←2019-05-25 2019-05-26 2019-05-27→ ↑2019 ↑all
00:02:14 -!- moony has changed nick to pushpoppeekbop.
01:08:03 -!- sprock1em has quit (Quit: brb).
01:09:18 -!- sprocklem has joined.
01:35:28 -!- sebbu3 has joined.
01:37:23 <esowiki> [[Talk:Bitch]] M https://esolangs.org/w/index.php?diff=62618&oldid=62611 * A * (+725) Paste the discussion here due to "request".
01:38:35 <esowiki> [[Talk:Bitch]] M https://esolangs.org/w/index.php?diff=62619&oldid=62618 * A * (-222) Less wordier discussions
01:39:27 -!- sebbu has quit (Ping timeout: 268 seconds).
01:40:37 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62620&oldid=62615 * A * (+22)
01:45:26 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62621&oldid=62620 * A * (+59)
01:45:50 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62622&oldid=62621 * A * (-78) /* Julie Andrews */
01:50:43 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62623&oldid=62622 * A * (+578) Add another negative comment
01:52:05 <zzo38> Using MSG_PEEK to find out where the line ends, and then receiving it again this time with MSG_WAITALL, seems to work fine as far as I can tell so far.
01:52:34 -!- mniip has quit (Quit: This page is intentionally left blank.).
01:54:31 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62624&oldid=62623 * A * (+616) Break my rule to add positive comments
02:03:34 -!- mniip has joined.
02:21:29 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62625&oldid=62624 * A * (+163)
02:22:12 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62626&oldid=62625 * A * (-719) Make the contest fairer
02:22:19 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62627&oldid=62626 * A * (-678) /* Julie Andrews */
02:22:40 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62628&oldid=62627 * A * (-6)
02:25:04 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62629&oldid=61480 * A * (+153)
02:25:33 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62630&oldid=62629 * A * (+1) /* Computational Class */
02:26:27 <kmc> zzo38: I think most programs would buffer and separately take lines from that buffer
02:26:37 <kmc> but your solution seems basically fine too
02:27:08 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62631&oldid=62630 * A * (+28) /* Java implementation */
02:28:50 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62632&oldid=62631 * A * (+17) /* Computational Class */
02:39:01 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62633&oldid=62628 * A * (+170) Add description
02:43:48 <int-e> fungot: are you famous?
02:43:48 <fungot> int-e: mr president, i would remind us all of what we have to deplore and condemn it at every election. the responsibility for not vetoing it. that is why mr pimenta invited me to do but which they feel " will create substantial pressure on social security for migrant workers. seventh: barriers to the internal market was a relatively easy measure to implement the internal market.
02:48:14 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62634&oldid=62633 * A * (-23)
02:49:17 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62635&oldid=62634 * A * (+39)
02:51:50 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62636&oldid=62635 * A * (+43)
02:52:12 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62637&oldid=62636 * A * (-71)
03:21:25 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62638&oldid=62632 * A * (+76)
03:22:11 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62639&oldid=62638 * A * (+21)
03:22:37 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62640&oldid=62639 * A * (+20) Specify
03:23:26 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62641&oldid=62640 * A * (-16) No, otherwise it will not compile to P''.
03:25:25 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62642&oldid=62616 * A * (+286)
03:25:45 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62643&oldid=62642 * A * (+1) I made a mess.
03:47:40 <esowiki> [[Eodermdrome]] https://esolangs.org/w/index.php?diff=62644&oldid=59661 * Salpynx * (+130) /* Implementations */ add Clojure interpreter on github
03:48:31 -!- FreeFull has quit.
04:02:51 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62645&oldid=62637 * A * (-1)
04:03:42 <esowiki> [[Eodermdrome]] https://esolangs.org/w/index.php?diff=62646&oldid=62644 * Salpynx * (+339) add "see also" and good blog post link
04:24:45 <esowiki> [[Talk:Volatile]] https://esolangs.org/w/index.php?diff=62647&oldid=62643 * TuxCrafting * (+227)
04:29:35 <esowiki> [[Talk:Volatile]] https://esolangs.org/w/index.php?diff=62648&oldid=62647 * TuxCrafting * (+261)
04:31:23 <esowiki> [[Volatile]] https://esolangs.org/w/index.php?diff=62649&oldid=62641 * TuxCrafting * (-47)
04:42:12 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62650&oldid=62645 * A * (-400)
04:44:04 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62651&oldid=62648 * A * (+173) /* Possibly Turing-complete? */
04:44:58 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62652&oldid=62649 * A * (-10) Partial undo revision 62649 by [[Special:Contributions/TuxCrafting|TuxCrafting]] ([[User talk:TuxCrafting|talk]])
04:45:35 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62653&oldid=62651 * A * (+0)
04:47:54 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62654&oldid=62653 * A * (+174) /* Possibly Turing-complete? */
04:53:14 <esowiki> [[Talk:Volatile]] https://esolangs.org/w/index.php?diff=62655&oldid=62654 * TuxCrafting * (+162)
04:58:44 <esowiki> [[Eodermdrome]] https://esolangs.org/w/index.php?diff=62656&oldid=62646 * Salpynx * (+143) /* Example programs */ Add link to my +-= 'interpreter' / decimal counter, written in Eodermdrome
05:03:06 <esowiki> [[Talk:Volatile]] https://esolangs.org/w/index.php?diff=62657&oldid=62655 * TuxCrafting * (+317)
05:04:28 -!- Frater_EST has joined.
05:04:50 -!- Frater_EST has left.
05:22:53 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62658&oldid=62657 * A * (+151) /* Possibly Turing-complete? */
05:25:26 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62659&oldid=62658 * A * (+148) I got the algorithm wrong
05:26:06 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62660&oldid=62659 * A * (+11) /* Possibly Turing-complete? */
05:26:51 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62661&oldid=62660 * A * (+6)
05:27:18 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62662&oldid=62661 * A * (-2) /* Possibly Turing-complete? */
05:31:21 <esowiki> [[Talk:Volatile]] https://esolangs.org/w/index.php?diff=62663&oldid=62662 * TuxCrafting * (+227)
05:35:35 <esowiki> [[Talk:Volatile]] https://esolangs.org/w/index.php?diff=62664&oldid=62663 * TuxCrafting * (+244)
05:37:56 <esowiki> [[EsoInterpreters]] M https://esolangs.org/w/index.php?diff=62665&oldid=62570 * A * (-35) Undo revision 62570 by [[Special:Contributions/A|A]] ([[User talk:A|talk]]): bitch cannot <u>interpret</u> the UTM, though.
