00:01:25 wait it's not even mine... mumble... 00:01:44 then you must make one 00:04:30 -!- J_Arcane_ has joined. 00:06:51 -!- J_Arcane has quit (Ping timeout: 246 seconds). 00:06:54 -!- J_Arcane_ has changed nick to J_Arcane. 00:08:58 -!- MDude has joined. 00:13:04 ah, but this one is mine at least: http://esoteric.voxelperfect.net/files/kayak/src/sort3.kayak 00:14:36 Or perhaps more usefully, http://int-e.eu/~bf3/eso/mergesort.kayak 00:16:15 One thing that I find weird is there was a time when people didn't know about things like mergesort and quicksort 00:16:22 I've *met* the person who invented quicksort 00:17:10 did you grovel properly twh 00:19:13 Taneb: how did they invent it? 00:19:22 did they sit down one day and decide to write a better sort algorithm? 00:19:35 Hm, Hoare 00:19:47 a hoary algorithm 00:19:55 Very 00:21:22 -!- Caesura has joined. 00:21:50 oh back then oerjan went by "Orjan" 00:22:13 i did? 00:22:32 well, in one place (real name that went with the email address) 00:22:47 my institute mail address, probably 00:23:08 the signature reads =D8rjan (less doesn't do mime decoding ;-) 00:23:53 -!- heroux has joined. 00:24:14 FireFly, apparently he was working somewhere and asked "Could you use shellsort to sort these?" to which he said "I bet you a tenner I can do it quicker" 00:24:17 And invented quicksort 00:24:54 (this story has two errors in it) 00:24:57 int-e: probbly Örjan then? 00:25:00 (it was a sixpence bet) 00:25:00 hm or maybe i just had set my realname that way. charsets were brittle back then. 00:25:10 FireFly: don't be ridiculous. 00:25:10 (and he had alrady invented it, while a visiting student in Moscow) 00:25:38 FireFly: no, 0xD8 is Ø 00:25:42 Er, right 00:25:45 I meant that 00:25:48 quicksort was invented when it was noted that nobody had yet invented a sorting algorithm that was quick 00:25:51 the rest is history 00:25:53 * FireFly must be tired 00:25:56 Œrjan 00:26:22 Everyone knows oe is the ASCIIfication of ö though :p 00:26:43 I don't know which of Taneb's and Phantom_Hoover's stories to believe 00:26:56 FireFly, they don't actually contradict each other 00:26:57 mine was a lie 00:27:10 Taneb: true 00:27:21 Phantom_Hoover: why would you do that 00:27:28 quicksort isn't so named because it's quick at all; after all, it's O(n^2) in the worst case! 00:27:42 Phantom_Hoover: I don't want to hear your explanation of bucket sort--it's bound to be gross. 00:27:52 now i kind of want to. 00:28:04 its name in fact comes from the idiom 'to cut to the quick', a reference to its pivot algorithm 00:29:20 Phantom_Hoover, I thought it was because Tony Hoare was drinking a Nesquik when he authored the algorithm 00:29:57 Anyway, I have to go 00:30:02 I have a bed that needs sleeping in 00:30:51 i understand that people can be hired to do that for you 00:30:56 . o O ( what will happen to the bed if not ) 00:32:06 Maybe the bed is lonely 00:33:24 -!- tromp has joined. 00:44:18 * hppavilion[1] is Judge, Jury, and Instruction Execution Unit <-- . o O ( a CPU based on montesquieu principles ) 00:44:51 oerjan: Wasn't that the PDP-11? 00:45:07 i don't know? 00:45:36 *in the 00:46:24 -!- tromp has quit (Remote host closed the connection). 00:48:36 ahh, good ol' PDP-11. 00:53:55 -!- Caesura has quit (Ping timeout: 268 seconds). 00:54:32 -!- hppavilion[1] has joined. 01:06:49 Big-Λ notation is my favorite notation. 01:11:17 Which is doing what? 01:12:19 zzo38: Not sure yet. Probably it curries them together 01:13:27 So $\LAM{\{\lambda x.x, \lambda x.\lambda y.x, \lambda x.\lambda y.y\}} is... 01:13:28 Hm... 01:13:36 OK, I'm just bullshitting, it turns out. 01:14:22 I'm as surprised as you guys 01:30:28 hppavilion[1]: the one use of big lambdas i remember is in System F, where it denotes a type-level lambda 01:30:41 oerjan: Thank you 01:32:03 so e.g. id = \Lambda t. lambda x : t. x is the polymorphic id 01:32:19 * \lambda 01:32:33 or thereabouts 01:33:20 [http://www.tutorialspoint.com/artificial_intelligence/artificial_intelligence_overview.htm] Artificial Intelligence is a way of making a computer, a computer-controlled robot, or a software think intelligently, in the similar manner the intelligent humans think. 