←2017-04-26 2017-04-27 2017-04-28→ ↑2017 ↑all
00:00:57 -!- boily has joined.
00:02:24 <boily> `wisdom
00:02:25 <HackEgo> isomorphism//Isomorphism is isomorphic to Phantom_Hoover up to isomorphism.
00:03:43 <shachaf> `w
00:03:44 <HackEgo> ​`learn//`learn creates a wisdom entry and tries to guess which word is the key. Syntax (case insensitive): `learn [a|an|the] <keyword>[s][punctuation] [...]
00:04:04 <shachaf> not enough deep learning
00:04:22 -!- Nistur has quit (Ping timeout: 260 seconds).
00:04:29 -!- Nistur has joined.
00:07:05 <boily> H E L L O O C H A F !
00:07:21 <shachaf> H E L L O 
00:07:35 <shachaf> `w
00:07:36 <HackEgo> monomorphism//A monomorphism is just an epimorphism in the opposite category.
00:07:42 <shachaf> `? epimorphism
00:07:43 <HackEgo> An epimorphism is just a monomorphism in the opposite category.
00:08:12 <shachaf> An isomorphism is just a split epimorphism which is also a monomorphism.
00:08:25 <shachaf> Or: An isomorphism is just a split monomorphism which is also a epimorphism?!
00:09:15 <boily> they're all isomorphic to each other, modulo a natural transformation.
00:12:43 <shachaf> What does that sentence mean?
00:13:23 <boily> it means that's all fungobbledigook to me.
00:14:30 -!- iconmaster has joined.
00:14:51 <boily> `relcome iconmaster
00:14:53 <HackEgo> iconmaster: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: <http://esolangs.org/>. (For the other kind of esoterica, try #esoteric on EFnet or DALnet.)
00:15:21 -!- iconmaster has quit (Client Quit).
00:32:19 <boily> `w
00:32:21 <HackEgo> finn//Finns are helpful, albeit grossly overpopulated (cf. 'Finland').
01:05:12 -!- Zarutian has joined.
01:18:30 <boily> Zarutellon.
01:24:02 -!- sleffy has quit (Ping timeout: 260 seconds).
01:34:48 -!- erkin has joined.
02:25:31 -!- boily has quit (Quit: EMULSIFIED CHICKEN).
02:40:07 -!- ybden has quit (Ping timeout: 240 seconds).
02:40:54 -!- ybden has joined.
03:01:32 -!- Akaibu has quit.
03:20:11 -!- augur has joined.
03:25:49 -!- erkin has quit (Quit: Ouch! Got SIGABRT, dying...).
03:34:49 -!- sleffy has joined.
03:43:41 -!- MDude has joined.
04:06:02 -!- Zarutian has quit (Quit: Zarutian).
04:11:40 -!- erkin has joined.
04:56:49 <Jafet> `? deep learning
04:56:50 <HackEgo> deep learning? ¯\(°​_o)/¯
05:13:36 <Jafet> `learn Deep learning applies software engineering principles to AI. A deep learning network has numerous layers and no one understands what any of them do.
05:13:40 <HackEgo> Learned 'deep': Deep learning applies software engineering principles to AI. A deep learning network has numerous layers and no one understands what any of them do.
05:13:45 <Jafet> `revert
05:13:47 <HackEgo> Done.
05:14:01 <Jafet> er, I expected it to deep learn that
05:14:14 <Jafet> `le//rn deep learning//Deep learning applies software engineering principles to AI. A deep learning network has numerous layers and no one understands what any of them do.
05:14:16 <HackEgo> Learned 'deep learning': Deep learning applies software engineering principles to AI. A deep learning network has numerous layers and no one understands what any of them do.
05:20:08 -!- adu has joined.
06:18:34 -!- Jafet has quit (Ping timeout: 252 seconds).
06:23:04 -!- sleffy has quit (Ping timeout: 240 seconds).
06:24:19 -!- adu has quit (Quit: adu).
06:37:59 -!- Jafet has joined.
06:52:44 -!- FreeFull has quit.
07:38:47 -!- pikhq has quit (Ping timeout: 260 seconds).
07:55:49 -!- pikhq has joined.
08:25:31 -!- augur has quit (Remote host closed the connection).
08:26:06 -!- FreeFull has joined.
08:55:07 -!- augur has joined.
09:20:50 -!- oerjan has joined.
09:36:46 -!- augur has quit (Remote host closed the connection).
09:49:29 -!- AnotherTest has joined.
10:06:57 -!- erkin has quit (Quit: Ouch! Got SIGABRT, dying...).
10:13:59 -!- AnotherTest has quit (Ping timeout: 240 seconds).
10:15:51 -!- AnotherTest has joined.
10:36:01 <oerjan> fizzie: wiki bridge is down
10:39:48 <fizzie> So it is. I wonder if I could set up some automation for it.
10:40:02 <fizzie> Looks like this time it was a reboot: 05:39:40 up 4 days, 9:24, 1 user, load average: 0.25, 0.22, 0.09
10:40:13 <oerjan> seems about time, it's not seemingly about to stop happening.
10:42:26 <fizzie> Maybe I'll try to write a systemd service file for it, I think that should make it start automagically.
10:42:31 <fizzie> But not now, have to get to work.