05:38:36 <esowiki> [[Finite looping automaton]] M https://esolangs.org/w/index.php?diff=62666&oldid=62612 * TuxCrafting * (+3) grammar fix
05:42:28 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62667&oldid=62652 * A * (+96) /* Number tricks */
05:42:56 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62668&oldid=62667 * A * (+6) /* Computational Class */
05:43:33 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62669&oldid=62668 * A * (+5) /* Number tricks */
05:44:13 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62670&oldid=62669 * A * (+20) /* Syntax */
05:44:46 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62671&oldid=62670 * A * (-1)
05:49:28 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62672&oldid=62664 * A * (+179) /* Provided interpreter */
05:50:44 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62673&oldid=62671 * A * (+78) Inspired by Keg
06:04:02 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62674&oldid=62673 * A * (+30) /* Examples */
06:07:08 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62675&oldid=62672 * A * (+135) /* Possibly Turing-complete? */
06:09:04 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62676&oldid=62674 * A * (-1)
06:10:26 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62677&oldid=62676 * A * (+6) /* Number tricks */
06:13:51 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62678&oldid=62677 * A * (+2) The Java implementation does not count; it does not conform the spec
06:29:43 <kmc> `qutoe
06:29:44 <HackEso> ​/srv/hackeso-code/multibot_cmds/lib/limits: line 5: exec: qutoe: not found
06:29:45 <kmc> `quote
06:29:46 <HackEso> 91) <alise> like, just like I'd mark "Bob knob hobs deathly poop violation EXCREMENT unto;" as English <ais523> alise: that's great filler <alise> ais523: well it contains all the important words in the english language...
06:29:55 <kmc> `quote
06:29:56 <HackEso> 740) <Lumpio-> STOP CAPITALIZING <Lumpio-> It's making me feel weird <shubshub> the I has to be capitilized its proper grammer
06:30:00 <kmc> `quote
06:30:02 <HackEso> 539) <monqy> did you know: gravity was inspired by apples
06:30:06 <kmc> `quote
06:30:08 <HackEso> 1087) <boily> I prefer goat memory. I feel it's more reliable, like a vinyl over a CD.
06:30:12 <kmc> `quote
06:30:14 <HackEso> 1076) <+kmc> Harry Potter and the Tyranny of Structurelessness
06:30:17 <kmc> `quote
06:30:18 <HackEso> 1136) <oerjan> OKAY
06:30:20 <kmc> `quote
06:30:21 <HackEso> 1279) <zzo38> Why does my computer support only one colormap? <oerjan> because it's meta-racist
06:30:25 <kmc> `quote
06:30:26 <HackEso> 1301) <Gregor> No, I'm not alive, stop imagining I'm alive.
06:30:28 <kmc> `quote
06:30:29 <HackEso> 639) <Gregor> pikhq: And of course Rick Perry, saying that there's something wrong with a country where gays can serve in the military but we don't elect a douchebag as president.
06:30:33 <kmc> `quote
06:30:34 <HackEso> 312) <ZOMGMODULES> Felix's home page and Falcon's home page are actually the same page
06:37:45 <kmc> `quote
06:37:45 <HackEso> 829) <shachaf> Bike: Your client colours people? <Bike> it would be pretty boring to see everyone as white, i get that enough in real life
06:44:22 <kmc> `quote
06:44:22 <HackEso> 87) <fungot> alise: why internet is like wtf
06:53:58 <esowiki> [[Hello]] M https://esolangs.org/w/index.php?diff=62679&oldid=62534 * A * (-6) /* Interpreter */
06:54:28 <esowiki> [[Hello]] M https://esolangs.org/w/index.php?diff=62680&oldid=62679 * A * (+0) /* Interpreter */
06:56:19 <esowiki> [[Hello]] M https://esolangs.org/w/index.php?diff=62681&oldid=62680 * A * (-7) /* Interpreter */
07:08:28 <kmc> `quote
07:08:28 <HackEso> 556) <Ngevd> "Unlike other operating systems, Linux operating systems use Linux"
07:08:31 <kmc> `quote
07:08:32 <HackEso> 758) <elliott> I couldn't survive an apocalypse. I don't even have any bitcoins.
07:08:34 <kmc> `quote
07:08:35 <HackEso> 590) <Ngevd> Somehow I managed to read Haskell as Befunge
07:08:38 <kmc> `quote
07:08:38 <HackEso> 209) <zzo38> ais523: Maybe it is better, because I don't think the octopus will live very well in the tree. But the difference is that the Internet is lying and you cannot see such things; you could make modified picture, though, in order to lie more clearly, at least.
07:08:58 <kmc> `quote
07:08:59 <HackEso> 1206) <zzo38> Syntax highlighting is: sender in cyan, command in bright white, parameters in normal white, long parameters in bright blue (except for the preceding colon) <coppro> wait, you read IRC raw? <zzo38> It isn't quite raw; it is in colors.
07:09:12 <esowiki> [[Hello]] M https://esolangs.org/w/index.php?diff=62682&oldid=62681 * A * (+124) /* Interpreter */
07:09:21 <kmc> `quote
07:09:22 <HackEso> 707) <shachaf> Free as in unregistered, not free as in lunch or speech.
07:09:26 <kmc> `quote
07:09:27 <HackEso> 704) <zzo38> I think we are sort of this insane, and also sort of not as much as insane, and also sort of a bit more insane than that, and also somewhat more various other thing at various times whatever you are discussing at that time
07:09:32 <kmc> `quote
07:09:33 <HackEso> 923) <fungot> but when she saw him fnord and fnord. and then there's the fnord, as well as fnord reading sauce with fish, or fnord, that alice quite fnord for it hadn't spoken before.
07:09:36 <kmc> `quote
07:09:37 <HackEso> 617) <oklopol> also who it a tome, a small one
07:10:35 -!- AnotherTest has joined.