01:33:24 "The intelligent humans" 01:33:38 I think that Tutorialspoint may have produced a strong AI 01:33:45 oh that's the first example in wikipedia's article on System F. notation a bit different though. 01:34:02 \LAmbda k. \Lambda t : k. \lambda x : t. ... 01:34:13 this is why there are seven levels in the hierarchy 01:34:55 shachaf: Why? 01:35:07 shachaf: Shouldn't there be 64 levels? 01:35:22 :P 01:36:20 /j #ruby 01:36:27 -!- J_Arcane_ has joined. 01:37:51 fowl: ? 01:38:16 -!- J_Arcane has quit (Ping timeout: 252 seconds). 01:38:16 -!- hydraz has quit (Ping timeout: 252 seconds). 01:38:20 hppavilion[1]: I think there was a space in the buffer 01:38:21 -!- J_Arcane_ has changed nick to J_Arcane. 01:38:32 -!- Caesura has joined. 01:38:49 -!- Xe has quit (Ping timeout: 252 seconds). 01:38:49 -!- relrod has quit (Ping timeout: 252 seconds). 01:38:49 fowl: Ah 01:39:22 -!- ineiros has quit (Ping timeout: 252 seconds). 01:39:22 -!- acertain has quit (Ping timeout: 252 seconds). 01:39:22 -!- newsham has quit (Ping timeout: 252 seconds). 01:40:37 -!- ineiros has joined. 01:43:02 -!- hydraz has joined. 01:43:02 -!- hydraz has quit (Changing host). 01:43:02 -!- hydraz has joined. 01:43:07 -!- tromp has joined. 01:43:43 FireFly: the danger of asking in #canaima-social is that we might trigger some canaima users to come here for the right purpose... and then we cannot ban it... 01:43:59 Good point 01:44:28 -!- relrod_ has joined. 01:44:30 -!- relrod_ has quit (Changing host). 01:44:30 -!- relrod_ has joined. 01:46:01 -!- Xe has joined. 01:46:19 * hppavilion[1] is learning him a haskell under the promise of it bringing about great good <-- *MWAHAHAHA* i mean, exactly! 01:47:26 there is some blog post which claims LYAH isn't actually that pedagogical, though, and suggests a different path. 01:47:41 (based on some online courses iiuc) 01:49:03 -!- Caesura has quit (Ping timeout: 240 seconds). 01:50:31 if it's the post i'm thinking of, i don't recommend that advice too much 01:50:37 but maybe it happens to be correct 01:51:17 one of the courses is called FPsomenumbers iirc 01:51:56 -!- Phantom_Hoover has quit (Remote host closed the connection). 01:52:42 -!- acertain has joined. 02:03:01 -!- relrod_ has changed nick to relrod. 02:05:57 -!- Caesura has joined. 02:06:28 * hppavilion[1] walks in wearing his swatting shield 02:06:32 "Full-stack web developer" is a stupid concept 02:06:56 By the computer-scientific definition of a stack, it holds infinite data, and thus cannot be full. 02:07:40 I forgot to laugh. 02:11:06 -!- Caesura has quit (Ping timeout: 246 seconds). 02:16:34 -!- Kaynato has joined. 02:17:08 -!- centrinia has joined. 02:18:09 * oerjan whacks hppavilion[1] with the saucepan ===\__/ 02:18:43 * hppavilion[1] tries to deflect, but he is equipped only for swat-proofing 02:37:37 -!- newsham has joined. 03:05:16 -!- acertain has quit (Ping timeout: 268 seconds). 03:19:00 -!- jaboja has quit (Ping timeout: 276 seconds). 03:23:11 Well another channel that does not start with 'es' just had someone looking for #canaima-social join 03:23:30 So I think the theory that the 'es' is the crux of the problem is disproved 03:27:20 ic 03:28:35 -!- hppavilion[1] has quit (Ping timeout: 250 seconds). 03:28:46 but it's not _every_ channel. never seen any canaimas joining ##nomic. 03:29:16 Mysterious 03:29:34 Well, I don't know what would make #esoteric and #dolphin-emu similar 03:29:47 Except maybe that both are moderately large and not secret? 03:34:49 is anyone else in channels seeing many canaimans (or, if you don't notice the username/domain, people inexplicably speaking spanish) 03:37:03 wild option: it's all just a single troll faking this 03:37:52 -!- XorSwap has joined. 03:53:34 -!- MDude has changed nick to MDream. 03:59:22 -!- Caesura has joined. 04:07:35 well. 04:09:52 Well!??? 04:15:47 so it turns out that 200 kilofurlongs per fortnight is a reasonable highway driving speed in canada 04:16:19 this program to add 1 and 2, print the result, compare it to 1, if 1 is less than 3, print 3, else, print 0. 04:16:30 takes up 160 bytes. 