11:03:00 -!- LKoen has joined.
11:06:53 -!- Oxid_it_89 has joined.
11:06:59 -!- gsora_ has changed nick to gsora.
11:34:09 -!- boily has joined.
11:37:20 -!- augur has joined.
11:39:43 <boily> `w
11:39:45 <HackEgo> mirth//Mirths are juvenile moths. They giggle a lot.
11:40:39 <oerjan> hily
11:41:40 -!- augur has quit (Ping timeout: 258 seconds).
11:43:57 <boily> hellørjan.
11:46:08 <boily> `w
11:46:09 <HackEgo> welcome.bork//Velcume-a tu zee interneshunel hoob fur isutereec prugremmeeng lungooege-a deseegn und depluyment! Fur mure-a inffurmeshun, check oooot oooor veeki: <http://isulungs.oorg/>. (Fur zee oozeer keend ooff isutereeca, try #isutereec oon IFnet oor DELnet.)
11:46:25 <boily> bø̈rk.
12:09:54 -!- AnotherTest has quit (Ping timeout: 260 seconds).
12:25:17 -!- LKoen has quit (Remote host closed the connection).
12:26:02 -!- LKoen has joined.
12:30:20 -!- oerjan has quit (Quit: Later).
12:30:44 -!- boily has quit (Quit: REINFORCED CHICKEN).
12:30:54 -!- Oxid_it_89 has quit (Ping timeout: 260 seconds).
12:42:16 -!- LKoen has quit (Quit: “It’s only logical. First you learn to talk, then you learn to think. Too bad it’s not the other way round.”).
14:31:39 -!- `^_^v has joined.
15:34:17 -!- orby has quit (Quit: Page closed).
15:34:45 -!- orby has joined.
15:47:39 -!- Phantom_Hoover has joined.
15:47:39 -!- Phantom_Hoover has quit (Changing host).
15:47:39 -!- Phantom_Hoover has joined.
16:11:24 <HackEgo> [wiki] [[Brainfuck implementations]] https://esolangs.org/w/index.php?diff=51826&oldid=51817 * AsuMagic * (-11) update ashbf's description
16:20:50 -!- LKoen has joined.
16:32:38 -!- augur has joined.
16:43:31 -!- AnotherTest has joined.
17:14:53 -!- sleffy has joined.
17:35:04 <Taneb> O(n^2) : quadratic :: O(sqrt(n)) : ???
17:38:10 <izabera> squarerootatic
17:38:19 <Taneb> Any advances on that
17:38:46 -!- LKoen has quit (Remote host closed the connection).
17:57:07 -!- sleffy has quit (Ping timeout: 240 seconds).
18:07:02 <int-e> . o O ( semi-linear *runs* )
18:07:44 <shachaf> Did you see my post about "nines" of reliability?
18:08:05 <shachaf> They're a logarithmic unit and therefore they should be measured in decibels.
18:08:07 <int-e> Taneb: borrowing from geometry, circumferential (in terms of area) ;-)
18:08:27 <shachaf> -10 dB of unreliability = one "nine" of reliability
18:13:27 <int-e> this seems completely unrelated to blocked clauses
18:16:30 <shachaf> Blocked clauses?
18:17:30 * int-e is looking at SAT solving.
18:18:55 <int-e> And blocked clause elimination is a simplification technique for a set of clauses. And it's also totally unrelated to what you said, just something that's currently on my mind.
18:20:04 <shachaf> I see. I don't know that much about SAT solving techniques.
18:22:56 -!- DHeadshot has joined.
18:26:51 -!- DHeadshot_ has joined.
18:27:24 -!- hppavilion[1] has joined.
18:27:41 -!- DHeadshot has quit (Ping timeout: 268 seconds).
18:32:59 -!- oerjan has joined.
18:37:14 <oerjan> <Taneb> O(n^2) : quadratic :: O(sqrt(n)) : ??? <-- radical hth
18:39:14 <Taneb> oerjan, thanks
18:39:29 <Taneb> Doesn't sound right, though, so I'm gonna stick with "proportional to the square root"
18:40:14 <oerjan> it's etymologically right for "root", but adding the square is a bit verbose especially since latin is bad at compounds.
18:42:36 <oerjan> `le/rn SAT solving techniques//There are many SAT solving techniques, but none is completely satisfying.
18:42:47 <oerjan> hm
18:42:50 <HackEgo> Learned 'sat solving techniques': There are many SAT solving techniques, but none is completely satisfying.
18:43:25 <oerjan> `` mv wisdom/'sat solving technique'{s,}
18:43:29 <HackEgo> No output.
18:43:35 <oerjan> `cat bin/whoops
18:43:36 <HackEgo> OLD="wisdom/$1"; [ -z "$1" ] && OLD="$(lastfiles)"; NEW="${OLD}s"; if [ -f "$NEW" ]; then echo "«${NEW}» already exists"; exit 1; fi; mv "$OLD" "$NEW" && echo "«${OLD}» -> «${NEW}»"
18:45:52 <oerjan> `before
18:46:07 <HackEgo> wisdom/sat solving technique//There are many SAT solving techniques, but none is completely satisfying. \ wisdom/sat solving techniques//There are many SAT solving techniques, but none is completely satisfying.
18:46:18 -!- Zarutian has joined.