07:10:53 <esowiki> [[Hello]] M https://esolangs.org/w/index.php?diff=62683&oldid=62682 * A * (-73) /* Interpreter */
07:11:10 <shachaf> the other day someone said "free as in use-after"
07:14:07 <esowiki> [[Hello]] M https://esolangs.org/w/index.php?diff=62684&oldid=62683 * A * (+66) /* Interpreter */
07:16:00 <esowiki> [[Hello]] M https://esolangs.org/w/index.php?diff=62685&oldid=62684 * A * (-73) /* Interpreter */
07:16:07 -!- Frater_EST has joined.
07:16:13 -!- Frater_EST has left.
07:18:20 <esowiki> [[Hello]] M https://esolangs.org/w/index.php?diff=62686&oldid=62685 * A * (-1) Deleted an extra character
07:33:12 -!- LKoen has joined.
07:40:53 <esowiki> [[Hello]] M https://esolangs.org/w/index.php?diff=62687&oldid=62686 * A * (+84) /* Interpreter */
07:49:45 <esowiki> [[Hello]] M https://esolangs.org/w/index.php?diff=62688&oldid=62687 * A * (-42) /* Interpreter */
07:55:22 <esowiki> [[Volatile]] https://esolangs.org/w/index.php?diff=62689&oldid=62678 * JonoCode9374 * (+862) Added a python interpreter
07:56:15 <esowiki> [[Volatile]] https://esolangs.org/w/index.php?diff=62690&oldid=62689 * A * (+25)
08:01:19 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62691&oldid=62690 * A * (-8) A newline is fine.
08:14:00 <esowiki> [[Talk:Volatile]] https://esolangs.org/w/index.php?diff=62692&oldid=62675 * JonoCode9374 * (+317) /* Output Methods */ new section
08:15:19 <esowiki> [[Volatile]] https://esolangs.org/w/index.php?diff=62693&oldid=62691 * JonoCode9374 * (+198) /* Examples */ -> Print "Hi"
08:30:18 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62694&oldid=62693 * A * (+80) /* Syntax */
08:36:23 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62695&oldid=62694 * A * (+47) Python interpreter counts
08:36:38 <esowiki> [[EsoInterpreters]] M https://esolangs.org/w/index.php?diff=62696&oldid=62665 * JonoCode9374 * (+407) Added Keg as an implementing language
08:36:47 -!- oklopol has joined.
08:39:12 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62697&oldid=62692 * A * (+197) /* Output Methods */
08:39:24 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62698&oldid=62697 * A * (+1) /* Output Methods */
08:45:46 <esowiki> [[Talk:Volatile]] https://esolangs.org/w/index.php?diff=62699&oldid=62698 * TuxCrafting * (+1130)
08:46:30 <b_jonas> ``` recipe; starwars 5
08:46:31 <HackEso> ​ 1 c Sugar \ 1/4 c Shredded coconut \ 1/4 ts Cumin seeds \ \ Preheat oven to 350. In a medium bowl, combine milk, the sugar, vanilla and seasoned \ flour and water and then cornstarch. add tomatoes, oregano, and \ nutmeg; serve. \ \ MMMMM \ \ MMMMM----- Recipe via Meal-Master (tm) v8.05 \ \ Title: CHINESE MEAT OF 2 OR. SALAD \ Categories: Chinese, Candies, Fat \ Yield: 1 Servings \ \ Unsweetened chocolate canned \ 1 Medium potatoes \
08:46:36 <b_jonas> `starwars 5
08:46:36 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62700&oldid=62699 * TuxCrafting * (+7)
08:46:36 <HackEso> R2-D2 \ Admiral Crix Madine \ Lor San Tekka \ Rose Tico \ Jyn Erso
08:47:57 <shachaf> Has someone suggested to A to make more substantial edits rather than dozens and dozens of tiny ones?
08:48:47 <b_jonas> `? level
08:48:48 <HackEso> level? ¯\(°​_o)/¯
08:48:51 <b_jonas> hmm
08:48:59 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62701&oldid=62695 * A * (+2697) /* Display the word "Hi" */
08:50:40 <esowiki> [[Volatile]] https://esolangs.org/w/index.php?diff=62702&oldid=62701 * A * (+27) /* Hello, world! program */
08:50:43 <int-e> shachaf: they cite this on their user page: 00:21:49 <int-e> oerjan: A's microedits are annoying. and sometimes disruptive...
08:51:34 <shachaf> Hmm.
08:51:48 <shachaf> That userpage seems pretty trolly.
08:52:06 -!- tromp has quit (Remote host closed the connection).
08:52:07 <int-e> Yeah it reminds me of how hard it was to be 13 years of age.
08:52:37 <int-e> With all the world against me ;-)
08:53:24 <shachaf> I was that age in the past and I don't believe I did that kind of thing.
08:54:00 <int-e> Well, I didn't have Internet access at that age.
08:54:11 <b_jonas> int-e: and how hard it may be to be less than 13 years of age _now_, when all online services ban you from using it just because of your age. imagine living in an age when everything happens on the internet, but not being allowed to use it. you can't even register an email address, or an account practically required to use the operating system of your phone.
08:54:20 <shachaf> One thing I remember is being quite annoyed when people treated me differently for being that age. For example by making excuses for me, or being impressed by unimpressive things.
08:54:30 <int-e> I'm also being facetious to some extend. At least I'd like to believe that I am.
08:54:32 <esowiki> [[EsoInterpreters]] M https://esolangs.org/w/index.php?diff=62703&oldid=62696 * A * (+453) Hey!
08:54:48 <int-e> extent?
08:54:53 <int-e> spelling is hard.
08:55:59 <b_jonas> we didn't have any of that back when I was 13 years of age. heck, when I was aged 16, the rules that people under age of 18 can't buy tobacco or alcoholic drinks in shops were pretty new and nobody took them seriously yet. I could drink, and I could have smoked, and nobody would find 18 years to be the magical threshold.
08:56:30 <shachaf> In retrospect maybe I should've taken advantage of people being impressed by unimpressive things and being unreasonably helpful.
