elliott: so the stack is allocated then...?
20:46:35 <elliott> kallisti: Try again when you have a coherent question.
20:46:45 <oerjan> elliott: no, i mean, $X^n$ may usually denote the product of n $X$ copies, but for S it means something different.
20:47:03 <elliott> oerjan: It's S that's the function with weird syntax, then.
20:47:16 <kallisti> elliott: in order for the segfault to occur, it has to be trying to access memory that the program isn't allowed to access, right?
20:47:32 <kallisti> or other trying to access something like NULL, for example.
20:48:00 <kallisti> doesn't that imply some sort of allocation of the memory?
20:48:06 <elliott> kallisti: Yes, CPUs have memory protection.
20:51:17 <quintopia> C programs can protect parts of their own memory too. i thought that stacks typically start at the end of a program's memory and grow downward toward the heap though?
20:51:30 <kallisti> so what's incorrect about the assumption that the CPU allocates memory?
20:51:39 <elliott> quintopia: Yeah, stacks grow downwards on most CPUs.
20:51:47 <elliott> kallisti: The incorrect part is that CPUs don't allocate memory.
20:52:28 <kallisti> what else could possibly be allocating memory....
20:52:39 <elliott> kallisti: Allow me to introduce you to the "operating system".
20:52:42 <zzo38> Does the C computer allocate memory?
20:52:53 <olsner> the allocation is done by the "Dynamic" in DRAM
20:53:05 <kallisti> elliott: my computer doesn't have one of those. I think I forgot to order the part.
20:53:19 <olsner> before that we used SRAM where everything was preallocated in the factory
20:53:44 <fizzie> olsner: So what's this SDRAM then?
20:53:59 <zzo38> NES/Famicom does not have enough RAM for dynamic allocation; everything is allocated at compile-time
20:54:10 <kallisti> elliott: but yes I understand what you meant now.
20:54:24 <kallisti> by "CPU allocates memory" you meant something like there's a specific memory allocation instruction, or something.
20:54:33 <olsner> fizzie: synchronized, it's for multithreading
20:54:40 <elliott> Yes, which would be required for it to know about RAM.
20:54:44 <elliott> Yes, which would be required for it to know about sizes of things.
20:54:48 <elliott> Or else the OS would have to tell it.
20:54:52 <elliott> `addquote <olsner> the allocation is done by the "Dynamic" in DRAM <olsner> before that we used SRAM where everything was preallocated in the factory <fizzie> olsner: So what's this SDRAM then? <olsner> fizzie: synchronized, it's for multithreading
20:54:55 <HackEgo> 809) <olsner> the allocation is done by the "Dynamic" in DRAM <olsner> before that we used SRAM where everything was preallocated in the factory <fizzie> olsner: So what's this SDRAM then? <olsner> fizzie: synchronized, it's for multithreading
20:57:17 <kallisti> to what degree is a typical processes virtual address space physically contiguous?
20:57:37 <fizzie> cat some /proc/pid/maps to find out?
20:58:01 <fizzie> Oh, "physically continuous".
20:58:21 <fizzie> Never mind, then. maps just lists the ranges, not where they end up.
20:58:27 -!- monqy has quit (Read error: Operation timed out).
20:58:48 -!- oerjan has set topic: TKOE | Breeding zebras as an information storage medium | http://codu.org/logs/_esoteric/.
20:59:01 <kallisti> ah. I'm not even sure why I'm curious
20:59:32 <oerjan> elliott: i think i shall leave that up to your imagination.
20:59:54 <olsner> there's also a /proc/pid/pagemap that has the physical map in some binary format
21:00:06 <oerjan> you _may_ have guessed the word i initially intended.
21:00:08 -!- monqy has joined.
21:00:57 <HackEgo> lycopel ipha substip lhware mstruri pel neve vieredereoplatio imporph montific mer gray pischon ciparti systo wibate bonessorlverha boee ypera prispea peti admont phocea nouver arthemato
21:01:02 <elliott> zzo38: I already said The :(
21:01:22 <elliott> Busty Testicles' Kind of Esoteric?