18:46:24 <oerjan> hm that is not right.
18:46:44 <Taneb> none are
18:46:47 <oerjan> `cat bin/before
18:46:47 <HackEgo> lastfiles "$@" | while read f; do echo -n "$f//"; hg cat -r "$(hg log --removed --template '{rev}\n' "$f" | tail -n+2 | head -n1)" "$f"; done
18:47:02 <oerjan> Taneb: i thought none is singular
18:47:26 <oerjan> anyway that's not what wasn't right, but `before
18:47:58 <int-e> oerjan: ITYM satisfactory, hth.
18:48:04 <oerjan> oh.
18:48:27 <oerjan> `le/rn SAT solving technique//There are many SAT solving techniques, but none is completely satisfactory.
18:48:29 <HackEgo> Relearned 'sat solving technique': There are many SAT solving techniques, but none is completely satisfactory.
18:49:14 <oerjan> my short term memory gets worse and worse. i truly intended to change that to "are" 20 seconds before i pushed enter.
18:49:32 <Taneb> "It is sometimes held that none can only take a singular verb, never a plural verb: none of them is coming tonight rather than none of them are coming tonight. There is little justification, historical or grammatical, for this view. None is descended from Old English nān meaning ‘not one’ and has been used for around a thousand years with both a singular and a plural verb, depending on the context and the emphasis needed"
18:49:42 <Taneb> I'm used to it taking the plural
18:50:10 <oerjan> `slwd SAT solving technique//s, is, are,
18:50:11 <HackEgo> Roswbud!
18:50:30 <oerjan> `slwd sat solving technique//s, is, are,
18:50:32 <HackEgo> sat solving technique//There are many SAT solving techniques, but none are completely satisfactory.
18:51:16 <oerjan> `before
18:51:25 <HackEgo> wisdom/sat solving technique//There are many SAT solving techniques, but none is completely satisfactory.
18:52:48 <Vorpal> hi
18:52:55 <oerjan> hirpal
18:53:24 <Taneb> oerjan, I admit my English grammar is probably influenced by a combination of northern English and Australian dialects so may be odd in places
18:55:38 <oerjan> shocking
18:55:39 <\oren\> I need a while(){}else{} structure, blah...
18:56:38 <\oren\> if(thing){while(the same thing){}}else{} is ugly
18:57:00 <Taneb> \oren\, just use gotos
18:57:23 <Taneb> ;)
18:57:36 <Vorpal> \oren\: I seem to remember seeing some language that had a for ... else construct at least
18:58:06 <Vorpal> Can't remember which one it was
18:58:23 <Vorpal> ruby maybe? Some scripting language that I don't personally no, but I just read about anyway
18:58:30 <Vorpal> know*
18:58:44 <Taneb> Vorpal, Python does, I think
18:58:57 <Vorpal> Really? I know python. I thought at least
18:59:24 <int-e> Hmm, but you're treating the failed test in two different ways so it's awkward to implement the for...else thing; it'll either duplicate the test or have an extra boolean variable.
18:59:58 <Vorpal> int-e: well the python for loop is for variable in collection:
19:00:09 <Vorpal> so that doesn't really apply
19:00:22 <Vorpal> I guess it would be if the loop count is zero simply
19:02:22 <Vorpal> int-e: https://gist.github.com/anonymous/4cc8c1f2b9c8c9e88599b2f6de02fc9b
19:02:37 <Vorpal> It apparently works
19:03:41 <Vorpal> \oren\: so the answer is: use python
19:03:42 <Vorpal> Maybe
19:03:56 <Vorpal> It would also be reasonably simple in funge-98
19:09:38 -!- augur has quit (Remote host closed the connection).
19:09:58 <Vorpal> you could do it via self modifying code
19:10:12 -!- augur has joined.
19:10:26 <Vorpal> assuming you don't need concurrency
19:12:28 <\oren\> https://www.youtube.com/watch?v=o5muHHMWIMc
19:15:02 -!- augur has quit (Ping timeout: 260 seconds).
19:21:55 <oerjan> i wonder if the forth control flow primitives can do this, some of them apparently work in unusual combinations.
19:27:34 <oerjan> oh hm the problem is that the branch works differently on the first iteration.
19:28:08 <oerjan> so i guess it really _does_ need a flag internally.
19:30:57 -!- augur has joined.
19:31:22 -!- hppavilion[1] has quit (Ping timeout: 260 seconds).
19:35:13 -!- augur has quit (Ping timeout: 246 seconds).
19:37:55 -!- b_jonas_ has joined.
19:38:51 -!- ineiros_ has joined.
19:39:00 -!- b_jonas has quit (Write error: Broken pipe).
19:39:01 -!- ineiros has quit (Write error: Broken pipe).
19:56:34 -!- Zarutian has quit (Ping timeout: 260 seconds).
19:58:27 -!- Phantom__Hoover has joined.
20:06:20 -!- DHeadshot_ has quit (Ping timeout: 240 seconds).
20:08:36 -!- DHeadshot has joined.
20:13:20 -!- Phantom__Hoover has quit (Quit: Leaving).
20:33:10 -!- LKoen has joined.