08:56:34 <b_jonas> mind you, even now people under 18 years of age can drink and smoke, they just aren't allowed to buy alcohol or tobacco anywhere, and that's more or less enforced.
08:57:13 <shachaf> The world is a cold uncaring place, y'know, you gotta get every advantage you can.
08:57:46 <int-e> shachaf: you may be right. otoh there's a lot of value in being taken seriously, as a confidence boost.
08:57:49 <b_jonas> shachaf: you don't realize how cold uncaring it is while you're a child if you're in a good family and reasonable school
08:58:34 <int-e> (that remark was about taking advantage of the generosity of strangers)
08:59:29 <shachaf> I'm probably putting too cynical a spin on it.
09:00:00 <int-e> And heck, it can be fun to teach somebody who's visibly learning and improving.
09:00:09 -!- adu has quit (Quit: adu).
09:00:40 <shachaf> I agree.
09:00:42 <esowiki> [[EsoInterpreters]] M https://esolangs.org/w/index.php?diff=62704&oldid=62703 * A * (-453) You should call it "bitch", despite the fact that it is insulting... it is the language name.
09:03:15 <int-e> b_jonas: it's funny but I don't see what smoking and drinking have to do with being an adult, except legally.
09:03:16 <b_jonas> int-e: sure, that's why we're all on irc
09:03:36 <b_jonas> we come here to either learn something, or teach other people something back to give back all the help we've got on irc
09:03:38 <shachaf> I would certainly prefer respect over condescension or pity or whatever.
09:04:03 <b_jonas> and if we're lucky, we also run into a good community
09:04:06 <int-e> shachaf: . o O ( I also thought that when I was your age. )
09:04:07 <int-e> ;-)
09:04:25 <shachaf> At my age I get neither!
09:04:29 <int-e> <-- half troll.
09:04:46 <b_jonas> int-e: you're older than shachaf?
09:05:18 <int-e> Yeah.
09:05:33 <shachaf> whoa, int-e knows when I was born?
09:05:40 <shachaf> That's supposed to be confidential information.
09:05:47 <shachaf> Though I haven't done a good job of keeping it that way.
09:05:50 <b_jonas> shachaf: no, probably only a bound
09:05:52 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62705&oldid=62700 * TuxCrafting * (+4) what is english
09:06:08 <shachaf> Well, time is continuous, you can only ever measure it with a bound.
09:06:24 <b_jonas> iirc you can find out how old I am from the channel logs
09:06:44 <int-e> Hmm maybe I'm not 100% sure. :)
09:07:03 <shachaf> Time for the millionaire protocol.
09:07:09 -!- oklopol has quit (Ping timeout: 268 seconds).
09:08:10 -!- tromp has joined.
09:08:40 <shachaf> By the way, someone told me a protocol which was significantly simpler than the standard socialist millionaire protocol, and claimed it solved the problem of verifying that two people were thinking of the same number.
09:09:15 <int-e> Let's say I was born before 1980 :) that should be good enough for this.
09:09:16 <rain1> how does it work
09:10:29 <int-e> You can do something analogous to DH?
09:10:39 <shachaf> It did a thing analogous to DH, yep.
09:10:49 <b_jonas> as in diffie-hellman?
09:10:55 <int-e> yes
09:11:09 <shachaf> I can't remember what it was called.
09:11:14 <rain1> analogous?
09:11:33 <b_jonas> `witchcraft
09:11:34 <HackEso> ​/srv/hackeso-code/multibot_cmds/lib/limits: line 5: exec: witchcraft: not found
09:12:06 <int-e> I imagine something like this: Agree on a random modulus modulo m a large (safe) prime p, compute m^H(x) mod p where x is your number and H is a hash function, compare to m^H(y), where y is the other party's number.
09:12:39 <int-e> Repeat with different hashes if you are unhappy with accidental matches.
09:12:46 <b_jonas> I don't understand how some of these crazy protocols can work
09:13:04 <int-e> (without the hash, you could afterwards check whether ax = by for given integers a and b)
09:13:40 <esowiki> [[Talk:Volatile]] M https://esolangs.org/w/index.php?diff=62706&oldid=62705 * A * (+2) Group the operators
09:13:48 <b_jonas> int-e: um, how does that exclude that the other party tries with all likely values of x and compares the result you tell?
09:13:57 <int-e> b_jonas: p is too big
09:14:08 <int-e> b_jonas: so that's infeasible.
09:14:16 <b_jonas> int-e: huh what?
09:14:50 <b_jonas> I don't understand this protocol
09:15:27 <shachaf> That's not quite the protocol my friend described.
09:15:30 <shachaf> Let me see if I can remember.
09:15:36 <b_jonas> I mean, how does it exclude a replay where the other person tries to redo this for multiple different versions of y
09:15:38 <int-e> b_jonas: oh I see what you mean, yes it needs some tweaking for that, hmm.
09:16:08 <int-e> b_jonas: I was assuming that x and y are also large in themselves, which is silly...
09:16:11 <rain1> I think there may be a fundamental obstraction to symmetry in these kinds of protocols
09:16:36 <rain1> often one player gets perfect security and the other player gets n-bit security
09:17:33 <shachaf> Oh, it was called SPEKE
09:17:35 <shachaf> https://en.wikipedia.org/wiki/SPEKE
09:19:56 <int-e> Aha. That uses more DH. :)
09:20:13 <rain1> I guess what i said is wrong for the Diffie-Hellman based ones
09:20:24 <rain1> but it applied to the problem of intersecting two bitsets
09:20:30 <int-e> shachaf: thanks
09:20:40 <shachaf> I think there's an extra step past what's described there where I give you H("A"++K) and you give me H("B"++k) or something.
09:20:56 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62707&oldid=62702 * A * (+6) /* Java implementation */
09:21:01 <shachaf> Which is unfortunately asymmetric because one of us has to send it first?
09:21:04 <int-e> (I'll console myself with the fact that I recognized the importance of hashing the numbers :-))
09:21:33 <shachaf> Anyway, this protocol is quite simple, and the socialist millionaire protocol seems very complicated.
09:21:48 <shachaf> As I remember it, anyway.
09:21:55 <shachaf> So what's the advantage of the complicated one?