21:01:52 <kallisti> ....I'm wondering why the ' is there.
21:01:56 <elliott> Out of curiosity, do I have to sign up for GoDaddy to get a GoDaddy domain transferred to me if I plan to transfer it from GoDaddy?
21:02:06 <elliott> kallisti: Possssssszessive.
21:02:10 <kallisti> I think "Busty Testicles Kind of Esoteric" would make more sense
21:02:15 <kallisti> it reads like a news headline.
21:02:33 <oerjan> elliott: banach-tarski, duh
21:02:34 -!- monqy_ has joined.
21:03:36 <elliott> oerjan: SO WHAT'S THE T IN THE TOPIC
21:03:45 <elliott> kallisti: Oh, that works too.
21:04:43 <kallisti> at least, that's the main reason I'm here..
21:04:58 <elliott> kallisti: What, busty testicles?
21:05:16 <Gregor> <elliott> Out of curiosity, do I have to sign up for GoDaddy to get a GoDaddy domain transferred to me if I plan to transfer it from GoDaddy? // you shouldn't have to, and typically registrars make it about as difficult as legally possible to transfer domains to a different registrar.
21:05:32 <elliott> Gregor: Shouldn't have to what?
21:05:47 <elliott> And yeah, I know they do, but it's obviously not that hard given that mass migration away from GoDaddy circa SOPA.
21:05:51 -!- monqy has quit (Ping timeout: 244 seconds).
21:06:02 <elliott> Now I am going to sign a song for Alan Dipert.
21:06:09 <Gregor> elliott: You shouldn't have to sign up, no.
21:06:14 <elliott> Alan Dipert / Wake the fuck up / Fuck fuck fuck fuck / Wake the fuck up.
21:06:32 <elliott> Gregor: So someone could transfer a GoDaddy domain to someone else using Gandi?
21:06:32 <HackEgo> 509) <monqy> beautiful summer / fuck fuck fuck fuck fuck fuck fuck / fuck fuck fuck fuck fuck
21:06:34 <Gregor> You'll need the current owner to give you a transfer code.
21:06:43 -!- zzo38 has set topic: Iuckqlwviv Kjugobe was not here yet. | TKOE | Breading zebras as an information shortage medium | http://codu.org/logs/_esoteric/ | httP://64.62.173.65/%49%27/.%2E/lo%67s/_esoteri%63/#THIS_IS_NOT_A_SCAM.
21:06:59 <olsner> I think the registrar you transfer to takes care of transferring once you have the appropriate proof that you should be able to
21:07:03 <Gregor> Then GoDaddy has 30 days (IIRC) to sit on their ass and not transfer it while giving you the opportunity to confirm that the transfer was a fake, but not giving you the opportunity to confirm that it was legitimate.
21:07:34 <kallisti> are these topics supposed to be making sense?
21:07:36 <Gregor> PiRSquared17: It says it's not a scam right there!
21:07:41 <kallisti> perhaps I haven't divined enough esoteric knowledge.
21:08:03 <elliott> PiRSquared17: to look more Professional
21:08:25 -!- elliott has set topic: For direct log access , 10 year of experience. See httP://64.62.173.65/%49%27/.%2E/lo%67s/_esoteri%63/#THIS_IS_NOT_A_SCAM ..
21:08:51 <zzo38> Someone told me that GoDaddy turned it off one week before it is time for renew payment if it is not automatic
21:09:12 -!- monqy_ has changed nick to monqy.
21:10:00 <Gregor> But stalker mode is still read-only.
21:10:28 <kallisti> do you intend to make it read-write?
21:10:35 <olsner> I don't get this stalker mode thing, what is it?
21:10:45 <PiRSquared17> olsner: http://64.62.173.65/%49%27/.%2E/lo%67s/log/_esoteric/s
21:11:31 <Deewiant> https://64.62.173.65/logs/log/_esoteric/s looks unsafe!
21:11:39 <olsner> looks like it's displaying 20 random lines from the log
21:11:54 <olsner> not the latest, but always the same
21:12:31 <elliott> olsner: r u SURE????????????