20:34:23 <int-e> izabera: Echo Look is perhaps even more creepy than Faceapp
20:34:51 <izabera> `echo look
20:34:51 <HackEgo> look
20:34:54 <izabera> creepy
20:36:13 <int-e> `? amazon
20:36:14 <HackEgo> amazon? ¯\(°​_o)/¯
20:45:35 <int-e> . o O ( who makes furniture sturdy enough to do http://questionablecontent.net/view.php?comic=3469 this? )
21:01:30 <izabera> ikea?
21:02:53 <\oren\> "I know what you're thinking. Did I hit space 6 times or only 5? well to tell you the truth, in all this typing, I've kinda lost track myself. but being as this is a bog standard keyboard, the crappiest keyboard in the world, and would not register key presses cleanly, you've got to ask yourself one question: "Do I feel lucky?" well, do ya, punk?" - clint eastwood on python's indentation rules
21:03:31 -!- augur has joined.
21:03:55 <shachaf> Do you indent Python differently from other languages?
21:06:17 <\oren\> you're forced to
21:06:53 -!- `^_^v has quit (Quit: This computer has gone to sleep).
21:07:50 <izabera> no?
21:09:02 <shachaf> I'm not forced to indent Python differently from the way I indent other languages.
21:09:11 <shachaf> Well, I guess I'm forced to indent Haskell and Python differently.
21:14:47 -!- `^_^v has joined.
21:14:56 <\oren\> well for starters you're forced to indent things at all
21:15:39 <shachaf> Yes, but that's not what I said.
21:16:32 <\oren\> I don't indent anything in bash for example
21:17:50 <\oren\> in python you are forced to indent the body of a class
21:18:09 <shachaf> In bash you can't indent the body of a class.
21:18:24 <\oren\> right but in C++ I don't do that
21:18:52 <izabera> \oren\: humans indent code, usually
21:21:12 <\oren\> class Fred {
21:21:33 <\oren\> int somethingFredDoes(){
21:21:43 <\oren\> //code here
21:21:46 <\oren\> }
21:21:47 <\oren\> }
21:23:27 <shachaf> You should just run your C++ code through clang-format.
21:23:43 <\oren\> eww no
21:25:50 <\oren\> too many spaces
21:26:18 <int-e> what an odd reason for rejecting a programming language
21:28:03 <FireFly> \oren\: use two-space indentation, and the code in your function bodies ends up being indented the same as what you currently have :p
21:29:03 <\oren\> it also is problematic with putting too many spaces inside the code
21:29:42 <izabera> yeah it increases the file size and fills your hard drive
21:29:45 <int-e> hmm, I have (setq python-indent 3)... wonder if I still like that
21:30:00 <fizzie> What an odd number to use.
21:30:10 <int-e> it's a prime choice
21:31:13 <\oren\> and some idiots also do crap like
21:31:15 <\oren\> int
21:31:18 <\oren\> main()
21:31:20 <\oren\> {
21:31:22 <\oren\> WTF
21:31:29 <izabera> it's easier to grep for
21:31:31 <fizzie> The canonical justification is, you can find functions...
21:31:32 <fizzie> Right.
21:31:37 <int-e> it's so you can grep for a definition of main by ^main, I think
21:31:44 <int-e> ...
21:31:52 <int-e> `? redundancy
21:31:54 <HackEgo> Since redundancy exists, it's redundant for Taneb to invent it.
21:32:14 <int-e> @quote fugue
21:32:14 <lambdabot> monochrom says: Welcome to #haskell, where your questions are answered in contrapuntal fugues.
21:32:23 <\oren\> the definition of main is at /main.*{$/
21:32:35 <int-e> no it's not
21:32:42 <fizzie> That's also the definition of remainder.
21:32:43 <int-e> the { belongs on a seaparate line ;)
21:32:48 <\oren\> BAH
21:33:05 <fizzie> Admittedly ^main finds also the function mainstay.
21:33:12 <int-e> (but only for function bodies, not for if-then-else or loops)
21:33:23 <fizzie> This grepping for source code should be obsolete anyway.
21:33:49 <int-e> int main()
21:33:50 <int-e> {
21:33:55 <shachaf> fizzie: People should just use codesearch, right?
21:33:59 <int-e> ...; // is what I'd do
21:34:31 <shachaf> int
21:34:36 <shachaf> main(argc, argv)
21:34:39 <shachaf> char **argv;
21:34:40 <shachaf> {
21:34:43 <izabera> ew
21:34:45 <\oren\> char*somefunction(char*arg,int arg2){
21:34:46 <izabera> shachaf: plz no
21:35:02 <\oren\> spaces only belong in between letters
21:35:12 <int-e> shachaf: at least indent the char **argv properly, if you must use K&R C.
21:35:18 <\oren\> char**arv
21:35:36 <\oren\> and indentation is tabs only
21:36:09 <int-e> . o O ( I wonder who that \oren\ guy is, has he been here before? )
21:36:32 <fizzie> shachaf: I was looking at Doxygen the other day, I've remembered it being kinda crappy, but it wasn't that bad anymore.
21:36:41 <fizzie> shachaf: They've embraced Markdown and all that.
21:36:54 <shachaf> What's the point of using a documentation generation tool?
21:36:59 <shachaf> You can just read the source code.
21:36:59 <\oren\> why isn't Doxygen an automatic doxxing tool
21:37:04 <fizzie> shachaf: You could even imagine putting .md files sprinkled throughout the code.