09:22:22 <int-e> "A passive attacker doesn't lear whether x = y."
09:22:25 <int-e> *learn
09:22:34 <int-e> (paraphrased from wikipedia)
09:22:35 <b_jonas> "tweaking"
09:23:08 <int-e> b_jonas: this happens when you treat cryptographic protocols as a puzzle... you come up with flawed solutions :)
09:23:22 <shachaf> Do they learn it here?
09:23:27 <int-e> (At least in my case. I'm *not* a cryptographer.)
09:23:29 <b_jonas> int-e: sadly it also comes up when people try to use crypto seriously in production, not as a puzzle
09:23:41 <int-e> shachaf: K is communicated by both parties.
09:23:55 <int-e> shachaf: So an eavesdropper can do the comparison.
09:24:10 <b_jonas> cryptography is dangerous, you have to pay someone competent to do it, or at least pay them to tell you that your design is wrong
09:24:15 <shachaf> Why is K communicated?
09:24:33 <int-e> shachaf: how else are the parties comparing the two Ks?
09:25:00 <b_jonas> shachaf: I still don't understand how that's supposed to work
09:25:00 <esowiki> [[Talk:Volatile]] https://esolangs.org/w/index.php?diff=62708&oldid=62706 * JonoCode9374 * (+135) /* Output Methods */
09:25:06 <shachaf> The way I described, by hashing them?
09:25:33 <shachaf> Doesn't sending K defeat the whole protocol?
09:26:11 <int-e> Ah, what you described up there. Hmm.
09:26:43 <b_jonas> `? Eurytion
09:26:44 <HackEso> Eurytion? ¯\(°​_o)/¯
09:27:39 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62709&oldid=62707 * A * (+172) /* Examples */
09:28:03 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62710&oldid=62709 * A * (+0) Oops
09:28:57 <b_jonas> perhaps my hon. and learned friend fungot wants to add something to the campaign statements you told us yesterday, before I leave for the voting chambers?
09:28:58 <fungot> b_jonas: i am, however, the globalisation of financial markets. in the case of joseph o'dell, and with us in doing battle for the new citizens of the european parliament and the council on the basis of democracy.
09:29:37 <shachaf> I think int-e's real contention is that my behavior is or has been too immature to have been born before 1980.
09:29:40 <esowiki> [[User:JonoCode9374]] M https://esolangs.org/w/index.php?diff=62711&oldid=61776 * JonoCode9374 * (+170) Added something I've learnt
09:29:50 <shachaf> Which is reasonable.
09:29:54 <b_jonas> the only champion of democracy, protecting us from the evil forces who threaten it, that's what you always want to paint yourself, don't you, my hon. and learned friend fungot
09:29:56 <fungot> b_jonas: mr president, the question of the legitimacy in international trade, especially in the context of the internal market, which will take place, at community level. with this in mind and we should be ready to accept, and why we are now awaiting the commission's proposal for a new millennium round, which contradicted the very principle of preferential agreements that the council will support the latter view. it comes late
09:30:36 <b_jonas> oh, now that is more informative
09:31:01 <b_jonas> so you want to encourage international trade at the community level, such as individual people ordering small electronics from the far east through ebay?
09:31:47 <esowiki> [[Volatile]] https://esolangs.org/w/index.php?diff=62712&oldid=62710 * TuxCrafting * (+41)
09:32:42 -!- LKoen has quit (Remote host closed the connection).
09:33:34 <b_jonas> with its striped brown pattern, the building of the state archives in Amsterdam looks like it's made of wafers. it's so appetizing, perhaps more so than the gingerbread house of the witch
09:34:01 <shachaf> ^style
09:34:02 <fungot> Available: agora alice c64 ct darwin discworld enron europarl* ff7 fisher fungot homestuck ic irc iwcs jargon lovecraft nethack oots pa qwantz sms speeches ss wp ukparl youtube
09:34:34 <shachaf> ^style pa
09:34:34 <fungot> Selected style: pa (around 1200 transcribed Penny Arcade comics)
09:34:41 <shachaf> What? I wanted Peano Arithmetic.
09:34:55 <shachaf> ^style discworld
09:34:55 <fungot> Selected style: discworld (a subset of Terry Pratchett's Discworld books)
09:35:03 <shachaf> fungot: hit me
09:35:04 <fungot> shachaf: ' a mixture of fnord and violence. but, well, when i worked for my brother-in-law, fnord. the silicon anti-defamation league had been going on for some time afterwards.'
09:36:02 <shachaf> fizzie: Shouldn't there be an Illuminatus! trilogy style?
09:36:34 <b_jonas> don't change it! the europarl elevtion is not over yet
09:36:51 <b_jonas> other people may want to ask questions to the candidates
09:36:54 <shachaf> Which means you can get plenty of the real thing.
09:46:28 <b_jonas> yes, I'll go cast a real vote in a few hours. with good old paper trail, no fancy cryptographical protocols.
09:53:45 <b_jonas> although there is sort of a cryptographical element
09:54:44 <b_jonas> to cast the vote, I use the pen provided, which feels wrong, because I prefer to use a nice fancy pen for important stuff like this, but the vote is supposed to be anonymous, so it's preferable to have my ballot be as unrecognizable and generic as possible
09:56:22 -!- LKoen has joined.
09:58:06 <shachaf> bring enough nice fancy pens for everyone
10:01:46 -!- arseniiv has joined.
10:02:23 <b_jonas> shachaf: I'd have had to arrive there six hours before for that, and even then I don't know if it's allowed for me to do that
10:02:27 <fizzie> shachaf: Maybe. I'd need an electronic copy of the text though.
10:03:28 <shachaf> kmc gave me a printed copy of the text
10:03:32 <shachaf> it contains many electrons
10:03:54 <fizzie> I've got a printed copy as well.
10:04:38 -!- LKoen has quit (Remote host closed the connection).
10:04:42 <shachaf> You can have all the electrons out of mine if you want.
10:04:45 <shachaf> And the positrons.
10:04:59 -!- LKoen has joined.
10:06:17 <fizzie> I should get mosh set up, this train wifi disconnects TCP connections every time it either stops at a station or starts moving again.
10:10:53 -!- Lord_of_Life has quit (Ping timeout: 245 seconds).