21:12:37 <olsner> it's less than one page, no way to scroll down
21:13:07 <olsner> elliott: yes, is it broken in opera?
21:13:12 <elliott> Hey Gregor, an Opera user is having problems with your JS.
21:13:50 <olsner> reloading in the debugger actually put some more lines on there
21:13:59 <elliott> Gregor: Whoa, what's this require() thing that isn't in a library?
21:14:05 <olsner> Last line is "06:13:32: * Sgeo is reading the BOFH archives"
21:14:38 <Gregor> Never tested it in Opera, I guess all three Opera users will just have to live without stalker mode.
21:16:49 <elliott> How much does it cost to steal a domain from someone
21:17:10 <kallisti> elliott: you could negotiate an offer with them
21:18:29 -!- Chef_ has quit (Remote host closed the connection).
21:18:31 <elliott> kallisti: I'm not going to dignify that with a response.
21:19:21 <fizzie> I think the canonical form is "I'm not going to dignify that with a response other than 'I'm not going to dignify that with a response.'"
21:19:46 <elliott> But I didn't dignify them with that response.
21:19:48 <elliott> It's missing the second half.
21:19:50 <fizzie> Possibly with some sort of "other than #1" notation to make it recursive.
21:19:51 <monqy> but that's a response other than "I'm not going to dignify that with a response."
21:20:11 <fizzie> It's a thing I've seen said.
21:20:48 <monqy> and I've heard it written
21:21:00 <fizzie> #0=I'm not going to dignify that with a response other than '#0#'.
21:21:24 <fizzie> I think MzScheme used to show recursive things like that. Or at least somewhat like that.
21:23:04 <Sgeo> I want to learn a lisp.
21:23:11 <Sgeo> But I always get hung up on the choices.
21:23:48 <oerjan> > var . intercalate "I'm not going to dignify that with a response other than " $ cycle "'\""
21:23:49 <lambdabot> Couldn't match expected type `[GHC.Types.Char]'
21:24:24 <oerjan> > var . intercalate "I'm not going to dignify that with a response other than " $ cycle ["'","\""]
21:24:26 <lambdabot> 'I'm not going to dignify that with a response other than "I'm not going to...
21:26:27 <fizzie> I think it would be nicer if it had the matching-quote ". at the hypothetical end.
21:30:50 <kallisti> wow I leave for several minutes
21:30:54 <kallisti> and... we're still talking about that.
21:31:05 -!- Chef_ has joined.
21:33:16 <elliott> oerjan: can you add the end quotes thanx
21:33:33 <oerjan> but but i'd just decided not to bother D:
21:35:40 <fizzie> > let resp = "I'm not ... other than 'I'm not ... other than \""++resp++".\".'" in resp
21:35:41 <lambdabot> "I'm not ... other than 'I'm not ... other than \"I'm not ... other than 'I...
21:36:35 <elliott> oerjan: wtf, why is it not valid to use a (forall a. (C a) => a) as a b given C b?
21:36:57 <fizzie> (You could argue about the .; as a non-native speaker, I fortunately don't have to.)
21:37:36 <elliott> must be some other problem here
21:40:10 <elliott> oerjan: fizzie: Relevant: http://sprunge.us/CBjS
21:40:22 <elliott> Now you can *check* that your strings include the required endings!
21:41:02 <elliott> Oh, Reverse is just Dual String.
21:41:53 <elliott> *Main> putStrLn $ summarise resp
21:41:53 <elliott> I'm not ... other than 'I...'.".'.".'.".'.".'.".'.".'
21:42:00 <elliott> As we can see, fizzie's is good but omits the final .
21:42:15 <elliott> *Main> putStrLn $ summarise resp
21:42:15 <elliott> I'm not ... other than 'I[...]'.".'.".'.".'.".'.".'.".'
21:42:17 <elliott> That's a bit more readamable.
21:44:09 <elliott> oerjan: Come on, that's a cute hack. :(
21:45:30 -!- MoALTz_ has joined.
21:45:35 <fizzie> elliott: Yes, I was going to end it in ... in resp++"." but forgotteded.