04:18:05 that's not -too- bad. 04:18:46 but it's lame. 04:19:38 coppro: What highway? Different highways will often have different speeds 04:21:42 `frink 200 200 kilofurlongs per fortnight => km/h 04:21:45 oops 04:21:54 `frink 200 kilofurlongs per fortnight => km/h 04:22:57 No output. 04:22:57 No output. 04:23:08 `frink 200 kilofurlongs / fortnight => km/h 04:23:08 oerjan: it's about 1km/hr. 04:23:15 1.19km/hr 04:23:26 wtf is with HackEgo 04:23:44 fizzie: :( 04:23:49 No output. 04:23:54 eh what? 04:24:08 200 kilofurlongs is around 402 kilometers, one fortnight is 336 hours. 04:24:16 http://www.wolframalpha.com/input/?i=200+kilofurlongs+per+fortnight+in+km%2Fh 04:24:25 `echo hi 04:24:26 hi 04:24:29 do out the math. 04:24:32 `frink 200 kilofurlongs / fortnight => km/h 04:24:49 am i doing something wrong 04:25:05 No output. 04:25:06 or is frink beyond HackEgo's resources 04:25:13 `frink 2*2 04:25:30 shit, I fucked up a couple decimal places lmao. 04:25:40 `file bin/frink 04:25:53 No output. 04:25:53 bin/frink: POSIX shell script, ASCII text executable 04:26:13 did someone break frink 04:26:33 `url bin/frink 04:26:35 what's frink? 04:26:36 http://codu.org/projects/hackbot/fshg/index.cgi/file/tip/bin/frink 04:26:46 picobit: a programming language with unit conversions 04:26:58 nice. 04:27:02 useful. 04:27:24 `file lib/frink 04:27:27 lib/frink: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.15, BuildID[sha1]=0x3d3ad786ec6233455da8a3371b38b238b692f3e1, not stripped 04:27:32 well it used to be when it worked 04:27:55 it is possible i have just forgot the syntax 04:28:27 mayhaps. 04:28:45 `frink 2 m/s => km/h 04:29:07 `` ls -l lib/frink 04:29:27 -rwxr-xr-x 1 5000 0 4586628 Dec 9 04:12 lib/frink 04:29:28 No output. 04:29:37 `frink 2 m/s in km/h 04:29:56 `frink 2 m/s -> km/h 04:30:17 i guess it shouldn't be taking this long regardless of what the syntax is. 04:30:22 No output. 04:30:33 No output. 04:30:46 `` frink "2*2" (just testing if it's connected to reading input) 04:31:07 4 04:31:10 huh 04:31:27 `` frink "2 m/s => km/h" 1 error(s) occurred during parsing. \ Syntax error: frink.expr.b9@45c7efd8, line 1, near column 7 04:32:03 `` frink "2 m/s -> km/h" Conformance error \ Left side is: 2 m s^-1 (velocity) \ Right side is: 1.5091904506831453200e+36 m^-1 s kg^-1 (unknown unit type) \ Suggestion: divide left side by energy \ \ For help, type: units[energy] \ to list known units with these dimensions. 04:32:51 `` frink "2 m/s -> km/hour" 36/5 (exactly 7.2) 04:33:26 i certainly seem to recall `frink being faster than this 04:34:01 `frink 200 kilofurlongs / fortnight -> km/hour 04:34:11 Try writing the entire words "kilofurlong/fortnight" and "kilometre/hour" I think is what you need? 04:34:19 km is ok 04:34:20 3300000/27559 (approx. 119.7430966290504) 04:34:32 zzo38: nah that was close enough 04:34:57 zzo38: it just was the h it didn't understand (and the => i used, as well as timing out) 04:35:03 -!- iconmaster_ has quit (Ping timeout: 276 seconds). 04:35:07 i assuming it interprets h as planck's constant 04:35:08 On my computer I used the "units" program and got 119.74286 (so perhaps it is not quite as accurate? I don't know) 04:35:10 *assume 04:35:34 Maybe frink calculate directly by the fractions so it can be more accuracy 04:35:40 maybe 04:35:55 > 3300000/27559 04:35:56 119.7430966290504 04:36:38 -!- bb010g has joined. 04:39:54 so I am faced with a dillema. 04:40:24 in this virtual machine, I have no "instructions". everything consists of moving data from one place or another. 04:40:50 which means that programs consist of source:destination pairs. 04:40:56 as well as some raw data. 04:41:00 like constants, etc. 04:41:32 mhm 04:42:10 you perform calculations via writing values to specific mapped locations in memory. 04:42:15 * oerjan reminded of MOV-only programming, and also Resplicate. 04:42:37 -!- Marcela_Gandara has joined. 04:43:00 so your arithmetic logic unit is mapped to one location, your branch unit another.. etc. 04:43:37 -!- variable has joined. 04:43:50 ah. i think that's also a known OISC type 04:44:07 I can say I have once designed schematics for something like that 04:44:08 it's a modified transport-triggered architecture. 04:44:32 All instructions were move one register to another register and jump unconditionally. 