21:38:07 <fizzie> Anyway, even disregarding the special documentation comments, it generates a slightly codesearch-like thing if you ask it.
21:38:20 <fizzie> With crossrefs and a file tree and that sort of thing.
21:38:31 <shachaf> For reading the actual code?
21:38:34 <fizzie> Yes.
21:38:38 <shachaf> Sounds good.
21:38:52 <shachaf> Does it search millions of lines of code with regular expressions?
21:38:56 <fizzie> You just say SOURCE_BROWSER = YES and VERBATIM_HEADERS = YES.
21:38:59 <fizzie> It doesn't do that.
21:39:17 <\oren\> so it generates some sort of database
21:39:21 <shachaf> Does it have all sorts of fancy layers?
21:39:28 <fizzie> It doesn't really do that either.
21:39:46 <fizzie> It does have integration with something called htags, which seemed a little grok-like.
21:39:59 <fizzie> GNU GLOBAL, it's called.
21:40:03 <fizzie> I didn't look too closely.
21:40:10 <shachaf> itym kythe hth
21:41:19 -!- augur has quit (Remote host closed the connection).
21:45:12 <\oren\> hey I did figure out something worse
21:45:14 <\oren\> if(x==1)runstuff();else(runotherstuff());
21:45:33 <\oren\> you can avoid spaces in many places by using ()
21:46:40 <\oren\> int(x)=1;
21:47:24 -!- hppavilion[1] has joined.
21:48:19 <\oren\> int(main)(int(argc),char**argv){
21:48:32 <\oren\> compiles with no warnings on -Wall -Wextra
21:48:50 <shachaf> You know that C++ parsing ambiguity.
21:49:44 <shachaf> https://en.wikipedia.org/wiki/Most_vexing_parse
21:51:58 <\oren\> hmm, https://arin.ga/E3Z9Ll
21:53:17 <shachaf> aooooo.ga
21:54:22 <\oren\> shachaf: that's not a real url is it
21:55:42 -!- wob_jonas has joined.
21:55:45 <shachaf> It's not a URL at all.
21:58:22 <\oren\> well, maybe not *technically* but
21:58:32 <\oren\> it's a domain name
21:58:38 -!- LKoen has quit (Remote host closed the connection).
21:59:01 <\oren\> wob_jonas: hellwob!
21:59:06 <wob_jonas> hi
21:59:22 <\oren\> https://arin.ga/E3Z9Ll what do you think of this abomination
22:00:24 <wob_jonas> um, what about it? it looks like a simple hello world in C
22:00:31 <\oren\> I was thinking about how to avoid using any spaces
22:00:59 <\oren\> and I realized that int(main)(int(argc),char**argv{) is valid
22:01:04 <wob_jonas> ok... any reason for wanting to avoid them?
22:01:10 <Taneb> space bar broken
22:01:11 <\oren\> for fun
22:01:21 <wob_jonas> ok, have fun then
22:01:21 <\oren\> Taneb: yeah let's go with that
22:01:30 <wob_jonas> Taneb: nope. I'd replace keyboard immediately
22:01:44 <Taneb> wob_jonas, I do have two backup keyboards, but they are in York
22:01:47 <Taneb> `? Taneb
22:01:48 <HackEgo> Taneb is not elliott, no matter whom you ask. He also isn't a rabbi although has pretended in the past. He has at least two backup keyboards with dodgy SHIFT KEys, cube root of nine genders, one of which is a Czech woman, and above average, not too voluminous, but calm eyebrows. He sometimes invents without noticing it (see: tanebventions).
22:01:53 <\oren\> int(x)=0;
22:02:26 <wob_jonas> I actually bought a temporary keyboard when my previous one broke and the new one I'd ordered from abroad hadn't arrived yet. Used the temp for like two week. I absolutely refuse to use a keyboard with a key broken, or a phone with the screen broken except for emergencies,
22:02:41 -!- `^_^v has quit (Quit: This computer has gone to sleep).
22:03:08 <wob_jonas> and I look down on the lots of people I see in the city that use phones with a broken display, and pretend that they're people who buy expensive phones without being able to afford to replace it when they break.
22:03:30 <\oren\> if(x==1)runstuff();else(runotherstuff());
22:03:55 <wob_jonas> I have the temp keyboard here as a backup, but even if I didn't, I'm in Budapest, so I can just walk in a shop pretty much any time and buy a new temp keyboard.
22:04:13 <Taneb> Anyway, I'm off now
22:04:16 <Taneb> Goodnight!
22:04:37 <\oren\> hmm, const(int(x))=1; doesn't seem to work
22:04:59 <wob_jonas> There can be good eso-reasons to avoid using a character (like, you want to embed the code into some other text), but broken keyboard isn't a good one.
22:05:56 <wob_jonas> \oren\: can you afford one space? #define T(x)x and then whenever you'd need a space between foo bar write T(foo)bar instead
22:06:06 <wob_jonas> if not, try to find a macro that's already like that in a library
22:06:19 <wob_jonas> or just use newlines instead of spaces everywhere
22:06:21 <wob_jonas> const
22:06:24 <wob_jonas> int
22:06:32 <wob_jonas> x=1; is ugly, but so what
22:06:46 <wob_jonas> oh, and there's also always foo/*c*/bar
22:06:54 <wob_jonas> a comment serves as whitespace
22:07:00 <wob_jonas> that even works on the define line
22:07:09 <wob_jonas> #define/*c*/T(x)x
22:07:16 <wob_jonas> T(const)int(x)
22:07:27 -!- LKoen has joined.