21:47:19 <lambdabot> <no location info>: parse error on input `:>'
21:47:27 <lambdabot> Not in scope: data constructor `:>'
21:48:59 -!- MoALTz has quit (Ping timeout: 276 seconds).
21:53:27 <HackEgo> "Banach-Tarski" is an anagram of "Banach-Tarski Banach-Tarski".
21:54:37 <oerjan> quintopia: what's yellow and equivalent to the axiom of choice?
21:55:42 <quintopia> a good orange should not be yellow
21:55:51 <quintopia> a good orange is sweet and delicious
21:56:06 <fizzie> Much like a good yellow shouldn't be orange.
21:56:17 <oerjan> the answer, of course, is zorn's lemon.
21:56:18 -!- MoALTz__ has joined.
21:56:31 <quintopia> fizzie, did you have any other clever self-inverses?
21:56:51 <fizzie> quintopia: oerjan constructed an infinite amount of them.
21:57:09 <quintopia> an infinite number of families of self-inverses?
21:57:39 <quintopia> which of those families have a finite number of discontinuities? those interest me the most
21:57:46 <kallisti> a self-inverted banach-tarski of families of self-inverses. (what)
21:57:49 <fizzie> quintopia: f(x) = (2 - x^(1/3))^3 is a continuous one.
21:58:33 <olsner> Gregor: for some reason you only get one event from that xmlhttprequest that downloads the logs
21:58:45 <olsner> (you're probably doing something wrong, because Opera is perfect :P)
21:59:06 <elliott> olsner: Go stare at the Acid 3 test some more if you're losing faith.
21:59:18 <quintopia> fizzie: aha.to generalize that to a whole lot of continuous ones
21:59:21 -!- MoALTz_ has quit (Ping timeout: 240 seconds).
22:00:02 <fizzie> quintopia: And for any continuous bijection g(x) from R^+ to R^+ (i.e. any suitable wiggle) that has lim x->0+ g(x) = 0, there's f(x) = { 0 if x = 0; -g(x) if x > 0; g^-1(-x) if x < 0 }.
22:01:08 <elliott> Lag deletes keypresses now?
22:01:33 <quintopia> if those keypresses are transmitted in a subset of packets which get dropped
22:01:47 <elliott> fizzie: "Wiggle" is the bestest terminology.
22:02:01 <fizzie> quintopia: Anyway, oerjan's generic construction was g . f . g^-1 for any homeomorphism g, and any existing suitable involution f. The (2 - x^(1/3))^3 came from f(x) = -x and g(x) = 1+x^3 or something.
22:02:35 <oerjan> f(x) = 2 - x, g(x) = x^3 is rather more obvious in afterthought :P
22:03:14 <quintopia> the real part of (2 - x^(1/3))^3 does not appear to be symmetrical across y=x
22:03:19 <kallisti> chrome is not standards compliant!!!
22:03:39 <fizzie> quintopia: You're probably not using the right cube root, then.
22:03:49 <fizzie> quintopia: You want the completely real one.
22:04:10 <kallisti> "the animation has to be smooth" lol
22:04:50 <oerjan> that R^+ to R^+ version fizzie mentioned is of course a special case of mine
22:04:52 <kallisti> what. everytime I reload http://acid3.acidtests.org/ I get a different score.
22:05:32 <oerjan> although i suspect that it's also the other way around modulo a translation or so
22:05:59 <elliott> kallisti: yes, that's intentional
22:05:59 <fizzie> oerjan: Yes, and for my "wiggle", f(x) = { k if x = k; k-g(x-k) if x > k; k+g^-1(k-x) if x < k } was an obvious generalization to let the wiggle and mirrored wiggle (the /antiwiggle/) meet anywhere on the y=x line, not just at 0.