04:44:59 Hoever the address to read instruction from is not only the address in the previous instruction but there are a few extra bits which are the condition flags. 04:45:25 one sec, I'll show an example. 04:45:43 -!- variable has changed nick to constant. 04:46:11 http://ideone.com/cf3d5y 04:46:25 uncomment the debug define at the top to show the transfers as they execute. 04:46:44 on the one hand, this is stupidly simple. 04:46:59 and it's very cool. 04:47:35 on the other hand, the "instructions" are pretty large, and something like adding two numbers together requires a bit more overhead. 04:47:58 the intention is to form a TIS-100-like mesh of nodes that can pass values to eachother and perform calculations. 04:48:08 this is the implementation for a node. 04:48:19 (at least, just a small, simple example) 04:48:52 btw, line 82 lies. 04:49:45 -!- Kaynato has quit (Quit: Connection closed for inactivity). 04:50:46 so. either I go ahead and implement something like a RISC (and make an assumption about the bit-width of the machine), I implement a stack machine (all well and good, save for the odd primitives), or say "fuck it" and accept that dealing with this kind of sizing is kind of required. 04:51:14 -!- tromp has quit (Remote host closed the connection). 04:53:01 * Marcela_Gandara va a dormir 04:54:46 I think it's quite an interesting architecture that rightfully segments things up into functional units. 04:55:03 simple on implementation, somewhat simple programming.. generating code for it wouldn't be too hard. 04:56:24 -!- Marcela_Gandara has quit (Quit: Leaving.). 04:56:33 -!- tromp has joined. 05:16:36 http://ideone.com/cf3d5y 05:17:04 this is what it takes to add two arbitrary constants together. 05:20:26 -!- hppavilion[1] has joined. 05:24:16 -!- MoALTz has joined. 05:42:24 -!- tromp has quit (Remote host closed the connection). 05:47:18 -!- XorSwap has quit (Quit: Leaving). 05:49:03 I set up a Fossil repository and it is much easier than git. One thing is autosync mode, and you do not necessarily have to copy all of the files, also you do not need a directory structure with a lot of files because instead the repository is just one file and it is a SQL database so you can even execute SQL commands on it. The commands are also simpler to operate, and it is just a single executable file. 05:50:33 However the webpage lacks many keyboard commands. I could easily add them for the main menu and ticket menu, but not for the wiki. 05:53:02 damn, writing machine code for this thing is.. oddly relaxing. 05:57:42 My design was very different. It used a separate memory for program and data, and the data memory address was also a register and then another register to read/write data at that address. It used separate registers for the result of different arithmetic operations, but most operations shared the same registers for input (a few used the output register also for input). 05:58:30 Also each instruction included half of the address of the next instruction; the other half was a register that you could write to. 06:00:30 https://ptpb.pw/046m 06:00:40 a simple program to count from 5 to 0. 06:01:24 -!- Caesura has quit (Ping timeout: 246 seconds). 06:01:46 err, shit. 06:01:54 that 'add' should be a 'sub'. 06:18:42 -!- lambda-11235 has quit (Read error: Connection reset by peer). 06:28:26 nitrix: ping. 06:33:29 picobit: Counting from 5 to infinity is nice too 06:33:37 FreeFull: :P 06:36:04 the MIPS equivalent of that program is 16 bytes. that program is 80. a stack machine equivalent would be 40 bytes. 06:37:33 price you pay for simplicity, I suppose. this assumes a 32-bit architecture. 06:43:05 -!- tromp has joined. 06:48:18 -!- tromp has quit (Ping timeout: 276 seconds). 07:05:08 Functor:map::?:filter 07:07:20 -!- constant has quit (Ping timeout: 276 seconds). 07:13:33 hppavilion[1]: MonadPlus can fit there, although it's probably too powerful. 07:13:56 :t mfilter 07:13:57 MonadPlus m => (a -> Bool) -> m a -> m a 07:15:03 Huh 07:15:24 Alternative+Traversable might work too, i think 07:15:39 -!- rdococ has joined. 07:19:03 :t asum 07:19:04 (Foldable t, Alternative f) => t (f a) -> f a 07:20:27 i think you need MonadPlus to keep any structure more complicated than a list. 07:31:07 -!