22:07:42 <wob_jonas> can be chained: T(const)T(unsigned)T(long)T(long)T(int)x
22:08:08 -!- LKoen has quit (Client Quit).
22:08:31 <\oren\> brilliant
22:08:37 <\oren\> that works great
22:08:42 <wob_jonas> if you need it often, define a macro like #define U(x,y,z)x/*c*/y/*c*/z and then you can just U(const,unsigned,long)U(long,int,x);
22:08:54 <wob_jonas> still uglier than just using space, but whatever
22:09:34 <wob_jonas> all this works only in an ANSI compiler, can break in the very ancient ones where comments are handled differently
22:10:19 <hppavilion[1]> Is there a programming language based on grammar?
22:10:37 <wob_jonas> but these days, everyone uses ANSI compilers, it's more like old code that hasn't been updated and requires old practices that's the problem
22:10:55 <wob_jonas> and even that's becoming rarer and rarer
22:14:30 * hppavilion[1] . o O ( I wonder if I could set up a home proxy )
22:14:34 <oerjan> heh PPCG has actual advertisements for Esoteric IDE
22:14:42 <hppavilion[1]> oerjan: PPCG?
22:14:53 <hppavilion[1]> Ah, Programming Puzzles and Code Golf (on Stack Exchange)
22:14:54 <hppavilion[1]> I remember
22:14:57 <\oren\> hppavilion[1]: based on grammer?
22:15:05 <\oren\> what kind of grammar?
22:15:45 <hppavilion[1]> \oren\: The language would be based on parsing strings, performing actions incidentally
22:16:45 <hppavilion[1]> Like, a truth machine could be encoded as something like "your initial string is 's' and your initial rule is 'main'"
22:19:29 <hppavilion[1]> One option is that main parses one 's' token and incidentally (1) enqueues a new string [when the program finishes parsing, it parses the next string if there is one and ends if there isn't] obtains input (when input is obtained, it's appended to string at the back of the queue)
22:20:00 <hppavilion[1]> Main could alternatively parse one '0' token, which incidentally outputs '0'
22:20:32 <hppavilion[1]> Or it could parse one 'ONE' block
22:21:43 <hppavilion[1]> A ONE block parses one '1' token followed by a ONE block and incidentally adds a '1' token and outputs '1'
22:21:45 <hppavilion[1]> Or something like that
22:25:17 <\oren\> https://www.youtube.com/watch?v=hG6ZNNWLb6U
22:31:38 -!- augur has joined.
22:33:14 -!- AnotherTest has quit (Quit: ZNC - http://znc.in).
22:33:22 -!- DHeadshot has quit (Ping timeout: 260 seconds).
22:36:03 <hppavilion[1]> Hm...
22:37:09 <hppavilion[1]> If we could go back and do it all again; if we wouldn't need to voluntarily abandon decades of permanently-ingrained standards
22:37:21 <hppavilion[1]> I wonder how we could have made the internet work better
22:38:36 -!- wob_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
22:39:07 <hppavilion[1]> One of the things I think would have to be done: change our rich text design
22:39:17 <hppavilion[1]> HTML, in the state it's in, is pretty terrible
22:40:11 <hppavilion[1]> But if you strip away some of the ugly parts- make a more elegant design for the same thing, examine only the pure, idealized syntax of the format; ignore the flesh, look only at the skeleton
22:40:18 <hppavilion[1]> ...it's still pretty terrible.
22:40:27 <hppavilion[1]> Scrap it entirely, replace it with something half-decent
22:41:37 <int-e> . o O ( "This sentence is false or paradoxical." )
22:42:39 * hppavilion[1] . o O ( Hm, "pure, idealized syntax" was supposed to mean "look at the scheme for the abstract syntax tree" but came out more like "look at XHTML" )
22:43:18 <hppavilion[1]> int-e: Does formal logic provide the capacity to discuss paradoxes?
22:43:28 <int-e> well, foremost, take out all the 5 from HTML5.
22:44:10 <hppavilion[1]> int-e: Also all the HTML
22:44:25 <hppavilion[1]> XML for markup seems like a bad plan
22:44:32 <int-e> I don't mind the HTML at all
22:44:49 <int-e> (which is not XML)
22:44:58 <hppavilion[1]> int-e: I hate it. The most annoying part is the abuse of properties on tags
22:45:13 <hppavilion[1]> int-e: Yeah, but it's basically just XML with more freedom to make ugly stuff
22:45:27 <int-e> that's historically backwards though
22:46:00 <int-e> SGML existed before XML. HTML is still a subset of SGML, AFAIK.
22:46:05 <hppavilion[1]> int-e: Wait, I seem to remember that XML is a subset* of SGML and HTML is an abuse of XML
22:46:33 <hppavilion[1]> int-e: But it's still pretty XMLy
22:47:11 -!- augur has quit (Remote host closed the connection).