22:06:24 <elliott> my chrome passes here, although the "smooth" bit is dubious
22:06:49 <oerjan> oh hm wait not the other way maybe
22:07:16 <kallisti> though sometimes it stops at 97
22:07:23 <oerjan> no reason why R^+ and R^- parts have to reflect each other
22:07:28 <elliott> because of browser bugs, presumably
22:07:33 <elliott> view the source, it'll be in there somewhere
22:08:18 <elliott> //* COMMENTED OUT BECAUSE SVGWG KEEPS CHANGING THIS
22:08:31 <oerjan> quintopia: i also had another construction but that is breaking R into intervals so gives lot of discontinuities
22:08:51 <quintopia> fizzie described a specialization of such
22:09:03 <elliott> the previous acid tests are http://www.w3.org/Style/CSS/Test/CSS1/current/test5526c.htm and http://acid2.acidtests.org/, fwiw
22:09:16 <elliott> my chrome is one or two pixels off acid2 :(
22:09:18 <kallisti> elliott: does it do anything to throw off acceptance of non-standard SGML comment styles?
22:10:47 <kallisti> I can't recall exactly where I read it
22:10:54 <kallisti> but there's something that's commonly accepted but non-standard...
22:11:37 <kallisti> http://htmlhelp.com/reference/wilbur/misc/comment.html
22:12:16 <elliott> kallisti: Browsers do <!-- foo -- bar --> "correctly" these days, IIRC.
22:12:25 <elliott> I believe Hixie, the author of Acid3, pushed Mozilla to but regrets it.
22:12:41 <fizzie> oerjan: It sounds likely that y=f(x) has to be symmetric across y=x for any suitable f, so in some sense it doesn't sound impossible to handwave-argue that having a freely choosable "one half" should be enough to get all of them.
22:15:14 * elliott throws bricks into windows.
22:15:42 <olsner> *when* is alan dipert?
22:15:57 <olsner> is alan always dipert, or only sometimes dipert?
22:16:01 <elliott> olsner: ALAN DIPERT is the owner of esolangs.org.
22:16:08 <elliott> His name is always written in all-caps as of now.
22:16:20 <oerjan> elliott: NOOOO http://xkcd.com/1013/
22:16:39 <quintopia> i can't find a graphing tool on the internet that has x^(1/3) defined for x<0
22:16:42 <elliott> oerjan: No. Fuck you. That was a terrible xkcd and it was days ago so you have no right to remember it.
22:16:54 <elliott> Also, I only said it once, so it's not even a relevant reference.
22:16:58 <fizzie> quintopia: oerjan also cheated W|A to plot it.
22:17:16 <fizzie> quintopia: By doing sgn(x)*abs(x)^(1/3).
22:17:25 <fizzie> As opposed to x^(1/3).
22:17:32 <fizzie> He's quite a cheatoguy.
22:17:33 <olsner> elliott: always so negative against xkcd
22:17:41 <quintopia> fizzie: wtf! that shouldn't even be necessary!
22:18:21 <fizzie> quintopia: Mathematica's idea of "first" cube root is the one that has the smallest (positive) angle, or some-such.
22:18:34 <elliott> olsner: Says our favourite Python-hater.
22:19:02 <kallisti> I hate Python, and use my past experience with Python to help make me sound legit.
22:19:09 <quintopia> fizzie: try that again? which angle?
22:20:12 <fizzie> quintopia: The three cube roots of negative x are at angles pi/3, pi (the negative but real one), and -pi/3. Since they're spaced at intervals of 2pi/3.
22:20:41 <elliott> kallisti: Pretty sure olsner has used Python too.
22:21:09 <kallisti> elliott: what? no, I am the only Python user.
22:21:20 -!- Gregor has set topic: Best for direct log access , 10 year of experience and guarantee quality. See httP://64.62.173.65/%49%27/.%2E/lo%67s/_esoteri%63/#THIS_IS_NOT_A_SCAM.
22:21:46 <elliott> kallisti: BTW click the "A" in "Acid3" to see details about what took a long time.
22:21:51 <quintopia> fizzie: other tools just refuse to pick one, and leave it undef. for -x
22:21:58 <elliott> It reports a test taking many attempts for me, which could explain nondeterminism.
22:22:10 <elliott> kallisti: e.g. perhaps it's busylooping to wait for a DOM element to start existing or such.