- augur_ has quit (Ping timeout: 260 seconds). 07:33:48 -!- pelegreno_ has quit (Remote host closed the connection). 07:35:23 -!- pelegreno has joined. 07:42:25 Hm... 07:42:44 Has anybody Collatzized something >NN? 07:43:23 -!- J_Arcane has quit (Ping timeout: 268 seconds). 07:43:42 makes no sentence your sense 07:44:32 -!- J_Arcane has joined. 07:48:54 -!- oerjan has quit (Quit: Nite). 07:52:00 -!- augur has joined. 07:52:20 @tell oerjan NN is the natural numbers 07:52:20 Consider it noted. 07:55:22 -!- haavardp has changed nick to havard. 07:55:24 -!- havard has changed nick to haavard. 08:00:25 -!- zadock has joined. 08:09:10 -!- bb010g has quit (Quit: Connection closed for inactivity). 08:24:55 Most people seem to think logical argumentation is about witty comebacks. 08:25:04 Usually we would facepalm at how wrong they are 08:25:10 But let's try going with the flow 08:25:16 How do you formalize witty comebacks? 08:25:19 08:25:49 by realizing them for what they are. 08:25:58 an approximation of an argument. 08:26:03 not an actual argument 08:28:22 -!- zadock has quit (Quit: Leaving). 08:38:51 http://codepad.org/4BCgVrxO 08:39:12 although variables are kinda unhandy 08:39:28 (whilecall, ifcalls don't create a new scope so you inherit the scope) 08:43:00 -!- tromp has joined. 08:47:10 -!- tromp has quit (Ping timeout: 244 seconds). 08:57:03 -!- hppavilion[1] has quit (Ping timeout: 250 seconds). 09:44:12 -!- tromp has joined. 09:48:47 -!- tromp has quit (Ping timeout: 260 seconds). 09:50:32 -!- centrinia has quit (Ping timeout: 260 seconds). 09:58:42 -!- rodgort has quit (Ping timeout: 260 seconds). 10:04:32 -!- MDream has quit (Ping timeout: 260 seconds). 10:06:35 -!- rodgort has joined. 10:22:59 -!- picobit has quit (Ping timeout: 260 seconds). 10:31:55 -!- AnotherTest has joined. 11:31:02 -!- boily has joined. 11:43:54 `wisdom 11:44:21 metar/metar is a service that allows nerds to talk about the weather. 11:57:07 @metar lowi 11:57:08 LOWI 211050Z 25003KT 200V310 CAVOK 20/06 Q1021 NOSIG 11:58:11 int-ello. enjoying the summer? 11:58:59 -!- earendel has quit (Ping timeout: 276 seconds). 12:02:37 bon matily 12:02:59 @metar katl 12:02:59 KATL 211052Z 10005KT 10SM FEW180 13/09 A3009 RMK AO2 SLP185 T01280094 12:04:41 QUINTHELLOPIA! 12:04:56 you're morningy! 12:05:01 @metar CYUL 12:05:01 CYUL 211033Z 16004KT 30SM FEW180 FEW240 02/M03 A3010 RMK AC1CI1 CI TR SLP194 12:08:00 -!- nisstyre has quit (Ping timeout: 276 seconds). 12:08:20 -!- nisstyre_ has joined. 12:09:02 yes im always up but rarely get on irc 12:09:46 today im in a hotel 12:11:41 dun dun dun ♪ 12:14:30 @metar EFHK 12:14:31 EFHK 211050Z 28010KT 9999 FEW040 09/M06 Q1011 NOSIG 12:14:55 Not quite summer, but not bad either. 12:15:16 are you always so morningy? 12:15:34 Yesterday there was a bit of sleet. 12:15:56 It's 2:15pm here. 12:16:08 boily 12:16:52 -!- jessi has quit (Ping timeout: 250 seconds). 12:17:53 quintopia: yeah, between 6:30am and 7:20am. 12:18:20 oh thats mostly too early for me 12:18:27 have a good workday 12:18:38 I was AAAAAAAAAAAAAAAAH yesterday! 12:19:29 good hosteling! and/or whatever you are disappearing at because it'd be weird to not go outside, or something that you the way prefer to be! 12:19:34 (I need coffee...) 12:19:52 -!- boily has quit (Quit: HELICOIDAL CHICKEN). 12:25:28 -!- jaboja has joined. 12:46:03 -!- tromp has joined. 12:49:03 -!- J_Arcane has quit (Ping timeout: 240 seconds). 12:50:22 -!- tromp has quit (Ping timeout: 252 seconds). 13:25:12 -!- J_Arcane has joined. 13:48:54 i'm back did you miss me? 13:49:02 Hi, izabera 13:49:07 hi 13:50:38 `? Taneb 13:50:58 Taneb is not elliott, no matter who you ask. He also isn't a rabbi although he has pretended in the past. He has at least two backup keyboards with dodgy SHIFT KEys, cube root of five genders, and voluminous but calm eyebrows. (See also: tanebventions) 13:52:36 `run sed -i s/five/eight/ wisdom/taneb 13:52:49 No output. 13:52:51 `? Taneb 13:52:55 Taneb is not elliott, no matter who you ask. He also isn't a rabbi although he has pretended in the past. He has at least two backup keyboards with dodgy SHIFT KEys, cube root of eight genders, and voluminous but calm eyebrows. (See also: tanebventions) 13:53:13 I'm feeling a bit more integral 14:02:20 -!