10:11:59 -!- Lord_of_Life has joined.
10:24:41 <int-e> shachaf: AFAICS that socialist millionaire's protocol relies solely on discrete logs/computational Diffie-Hellman... in particular the description in the OTR protocol remarks that the quantity h^{\alpha\beta a b} cannot be computed by either party using the information transmitted in the protocol, making it effectively a random element modulo p, so no information about (x-y) beyond equality is...
10:24:47 <int-e> ...revealed.
10:26:47 <b_jonas> "solely"
10:27:33 <int-e> b_jonas: hmm, you think otherwise?
10:28:38 <shachaf> You mean it doesn't rely on a hash?
10:28:47 <int-e> SPEKE has stronger, less well researched, assumptions.
10:28:50 <int-e> shachaf: yes
10:29:17 <shachaf> I feel like a cryptography hash isn't that strong an assumption?
10:30:06 <int-e> (Of DH's x |-> x^a (mod p) has a lot of the relevant properties of a hash function, being a random permutation.)
10:30:18 <int-e> err pseudo-random.
10:30:56 <int-e> shachaf: Oh but SPEKE is also murky on the discrete log side... it's not a clearcut instance of DH.
10:30:58 -!- LKoen has quit (Remote host closed the connection).
10:31:12 <b_jonas> int-e: no, it's just that it's funny to say that a protocol relies "solely" on something as strong as a public key cryptography primitive.
10:31:49 <b_jonas> DH is a stronger assumption than a one-way function
10:32:04 <b_jonas> possibly still weaker than a full public-key crypto, but still
10:32:25 <int-e> b_jonas: No, it's not odd at all. "solely" means that there's nothing else, not that the thing referred to is insignifcant.
10:32:30 <b_jonas> to me the distinction goes between whether you use only symmetric crypto, and when you need public key crypto too
10:32:57 <shachaf> i,i what kind of coin is that?
10:33:48 <b_jonas> even if RSA or DH is a funnier thing to explain as a popular mathematics bling than any symmetric crypto stuff
10:34:01 <shachaf> Do you like this, where this is Lamport signatures?
10:34:13 <b_jonas> `hello darkness, my old friend
10:34:13 <HackEso> hello, World
10:34:36 <int-e> `? s&g
10:34:37 <HackEso> s&g? ¯\(°​_o)/¯
10:34:38 <shachaf> `hi darkness, my old friend
10:34:39 <HackEso> Hi darkness, my old friend. Harkness, my old friend.
10:35:06 <int-e> Playing 08 The Sounds Of Silence.ogg.
10:35:38 <b_jonas> `? hi
10:35:38 <shachaf> that song is from way before 08 hth
10:35:39 <HackEso> hi? ¯\(°​_o)/¯
10:35:46 <b_jonas> `? `hi
10:35:46 <HackEso> ​`hi? ¯\(°​_o)/¯
10:35:51 <b_jonas> hmm
10:35:55 <int-e> shachaf: it's number 08 on that particular best of album.
10:36:25 <int-e> Hmm.
10:37:44 <shachaf> Oh, you're playing The Sounds of Silence.
10:37:48 <shachaf> Maybe that song is from 08.
10:39:10 -!- LKoen has joined.
10:39:12 <b_jonas> as opposed to the movie?
10:39:49 <int-e> Probably not, but I've never noticed that extra 's' before. (Especially since the lyrics mix singular and plural? Or is it this version...)
10:41:22 <int-e> shachaf: It's more complicated: '"The Sound of Silence", originally "The Sounds of Silence", is a song by the American music duo Simon & Garfunkel.'
10:42:25 <shachaf> `quote sound
10:42:25 <HackEso> 263) <tswett> Of course, "b" is clearly just "pv". <tswett> Say "pvottle". It will sound... similar to exactly the same as "bottle". \ 290) <olsner> boston cream pie? sounds related to a cleveland steamer \ 298) [on Sgeo's karaoke] <not_nddrylliog> Sgeo: awesome <not_nddrylliog> sounds like a japan anime sound track \ 385) <SgeoN1> No nasty sounds for a while now. Going to turn off and on and see if the numbers get worse. \ 432) <Taneb> Well, I'm now ex
10:42:33 <b_jonas> oh wait, the film is called "The Sound of Music", not "The Sound of Silence"
10:42:58 <shachaf> `2 quote sound
10:42:59 <HackEso> 2/4:ow experimenting with clients <fizzie> It doesn't sound like good PR to say that out loud. \ 605) <ais523> also, why isn't monqy from Hexham? his name sounds like he should be \ 734) <kmc> i saw Godspeed You! Black Emperor live <kmc> it is hard to tell when the soundcheck ends and the set begins \ 789) <elliott> this sounds sort of like @ kmc <kmc> well @ is the least upper bound of all ideas in computer science \ 799) <HackEgo> 499) <zzo38> What i
10:43:22 <shachaf> I wonder why, when I run a program with posix_spawn instead of fork/exec, it wraps somewhat differently in the terminal.
10:43:34 <shachaf> Unfortunately I don't care quite enough to find out.
10:45:28 <int-e> . o O ( wrap? )
10:46:04 <shachaf> Well, the formatting is somehow different, when I run ls.
10:47:35 <shachaf> Different number of columns.
10:49:19 <b_jonas> shachaf: you can use ls -1 to force one column, ls -C to force multiple columns, and the COLUMNS variable will probably tell how many cells wide ls will format the multi-column output
10:49:43 <int-e> shachaf: perhaps a silly question, but what are you passing for envp?
10:49:54 <b_jonas> so eg. run (COLUMNS=132 ls -C) if you want to print the output to a 132-column printer
10:49:56 <shachaf> Aha. I just worked it out.
10:50:04 <shachaf> posix_spawn is indeed running with an empty environment.
10:50:10 <shachaf> I'm surprised anything worked.
10:50:28 <b_jonas> shachaf: yeah, we often feel that way in modern computing
10:51:12 <b_jonas> heck, in my job my current task is to try to get some stupid third-party mess of a software working, and I don't think I'll succeed
10:51:26 <b_jonas> all I can do is run installers at random and get uninformative error messages
10:51:38 <shachaf> So what are you supposed to do?