22:22:14 <oerjan> if f(x) = y > x, then f(x)-x > 0 > x-f(x) = f(y)-y so there exists x < z < y s.t. f(z) = z
22:22:46 <kallisti> Test 04 failed: expected 'null' but got '[object HTMLDivElement]' - expectation 39 failed
22:22:49 <kallisti> Test 05 failed: expected 'instructions' but got '' - expectation 21 failed
22:22:51 <kallisti> Test 26 passed, but took 197ms (less than 30fps)
22:22:54 <kallisti> Test 43 passed, but took 39ms (less than 30fps)
22:22:56 <kallisti> Test 69 failed: timeout -- could be a networking issue
22:24:27 <oerjan> oh and f is of course bijective again
22:25:41 <olsner> kallisti: you're a Python user? poor soul :(
22:25:41 <oerjan> i think only f(x) = x is increasing
22:26:19 <kallisti> but I do occasionally do Python work
22:26:37 <kallisti> also, before I knew any better, Python was my favorite language.
22:27:16 <olsner> kallisti: favorite. language. wat?
22:27:35 <kallisti> at the time I knew... Python, C, and Java.
22:27:49 <quintopia> where can i find info about how the memory is arranged and managed in GNU c programs?
22:27:49 <olsner> hmm, but you still had C as an alternative
22:27:52 -!- oerjan has set topic: Best for direct log access , 10 year of experience and guarantee quality. See httP://64.62.173.65/%49%27/.%2E/lo%67s/_esoteri%63/#THIS_IS_NOT_A_SCAM | Waiting for Dipert.
22:28:00 <Phantom___Hoover> quintopia, for some reason I know the correct line is "where is John Galt" and now I hate myself.
22:28:00 <kallisti> olsner: this was also before I liked static typing.
22:28:22 -!- olsner has set topic: Best for direct log access , 10 year of experience and guarantee quality. See httP://64.62.173.65/%49%27/.%2E/lo%67s/_esoteri%63/#THIS_IS_NOT_A_SCAM | Waiting for DIPERT.
22:28:23 <elliott> Phantom___Hoover: it isn't
22:28:42 <elliott> your mind remains unfucked by shittery
22:28:51 <quintopia> i just watched the movie last night cuz my mom wanted to
22:28:57 <quintopia> i was far more terrible than i expected
22:29:33 <kallisti> so I have one of those rolling chairs that let you adjust height
22:29:38 <kallisti> which means I'm not going to be productive anymore.
22:29:44 <elliott> quintopia: did u kkno it's only part 1
22:29:56 <kallisti> I'm just going to roll around while pulling levers
22:30:01 <quintopia> i'm not sure if i want to suffer part 2 tho
22:30:04 <elliott> yes EVEN MORE shrugging to come
22:30:25 <elliott> atlas shrugged, atlas shrugged 2: shrug harder
22:31:22 <fizzie> Which is a total shrugfest of a book.
22:32:17 <kallisti> Atlas Shrugged Redux: shrug hard, or shrug trying
22:33:18 <monqy> film that's just a guy shrugging, put on loop
22:33:44 <olsner> so what is atlas shrugged about?
22:34:21 <elliott> olsner: RATIONAL SELF-INTEREST
22:34:44 <elliott> (advanced enlightened self-interest)
22:40:09 <elliott> olsner: tl;dr trains trains trains BEST NEW METAL EVER "fuck no you can't do that" -govt WHO IS JOHN GALT WHO IS JOHN GALT WHO IS JOHN GALT WHO IS JOHN GALT "I AM JOHN GALT [speech that literally lasts tens and tens of pages]" and then they all go to live in a perfect objectivist captalist utopia
22:40:18 -!- tzxn3 has quit (Read error: Connection reset by peer).
22:41:07 <olsner> new metal? is that about rock music or about materials?
22:41:56 <elliott> I believe it predates metal.
22:43:14 <olsner> so it's about the stone age then?
22:44:00 <olsner> they have cars and stuff in the flintstones, why not
22:47:13 <elliott> How long do North Carolinians sleep?!
22:48:12 -!- MoALTz_ has joined.
22:48:36 <Phantom___Hoover> (He said that he wasn't going to subject the hamsters to *that* scene.)