- kline has quit (Changing host). 14:02:20 -!- kline has joined. 14:16:43 -!- jaboja has quit (Ping timeout: 252 seconds). 14:20:20 -!- Kaynato has joined. 14:45:18 -!- `^_^v has joined. 14:46:34 -!- jaboja has joined. 14:51:43 -!- p34k has joined. 14:52:40 do the genders have names? 14:53:41 quintopia, no 14:54:03 (I'm probably gender-fluid) 14:55:43 then you should have [1,2] genders. The entire interval. 14:56:20 although, to be more precise, you should have a distribution on that interval 14:56:32 as you're probably not uniform 14:56:49 I'm not a huge fan of applied maths 14:56:58 what about type systems? 14:57:33 They're not applied when I'm done with them 14:58:09 can we say you are Maybe-gendered? 14:59:15 That's probably inaccurate 14:59:44 i don't understand types 15:06:53 <\oren\> idea: a compiler from C into C 15:07:12 <\oren\> with optimization 15:12:51 -!- mroman has quit (Quit: Lost terminal). 15:24:07 -!- jaboja has quit (Remote host closed the connection). 15:29:48 -!- jaboja has joined. 15:47:40 -!- tromp has joined. 15:52:32 -!- tromp has quit (Ping timeout: 276 seconds). 16:12:42 -!- acertain has joined. 16:14:25 -!- acertain has quit (Client Quit). 16:20:09 -!- Kaynato has quit (Ping timeout: 268 seconds). 16:21:49 -!- gamemanj has joined. 16:40:57 -!- Reece` has joined. 16:45:35 -!- iconmaster_ has joined. 16:50:06 -!- earendel has joined. 16:58:48 -!- p34k has quit. 17:03:36 -!- nycs has joined. 17:05:20 -!- `^_^v has quit (Ping timeout: 276 seconds). 17:05:51 -!- Reece` has quit (Ping timeout: 250 seconds). 17:07:44 -!- Reece` has joined. 17:12:00 -!- J_Arcane has quit (Ping timeout: 246 seconds). 17:19:12 -!- nycs has quit (Quit: This computer has gone to sleep). 17:19:40 -!- `^_^v has joined. 17:22:09 -!- zgrep has quit (Quit: EOF). 17:22:53 -!- zgrep has joined. 17:23:40 -!- jaboja has quit (Ping timeout: 268 seconds). 17:36:04 -!- jaboja has joined. 17:43:10 -!- jaboja has quit (Ping timeout: 250 seconds). 17:44:02 -!- acertain has joined. 17:48:49 -!- tromp has joined. 17:53:05 -!- tromp has quit (Ping timeout: 250 seconds). 17:53:44 * FireFly . o O ( clang → llvm optimizer → llvm-to-C backend ) 17:54:35 -!- Kaynato has joined. 17:56:39 FireFly: I wonder if that would get better or worse for every iteration 17:57:53 -!- jaboja has joined. 17:58:00 Only one way to find out 17:58:28 I think/hope it would reach a fixpoint after one iteration 17:58:40 But maybe that's wishful thinking 17:59:15 that would be cool 17:59:17 do it 17:59:22 and subscribe me to your mailing list 17:59:55 -!- iconmaster__ has joined. 18:01:33 -!- nitrix has quit (Ping timeout: 268 seconds). 18:02:13 -!- Caesura has joined. 18:02:28 -!- MoALTz has quit (Ping timeout: 268 seconds). 18:02:31 ...people still use mailing lists? 18:03:12 -!- staffehn has joined. 18:04:01 -!- staffehn_ has quit (Ping timeout: 268 seconds). 18:04:02 -!- iconmaster_ has quit (Ping timeout: 268 seconds). 18:04:17 -!- nitrix has joined. 18:04:50 -!- fungot has quit (Ping timeout: 250 seconds). 18:05:13 -!- jaboja has quit (Ping timeout: 250 seconds). 18:05:33 -!- Kaynato has quit (Ping timeout: 240 seconds). 18:06:57 -!- picobit has joined. 18:08:47 -!- Lord_of_Life has quit (Ping timeout: 297 seconds). 18:09:23 -!- Froox has joined. 18:13:41 -!- Lord_of_Life has joined. 18:15:58 -!- Zoroaster has joined. 18:17:51 -!- shikhin_ has joined. 18:19:13 -!- rodgort has quit (*.net *.split). 18:19:13 -!- Alcest has quit (*.net *.split). 18:19:13 -!- Frooxius has quit (*.net *.split). 18:19:13 -!- shikhin has quit (*.net *.split). 18:19:13 -!- mbrcknl has quit (*.net *.split). 18:19:38 -!- Caesura has quit (Ping timeout: 244 seconds). 18:20:14 -!- rodgort has joined. 18:23:40 -!- mbrcknl has joined. 18:24:04 -!- hppavilion[1] has joined. 18:27:24 -!- MoALTz has joined. 18:28:21 gamemanj: of course? 18:32:51 -!- jaboja has joined. 18:39:04 -!- hppavilion[1] has quit (Ping timeout: 252 seconds). 18:41:35 -!- vanila has joined. 18:41:36 hello 18:51:51 -!- Zoroaster has quit (Ping timeout: 268 seconds). 18:52:03 -!- picobit has quit (Ping timeout: 240 seconds). 