22:47:15 <hppavilion[1]> I think one of the standard cases I like to point to: <img title="$title" src="$source" [...]/>
22:49:13 <pikhq> Modern HTML is firmly not a subset of SGML.
22:49:24 <pikhq> The HTML5 syntax is similar to SGML, but not SGML.
22:49:43 <pikhq> In large part because no web browsers parsed HTML *as* SGML.
22:50:10 <int-e> pikhq: I wasn't really talking about HTML5.
22:50:26 <pikhq> Pre-HTML5, the spec and the reality were two different subjects entirely.
22:50:47 <pikhq> The spec was, I agree, SGML.
22:51:34 <fizzie> I seem to recall one of the text-oriented browsers was pretty good at those bits of SGML-derived features that looked particularly non-HTMLy.
22:51:49 <int-e> I've heard this rumor that lynx was the only browser that implemented an SGML parser.
22:52:11 <int-e> (For HTML)
22:52:30 <fizzie> I think I was thinking of the SHORTTAG support.
22:52:34 <pikhq> In my admittedly-brief testing, Lynx seemed to be the only browser that implemented an SGML parser.
22:52:50 <pikhq> In that it supported things like SHORTTAGs.
22:53:33 <fizzie> There's an Appendix B.3.3 "SGML features with limited support" in the HTML 4.0 spec, which mentions some of these things.
22:54:05 <pikhq> I'd be unsurprised to discover there's a couple of other obscure browsers that also did it, though.
22:56:22 <hppavilion[1]> Oh, I have a sentence unfinished
22:56:38 <hppavilion[1]> And the text in $title has absolutely nothing to do with the rest of the document; you can't even put entities in there iirc
22:57:39 -!- boily has joined.
22:57:41 <boily> `w
22:57:43 <HackEgo> icfp//I see functorial people.
22:57:49 <oerjan> bohily
22:57:49 <hppavilion[1]> But I think the other major issue with modern markup is that the markup is inextricably baked into the content
22:57:56 <boily> hellørjan!
22:58:07 <boily> hppavellon[一].
22:58:23 <hppavilion[1]> If I want to get the text from a page, I have to get it with everything INCLUDING the markup, then I have to extract the text itself.
22:58:50 -!- augur has joined.
23:00:03 <int-e> hppavilion[1]: I think what happens with that sentence is that "paradoxical" is inherently a meta concept; the term "paradoxical" that appears inside the sentence will not have the same meaning as the term "paradoxical" that one uses when talking about it. For formal logic, note that truth (of a sentence given by its Gödel number) is not expressible in Peano arithmetic, so that will block...
23:00:09 <int-e> ...Gödel-like translations of it.
23:01:01 <hppavilion[1]> And since there's no way to tell what's part of the page's content and what's just there as boilerplate without using heuristics, layout detection, and even (*shiver*) site-specific extractors, you then have to get the things like "Wikipedia, the Free Encyclopedia" and "Read in other languages" and so forth and can't just extract the body
23:01:28 <int-e> `cwlprits icfp
23:01:37 <HackEgo> oerjän
23:01:45 <int-e> IC.
23:01:59 <pikhq> hppavilion[1]: And your best "general" bet is to just use HTML5's semantic tags for that, and give up on things that don't use it (which is most things)
23:02:58 <pikhq> (tags like <article>, which if correctly used tell you what's the "content" of the actual page.)
23:04:10 <int-e> . o O ( <article>The</article> sky is blue. )
23:04:16 <pikhq> LOL.
23:04:26 <\oren\> the interesting thing though is that although renting in tokyo is roughly 1/3 the price of the same apartment in Toronto, it's only cheaper if you look it up on a japanese web site
23:04:57 <pikhq> That... sounds like typical Japanese racism.
23:05:34 <int-e> or successful capitalism
23:05:39 <\oren\> nah, I think the cheap ones just don't bother
23:05:41 <shachaf> What is the price in Toronto?
23:06:02 <\oren\> a bachelor pad in toronto, even in a shit location, is >1000 CAD
23:06:06 -!- hppavilion[1] has quit (Ping timeout: 260 seconds).
23:06:11 <shachaf> What's a bachelor pad?
23:06:21 <\oren\> one bedroom, one bathroom
23:06:30 <\oren\> sleeps 1
23:06:47 <\oren\> smallest possible option
23:06:47 <shachaf> The bedroom is a separate room? Or is it a "studio"?
23:06:55 <boily> he\\oren\. bletch!
23:07:05 <shachaf> Anyway 1000CAD sounds cheap around here.
23:07:16 <\oren\> oh, heres one fro 950
23:07:41 <boily> that's horribly expensive! rent here is CAD 610/mo for a 3½.
23:07:44 <\oren\> ...and it isn't even really in toronto, it's up in north york
23:08:00 <pikhq> That'd be pricy *here*, but I'm also in a saner real estate market than Toronto or SF.
23:08:07 <\oren\> downtown toronto is like 3000
23:08:45 <pikhq> shachaf's in an area where to get to 1000 USD you'd need to be, what, 100 miles out?
23:08:54 <shachaf> 100 miles? No.
23:09:01 <pikhq> I may exaggerate a bit.
23:10:10 <shachaf> I pay <$1500/month for a 1br apartment ~700m from BART.
23:10:11 <pikhq> Oh, right, East Bay. Bit easier getting to a saner real estate market over there. There's rather a lot less bay in the way.