22:49:32 -!- pikhq has joined.
22:49:33 <elliott> Phantom___Hoover: Ayn Rand's sex scenes are usually (universally?) rapes.
22:49:44 <elliott> The victim comes out the other end liking it, obviously.
22:49:47 <elliott> It's Ayn Rand, what do you expect?
22:50:05 -!- pikhq_ has quit (Ping timeout: 276 seconds).
22:50:11 <elliott> I think there's something like 3 in Atlas Shrugged.
22:50:46 <quintopia> elliott: which is why i was surprised that the sex scene in the movie was only an extramarital affair (oh, and there was that one other one where it was boring and intra-marital)
22:51:21 -!- MoALTz__ has quit (Ping timeout: 244 seconds).
22:51:47 <elliott> quintopia: Well, Randians tend to internalise that crap and bullshit it off when anyone points it out, so it's not surprising they'd tune it down for the "masses".
22:52:15 <quintopia> elliott: ugh, i'd have rather seen 2 hours of rape than what i saw
22:55:16 <olsner> pretty sure you could've done that, why didn't you?
22:55:51 -!- elliott has quit (Remote host closed the connection).
22:59:52 -!- Chef_ has quit (Remote host closed the connection).
23:01:18 -!- Chef_ has joined.
23:05:09 <quintopia> she would not have preferred rape.
23:05:26 <quintopia> although she agrees it was a terrible movie
23:12:10 -!- elliott has joined.
23:14:27 <elliott> http://upload.wikimedia.org/wikipedia/commons/e/ec/Acid2_in_Opera_Mini_4.png
23:14:36 -!- itidus21 has joined.
23:16:54 <Phantom___Hoover> http://www.reddit.com/r/skyrim/comments/pwp2h/what_i_want_to_see_as_console_dlc_if_you_play_on/c3sw483
23:18:04 <elliott> Phantom___Hoover: Santa can't exist; telling whether someone is naughy or nice is the Halting problem. Q.E.D.
23:18:14 <elliott> Faster-than-light travel? Halting problem.
23:18:22 <elliott> Predicting the weather? Halting problem.
23:18:32 <elliott> Designing skyscrapers? Halting problem.
23:19:13 <quintopia> oh, so you meet lots of girls elliott?
23:19:26 <elliott> You don't have to do something if you formally prove it's possible!
23:19:29 <elliott> That's the Halting problem.
23:20:00 <quintopia> have you formally proved its my problem?
23:20:55 -!- MoALTz__ has joined.
23:21:45 <quintopia> really? could have sworn it was on hilbert's list
23:22:06 <oerjan> no, i think that was on feynman's list
23:23:17 -!- MoALTz_ has quit (Ping timeout: 244 seconds).
23:24:06 -!- Phantom___Hoover has quit (Read error: Connection reset by peer).
23:26:22 -!- azaq23 has joined.
23:26:33 -!- azaq23 has quit (Max SendQ exceeded).
23:27:25 -!- azaq23 has joined.
23:30:19 <elliott> Estimated time until 7:00 ADT: five years.
23:30:59 <zzo38> What is this ADT supposed to be?
23:31:19 <elliott> His alarm is set for 7:00, naturally.
23:31:32 <shachaf> elliott: So I installed Debian testing.
23:31:45 <shachaf> Life is good, except for all the bad parts.
23:33:18 <shachaf> Over the wireless network.
23:33:41 <shachaf> It involved a bit of trickery.
23:33:51 <shachaf> It also took 3 hours or more, over my connection.
23:34:13 <shachaf> I should've gone somewhere with a faster connection.
23:34:15 <elliott> Also, what are you using for X?
23:34:29 <shachaf> The trickery was something about something.
23:34:38 <shachaf> In general the support is on the netinst CD, more or lessish.
23:35:13 <shachaf> I'm using Xfce for the moment. Not particularly happy with it but I didn't want to set something more complicated up.
23:35:45 <elliott> You can enable dragging items on the taskbar, by the way, but it's hard to use (doesn't like to accept the drags).
23:35:48 <elliott> (That was one of my main annoyances.)