19:00:10 -!- gremlins has joined. 19:00:50 -!- Zoroaster has joined. 19:01:29 -!- Reece` has quit (Ping timeout: 244 seconds). 19:04:15 -!- shikhin_ has changed nick to shikhin. 19:04:42 -!- zzo38 has quit (Remote host closed the connection). 19:05:16 -!- centrinia has joined. 19:14:28 -!- Reece` has joined. 19:15:57 -!- gremlins has quit (Ping timeout: 244 seconds). 19:18:21 -!- hppavilion[1] has joined. 19:26:00 -!- Phantom_Hoover has joined. 19:26:15 -!- centrinia has quit (Ping timeout: 250 seconds). 19:31:29 -!- gremlins has joined. 19:32:29 -!- Reece` has quit (Ping timeout: 244 seconds). 19:37:22 Random thing of the day: Aalto University's electrical engineering building had an oscilloscope showing a fancy vector graphics counter towards first of May. (It's a big holiday-type thing in Finland, especially for students.) 19:37:55 -!- XorSwap has joined. 19:37:59 We have a screen in the CS building's foyer saying when the next buses are 19:38:04 It's quite useful 19:38:18 -!- XorSwap has quit (Read error: Connection reset by peer). 19:40:33 We have one of those too. 19:40:46 But it's just a boring LCD monitor. 19:40:58 An oscilloscope is much nicer. 19:41:20 if only 19:41:44 I'd share this video I took of it if I could figure out how to remove the audio by just fiddling with this phone the right way. 19:42:59 step 1: download an x86 VM 19:43:08 step 2: install arch linux & ffmpeg 19:43:18 step 3: ...(I think you can see where this is going) 19:43:57 Yeah, no. If that's the alternative, I'll just wait a few days until I'm at a computer. 19:45:04 fizzie: Well, you don't have to use a VM. If you have a box you can SSH into, then you can use SCP to copy the file there and do it. 19:45:10 Would the proper way to make a font format for Windows 19:45:28 Be to produce a DLL with a callable function renderString(char* s[]) 19:45:35 I've gotten the impression U-tube has an editor of some sort, which might work. 19:46:07 That returns an array of bitmaps representing each character in the string as a rasterized image in a pre-set font? 19:46:26 hppavilion[1]: that's the proper way to produce malware 19:46:39 here, have this "font" 19:46:57 * gamemanj gives hppavilion[1] a DLL which, upon loading, destroys the universes 19:47:03 (all the universes) 19:47:40 That's a lot of universes 19:47:53 or perhaps only one 19:47:56 also, the Microsoft-Approved Naming Scheme for functions seems to be FirstLetterOfEachWordIsACapitalLetter 19:48:08 https://drive.google.com/file/d/0B4J9OAzXNfZAcWdOdFY4eDFLT2s/view?usp=drivesdk is a still image, though the animation was kind of nice. 19:48:58 int-e: Oh 19:49:00 inaccessible. 19:49:11 Huh. 19:49:22 related: https://mobiforge.com/research-analysis/the-web-is-doom 19:49:28 I tapped "share link". 19:49:30 fizzie: not enabling javascript :P 19:49:37 Maybe that wasn't enough. 19:49:38 (Well, a tagged union of bitmaps OR grayscale OR RGBAmap) 19:49:45 int-e: What /is/ the proper way to do it? 19:51:41 int-e: There's a lower-level link-to-file thing that I believe doesn't involve javascript, but maybe that's not what happens by sharing from the Android app. :/ 19:52:06 truetype is already turing complete... (meta reference: https://yahoo-security.tumblr.com/post/123981052855/font-parsing-vulnerabilities ) 19:52:30 int-e: OK, and? 19:53:01 fizzie: actually one of the links that noscript helpfully extracts worked 19:53:39 hppavilion[1]: I'm not sure what problem with existing vector font formats you identified that needs to be solved. 19:53:57 int-e: There isn't a problem; it's just an attempt to understand computers 19:54:25 Good. It's probably the one I normally get from the command-line gdrive tool. 19:54:52 hppavilion[1]: there's value in not exposing turing-complete machines (and in the worst case, actual machine code) in data formats. 19:55:06 int-e: Huh? 19:55:12 hppavilion[1]: Other than the points int-e has raised, nobody sane uses DLLs to store resources. 19:55:14 int-e: I'm a bit confused 19:55:24 gamemanj: I'm not encoding the fonts as DLLs 19:55:49 gamemanj: I'm talking about how one makes a renderer /for/ a font 19:56:01 *font format 19:56:02 hppavilion[1]: Oh. That wasn't the question you described above, but OK. 19:56:10 gamemanj: I must have misspoken 19:56:20 hppavilion[1]: oh. "make a font format" is ambiguous. 