23:10:34 <pikhq> Also, higher distance from SV.
23:10:36 <\oren\> anyway, meanwhile I'm looking at apartments in the centre of tokyo, 3 minutes from train stations, 3.3万円
23:10:45 <shachaf> (Which is a pretty good price for the area. Mildly scow apartment, perhaps, but the location is good.)
23:11:30 -!- rdococ has quit (Ping timeout: 246 seconds).
23:11:35 <\oren\> which is roughly 330 CAD a month. so WTF
23:12:25 <\oren\> if I have to move out of my parents's house I'mma ask for a transfer to tokyo
23:12:26 <pikhq> From what I recall, Tokyo is rather remarkable for having managed to be a massive, dense city that *doesn't* refuse to build enough to meet demand.
23:13:09 <shachaf> How much is that in lakh INR?
23:13:37 -!- Lymia has quit (Ping timeout: 255 seconds).
23:13:40 <\oren\> pikhq: it's cheaper than both SF and Toronto, *and* you don't need to own a car
23:13:41 <shachaf> Didn't Tokyo build more housing one year recently than all of England?
23:13:56 <pikhq> Possibly.
23:14:08 <pikhq> \oren\: Of course, that's a *low bar*. :)
23:14:18 -!- augur has quit (Remote host closed the connection).
23:14:18 -!- izabera has quit (Ping timeout: 246 seconds).
23:14:18 -!- kmc has quit (Ping timeout: 246 seconds).
23:14:33 <\oren\> I don't ever plan on owning a car
23:14:46 <shachaf> imo why not make more money instead of moving somewhere cheaper hth
23:15:01 <\oren\> why not both
23:15:21 -!- kmc has joined.
23:15:35 <\oren\> little_girl_both.gif
23:16:44 -!- Lymia has joined.
23:16:50 -!- sebbu2 has joined.
23:16:51 -!- izabera has joined.
23:17:14 -!- sebbu has quit (Read error: Connection reset by peer).
23:17:22 -!- rdococ has joined.
23:17:46 <\oren\> http://knowyourmeme.com/memes/why-not-both-why-dont-we-have-both
23:18:50 -!- Hoolootwo has quit (Ping timeout: 240 seconds).
23:18:50 -!- ocharles has quit (Ping timeout: 240 seconds).
23:20:22 <boily> fungot: litagano motscoud.
23:20:22 <fungot> boily: in different bits. instructions may interpret the bits however they please. it's a true pain to try to become so knowledgable about unicode?
23:20:50 -!- Hoolootwo has joined.
23:20:56 <shachaf> pikhq: east bay? more like best bay
23:22:27 <shachaf> \oren\: Why not both, indeed?
23:24:03 <shachaf> But what if you can only do one or the other?
23:28:19 -!- ocharles has joined.
23:28:31 <\oren\> then go for cheap
23:28:53 <shachaf> Why?
23:29:11 <shachaf> What if income and expenses are linearly related?
23:29:19 <shachaf> Then you should go for expensive.
23:31:08 <\oren\> I suppose, depending on the other variables such as convenience, safety etc.
23:35:04 <\oren\> I mena if you have a 1 hour commute that is effectively dividing your salary by 1.25
23:35:28 <\oren\> if it's two hours then 1.5
23:35:33 <\oren\> etc.
23:36:04 <\oren\> if you have to own a car, then that subtracts as well, plus the inconvenience of having to learn to drive a car
23:36:14 <shachaf> Unless you work during your commute, I guess.
23:36:25 <\oren\> right
23:36:37 <shachaf> I don't know that 1.25 is the right factor, but it's certainly something to consider.
23:37:28 <\oren\> well, one hour each way is 2 hours+ 8 hour workday = 10 divided by 8 = 1.25 right?
23:38:08 -!- LKoen has joined.
23:38:17 <shachaf> If you want to measure your workday as 8 hours.
23:38:53 <\oren\> wtf is apache spark
23:40:29 <\oren\> and why are they sending me spam
23:40:58 <\oren\> I'm not a DBA
23:45:59 -!- sleffy has joined.
23:51:30 <\oren\> align me with a technical resource? WTF DOES THAT MEAN
23:53:07 <tromp> willing
23:53:12 <tromp> ignore
23:53:21 <\oren\> is thise person trying to say "give me the url of a manual"
23:53:28 <boily> `? \oren\
23:53:29 <HackEgo> ​\oren\ is an attempt to improve upon oren. The only thing it actually improved was name recognizability, and it made everything else... unenthickenable, eh?
23:53:50 <\oren\> or "give me the email of tech support"
23:54:03 <boily> do you consider yourself well-aligned?
23:54:50 <oerjan> i dunno, he looks a bit slanted
23:55:07 * boily diagonally mapoles oerjan
23:55:25 <Phantom_Hoover> in oren, as elsewhere
23:55:26 <\oren\> or maybe it means "give me the email of a software engineer"
23:57:25 <\oren\> the full sentence is: I'd love to see if there is a technology fit for us to work together. Please let me know; I'd be happy to align you with a technical resource.
23:58:01 <\oren\> and I just. have no idea what this is supposed to mean
←2017-04-26 2017-04-27 2017-04-28→ ↑2017 ↑all