23:36:12 <shachaf> I have bigger annoyances than that.
23:36:27 <shachaf> It keeps resetting my caps lock key to lockin' the caps. :-(
23:37:42 -!- MoALTz_ has joined.
23:39:58 <shachaf> Why are these things all huge evil monoliths?
23:40:36 <shachaf> Desktop environment things.
23:40:44 <elliott> Because that's the definition of a desktop environment.
23:40:49 <elliott> Anyway, you didn't install xfce-goodies, did you?
23:40:51 -!- MoALTz__ has quit (Ping timeout: 244 seconds).
23:40:54 <elliott> That contains all the terrible Xfce applications.
23:41:12 <oerjan> shachaf: don't ask us, ask the aliens who made them
23:42:08 <shachaf> Doesn't look like I installed it.
23:42:24 <shachaf> What happened to the GNOME people, anyway?
23:42:41 <shachaf> GNOME used to be an OK Windows clone, more or less. Now it's a terrible OS X clone, more or less.
23:42:54 <elliott> GNOME 3 is nothing like OS X.
23:43:09 <elliott> For that matter, GNOME 2 is not that much like Windows.
23:43:10 <shachaf> I assume that it'll eventually turn into an OK OS X clone. It was a terrible Windows clone once, after all.
23:43:28 <shachaf> Why does it use Alt-tab to switch between applications and Alt-` to switch between windows of an application?
23:43:52 <Jafet> The Windows desktop is designed with strange but consistent principles
23:44:28 <Jafet> The Gnome 2 desktop is like humans discovering ancient alien technology and copying it
23:44:50 <zzo38> I do know when I work on the Ubuntu at FreeGeek, I noticed many things that didn't work like Windows.
23:45:09 <zzo38> Although many things did work like Windows.
23:45:11 <elliott> shachaf: OS X doesn't use those, either.
23:45:30 <elliott> Jafet: GNOME 2 is far more consistent than Windows circa XP.
23:45:59 <shachaf> Windows circa XP is pretty consistent.
23:46:21 <elliott> Jafet: What's inconsistent about GNOME 2?
23:47:18 <Jafet> Many gtk dialogs have small inconsistencies, such as the fact that you cannot undo in a certain kind of text input box
23:47:52 <Jafet> Or some dropdown selectors cannot be used by keyboard
23:48:01 <elliott> That undo thing applies to Windows too, I'm pretty sure.
23:48:11 <elliott> I don't think you can undo those dialog-box-with-text-entry-field things.
23:48:14 <Jafet> Alien technology indeed.
23:48:30 <zzo38> I thought it was like Windows, and expected some things to work. Some things worked and some things didn't work. Double-clicking the control box to close the window didn't work. Dragging a scrollbar and then moving the mouse pointer out of range before releasing the button to snap back didn't work, double clicking the title bar to maximize did work though.
23:48:33 <Jafet> Actually, it's often impossible to use Gnome programs with only the keyboard
23:48:38 <elliott> Seriously though, GNOME 2 is by no means a Windows clone, and I'm not saying GNOME 2 is very good.
23:48:46 <elliott> But it's better than Windows.
23:48:48 <Jafet> The KDE desktop is worse in that area, though.
23:49:29 <shachaf> Jafet: It's much more often impossible in GNOME 3.
23:49:29 <elliott> For a start, Microsoft would never arrange the panels the way GNOME 2 did (which I personally like).
23:50:56 <elliott> OK, it's been 42 hours since I sent that email.
23:51:04 <elliott> When's the acceptable time to get impatient and send a follow-up?
23:51:54 <Jafet> 1337 microfortnights
23:52:11 <elliott> `frink 1337 microfortnights -> hours
23:52:22 <HackEgo> 28077/62500 (exactly 0.449232)
23:52:35 <elliott> Is that from now or from when I sent it?
23:53:15 <Jafet> Outlook hazy, please try again later
23:54:35 <elliott> I'll check back in 0.449232 hours.
23:58:30 -!- Chef_ has quit (Remote host closed the connection).
23:59:47 -!- PiRSquared17 has quit (Remote host closed the connection).