19:56:24 int-e: Ah 19:56:27 int-e: I see the issue now 19:56:55 int-e: I'm making a thing that accepts a font in a format (or, well, the name) and a string and produces the bitmap for the string 19:57:19 int-e: Or, well, I might make such a thing 19:57:38 hppavilion[1]: anyway in that case... I suspect the OS wants to be able to cache glyphs. 19:57:46 int-e: True 19:58:16 int-e: How exactly do I go about it though? 19:58:36 Well, modern font rendering is complex 19:58:43 gamemanj: I imagine so 19:59:06 There's ligatures, subsitutions, RTL, kerning, spacing, all the different types of curves, 19:59:20 gamemanj: Yes, correct 19:59:21 and 0x10000 characters. Before we get into Emoji. 19:59:28 gamemanj: I know 19:59:43 I think you generally don't. Because if it's not part of the "system" (FSVO), every program wanting to take advantage of it would have to do so separately, and they wouldn't. 19:59:50 Which means it really depends what font format you're trying to render. 19:59:59 gamemanj: OK, and? 20:00:17 gamemanj: I'm inventing a simplistic format as a project. Let's say it's for school. 20:00:41 gamemanj: And let's pretend that said school assignment has 3+ years to be completed. 20:02:10 It depends on how much you want to cover. 20:02:20 gamemanj: Not much. Yet. 20:02:45 gamemanj: I want to build it slowly; not as a single monolithic program that has all features right off the bat 20:03:05 gamemanj: Nobody is going to ever /use/ this /seriously/, so it doesn't have to be spectacular 20:03:07 If you don't have any practical goals, you could as well implement your format as an extension to FreeType, which already supports a number of formats. 20:03:11 Unifont, for example, is a bitmap font that more-or-less handles the Unicode BMP, and can be easily rendered. 20:03:47 gamemanj: My font format is a rather primitive vector-based thing. 20:04:27 gamemanj: I just need to know what kind of file I use to make an easily-renderable font that can be used by things /other/ than applications which include the entirety of the renderer code packaged with them 20:04:55 ...Ok, now I'm confused 20:05:06 Are you asking about making a font or making a font renderer? 20:05:28 gamemanj: I don't care about the code I'm putting into the file, I just care what I need to do to make it usable by external programs (that support it in their code) 20:05:56 gamemanj: Once I know what kind of file I'm producing, I can find how to make it online 20:06:15 hppavilion[1]: I'd look at http://www.cairographics.org/manual/cairo-User-Fonts.html for inspiration 20:06:17 gamemanj: It boils down to: Is making the renderer a DLL the right way to go for windows? 20:06:26 hppavilion[1]: No. 20:06:33 gamemanj: OK, then what is? 20:06:36 Make a TTF. FontForge can make it, but Birdfont also works (but is a bit of a mess) 20:06:51 gamemanj: The goal is a custom format. 20:06:51 (I wouldn't expect it to solve all problems... but it's something smart people have worked on.) 20:07:12 gamemanj: The goal isn't to make a font. It's to make a format for encoding fonts. 20:07:17 gamemanj: This isn't a practical thing. 20:07:21 gamemanj: This is a fun thing. 20:07:50 So, basically, your question boils down to "If I am inventing my own format, and using it in multiple applications, should I put loading/rendering code into a DLL?" 20:07:50 (and I'd call it a "font renderer" instead of a "font format") 20:08:00 int-e: OK 20:08:05 gamemanj: Yes. 20:08:09 gamemanj: That is the question. 20:08:27 gamemanj: And if it shouldn't be a DLL, what should it be? 20:08:35 hppavilion[1]: Keep your font format (your data) separate from your font renderer (your code) and you'll be fine. 20:08:45 gamemanj: Yes, that's the goal. 20:09:03 gamemanj: The renderer is the demonstration of the format. 20:09:03 I like it when (academic) spam comes right to the point. Subject: Publish your papers 20:09:33 hppavilion[1]: The renderer should read the font data from a file - having the DLL be the font is a great way to stop you from ever having more than one font. 20:09:37