10:51:53 -!- sebbu3 has changed nick to sebbu.
10:51:53 <shachaf> I see some examples that do "char **environ;".
10:51:58 <shachaf> That seems a bit sketchy.
10:52:04 <int-e> that seems to be the task: run installers at random and hope it works
10:53:09 <b_jonas> I'm getting three "connection failed" error message popups at bootup now, with no detail on what connection to what failed
10:53:49 <shachaf> Is using a magic global variable the standard way to do this?
10:53:54 <int-e> shachaf: hmm. `man environ` agrees with that... but doesn't have a "CONFORMING TO" section :-(
10:55:15 <int-e> (defining _GNU_SOURCE and using <unistd.h> would look better)
10:55:50 <b_jonas> the archæological museum in Leiden was awesome by the way. it's the archæological museum with the second largest exhibition I've ever seen (this may change when I go to London). it has _three_ egyptian sanctums rebuilt inside the building
10:56:21 <b_jonas> int-e: the gnu libc info documentation may tell some details
10:56:27 <int-e> shachaf: http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap08.html#tag_08
10:56:35 <shachaf> But I don't want _GNU_SOURCE!
10:56:36 <int-e> so it's posix, by all appearances.
10:56:42 <int-e> deal with it.
10:56:42 <shachaf> I mean, maybe I do.
10:56:59 <shachaf> So does setenv actually modify that array?
10:57:01 <shachaf> I guess it must.
10:57:25 <shachaf> Maybe I'll show them by not using libc at all.
10:57:25 <b_jonas> int-e: https://www.gnu.org/software/libc/manual/html_node/Environment-Access.html#index-environ
10:57:48 <shachaf> I'll write my own entry point that just gets the environment right from the place ELF or whatever specifies.
10:57:53 <b_jonas> it says "This variable is declared in the header file unistd.h", and also about the third parameter to main
10:58:26 <int-e> b_jonas: that is not in agreement with the manpage.
10:58:52 <int-e> "([The environ] variable must be declared in the user program, but is declared in the header file <unistd.h> if the _GNU_SOURCE feature test macro is defined.)"
10:59:40 <shachaf> OK, now ls is well-behaved.
11:00:31 <shachaf> I guess I'll go back to posix_spawnp.
11:01:07 <b_jonas> posix_swamp?
11:01:47 <shachaf> I feel like the posix_spawn interface is way too complicated, though.
11:02:38 <int-e> b_jonas: and experimentally, the manpage is right.
11:02:47 <shachaf> It consists of... 21 functions.
11:02:52 <shachaf> Isn't that ridiculous?
11:02:57 <b_jonas> why are you even spawning ls?
11:03:06 <b_jonas> or is that just to test the swamp?
11:03:16 <shachaf> I mean, I'm usually spawning other commands, but I was testing with ls and even that behaved badly.
11:03:44 <int-e> posix_spawn_file_actions_adddup2 <-- no further questions, your honor.
11:04:02 <shachaf> When did people start writing all their APIs like api_thing_t *x = api_malloc_my_thing(); api_set_thing_blah(x, 1); api_set_thing_blam(x, 2); ?
11:04:24 <shachaf> Instead of just using s a struct and maybe a version number or sizeof or something for ABI compatibility?
11:04:38 <int-e> (I understand it... everything that you might want to do between fork and exec has to be supported.)
11:04:51 <b_jonas> shachaf: since you want a C interface with binary compatibility decades into the future
11:05:05 <b_jonas> struct with version number... could work, sure
11:05:09 <int-e> shachaf: OO is good for you! Embrace it!
11:05:11 <shachaf> Look at this nonsense: https://www.alsa-project.org/alsa-doc/alsa-lib/group___p_c_m___h_w___params.html
11:05:17 <int-e> And in C, that's what OO looks like.
11:05:26 <shachaf> I was writing ALSA code and it's just such a mess.
11:06:00 <b_jonas> seriously, three sanctums
11:06:02 <shachaf> What happened to writing functions that accept data as argument?
11:06:07 <shachaf> s
11:06:13 <shachaf> datas as argument
11:06:37 <int-e> Also this way you can write hundreds of lines of code without thinking too much. It's, a huge productivity boost.
11:06:56 <b_jonas> shachaf: just be happy it's not one of those setter functions that take an argument with variable type, using C ... prototype, like setsockopt or the curl functions
11:07:11 <b_jonas> those are great ways to get silent type errors
11:07:16 -!- LKoen has quit (Remote host closed the connection).
11:07:30 <shachaf> Also what's the deal with APIs that malloc things just because their size might change in the future?
11:07:45 <b_jonas> shachaf: example?
11:08:12 <shachaf> https://tronche.com/gui/x/xlib/ICC/client-to-window-manager/XAllocSizeHints.html
11:08:17 <b_jonas> (and make sure it doesn't malloc things to get an aligned data storage on old C)
11:08:39 <shachaf> Instead of just telling you the size, they give you a function to malloc it and that's it.
11:08:47 <shachaf> What if there was a rule that APIs aren't allowed to call malloc?
11:08:50 <shachaf> imo good rule
11:09:07 <int-e> meh
11:09:46 <int-e> You'd end up having to pre-allocate everything yourself, with high buffer overflow potential.
11:10:03 <int-e> (especially because programmers are smart)
11:11:06 <shachaf> You can still include the malloc API if you want, I guess.
11:11:07 <b_jonas> int-e: and you'll end up with MS-style functions that take the size of the buffer in two arguments
11:11:22 <shachaf> Static allocation seems safer to me than malloc/free anyway.
11:11:25 <shachaf> No lifetime issues.
11:11:53 <b_jonas> you know, safe_memcpy(void *src, size_t srclen, void *dest, size_t destlen, size_t numbytes);
11:12:00 <shachaf> Another thing ALSA could have done was put a reasonable number of padding bytes in the struct, and require them to be 0.
11:12:08 <shachaf> That would let them do extensibility just fine.
11:13:29 <shachaf> You don't even need that many because you can put a pointer there in the future if you really need a lot of space.
11:14:10 <shachaf> I bet I shouldn't be using libasound anyway because it's LGPL.
11:14:33 <int-e> shachaf: I'm not opposed to providing a low-level API that leaves allocation to the user, but I would treat it like that, as a low-level API for special case use.
11:14:59 <int-e> For most uses, having to think about allocation will just lead to more bugs.
11:15:13 <shachaf> But you have to think about allocation anyway even if someone else does it for you.
11:15:39 <int-e> (It's bad enough that you will end up having to check for errors all the time, though there are tricks using NULL or static error objects to alleviate that need.)
11:15:45 <shachaf> pulseaudio's API is even worse, of course. They insist on doing reference counting.
11:15:50 <shachaf> harfbuzz too.
11:16:14 <int-e> glib :-P
11:16:20 <shachaf> APIs that do reference counting for no reason should just be illegal.
11:16:39 <int-e> and what makes you say that they do it for no reason?
11:17:06 <shachaf> They claim the reason is my convenience.
11:17:16 <int-e> (I'm not a huge fan of reference counting but that's because I'd prefer proper GC.)
11:17:24 <shachaf> But what's convenient for me is to clearly know the lifetime of every object.
11:17:48 <int-e> Well, you're special.
11:18:00 <shachaf> I don't think I'm that special?
11:18:16 <shachaf> s/clearly/be able to/
11:18:17 <b_jonas> or you can just have multiple apis, of different level, for the same underlying library. which already happens sometimes, with C and C++ and python apis.
11:18:36 <b_jonas> and sometimes even with multiple C apis or multiple C++ apis of different levels
11:18:40 <esowiki> [[Volatile]] https://esolangs.org/w/index.php?diff=62713&oldid=62712 * TuxCrafting * (+1908)
11:18:52 <shachaf> Sure, put whatever nonsense you want on top of the basic API.
11:19:12 <int-e> shachaf: the standard attitude today is not to care unless the object holds resources other than memory :P
11:19:41 <int-e> (Aka scarce resources. ;-) )
11:19:49 <int-e> *AKA
11:19:55 <shachaf> "memory is cheap, allocations are expensive"
11:21:08 <int-e> make inner loops fast.
11:21:17 * int-e shrugs.
11:22:02 <int-e> shachaf: It's not even that I don't care about these things. They just don't matter 99% of the time.
11:22:17 <shachaf> anyway here's my fancy program that uses posix_spawnp: http://slbkbs.org/tmp/tym.c
11:22:43 <int-e> In the remaining 1% of the cases I do want the low-level interfaces, I will complain about reference counting... and pretty much agree with all the other things you're saying.
11:23:09 <shachaf> My claim isn't merely that the reference counting is easier but less efficient.
11:23:21 <int-e> it's less efficient than doing GC properly, yes.
11:23:23 <shachaf> The claim is that it's both less convenient and less efficient.
11:24:21 <shachaf> GC is just not practical for anything that wants to be even moderately real-time as far as I can tell.
11:24:59 <int-e> real-time and fast are often at odds anyway.
11:25:31 <shachaf> GCs seem to be neither.
11:25:35 <int-e> (Real-time GC is a thing. And last I looked it was a trade-off between maximal response time and performance.)
11:26:11 <shachaf> Yes, GC trades throughput for latency.
11:26:17 <int-e> GC trades memory for speed. You want to do it as infrequently as possible, and you want to do it in a way where you don't pay for the garbage (that's the main attraction of copying GC).
11:26:20 <shachaf> And even in the best case the latency requirements are pretty high.
11:26:23 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62714&oldid=62713 * A * (-57)
11:26:46 <shachaf> But often you can do better than sophisticated GC in both latency and throughput, using much less sohisticated code.
11:27:20 * int-e shrugs
11:27:26 <shachaf> (Because GCs have to do everything dynamically and you can do it statically.)
11:27:51 <shachaf> Are any real-time GCs even available to me?
11:27:53 <int-e> The reference counting (poor person's GC) comes up in libraries for a reason... you can't predict how things are going to be shared at that level.
11:28:09 <shachaf> The library can't predict it, but I sure can.
11:29:01 <int-e> https://docs.oracle.com/javase/realtime/doc_2.2/release/JavaRTSQuickStart.html <-- really just soft real time, last I checked. But they tried.
11:30:17 <shachaf> What's the maximum pause time?
11:31:08 <shachaf> In https://wiki.openjdk.java.net/display/zgc/Main which I think is a newer project, they give a maximum pause time of 10ms which is just ridiculously long.
11:33:14 <shachaf> Ah, hmm, that page seems to be claiming 100µs? I'm suspicious.
11:33:27 <b_jonas> lol
11:34:33 <b_jonas> I tried to recreate a directory structure inside a subdirectory, like (find -type d -exec mkdir -v foo/{} \;)
11:34:37 <b_jonas> guess why that's a bad idea
11:35:55 <int-e> I've stopped following this a decade ago... back then the status was that a) academic work on real-time GC exists -- basically, if you have an upper bound on the allocation rate in addition to the execution times of your jobs, you can make things work out, given enough memory b) there are real-time Java efforts with somewhat murkier guarantees, making them soft real time. apparently in the...
11:36:01 <int-e> ...meantime they promise hard real time... under suitable conditions (which you have to check as a programmer).
11:37:03 <int-e> What about Erlang, I wonder...
11:37:20 <shachaf> Erlang has per-thread GC with no shared memory.
11:37:32 <shachaf> So no long global pauses, I assume.
11:37:58 <int-e> (In Haskell the state of the art seems to be to limit allocations *and* the working set size dramatically, and possibly do eager GC (like once per frame for real-time graphics))
11:38:37 <int-e> IOW, there's no designed-to-be real-time GC for Haskell that I'm aware of.
11:41:25 <shachaf> One good "GC" you can do once-per-frame is to allocate everything in an arena and then clear it at the end of the frame.
11:41:45 * int-e wouldn't use Haskell beyond small, best-effort, soft real time tasks.
11:42:01 <shachaf> This is very simple and cheaper than the fanciest GC.
11:42:28 <int-e> no, it's exactly as cheap as a copying GC when you've lost all the references to everything you've allocated during the frame :-P.
11:42:49 <shachaf> How?
11:43:01 <b_jonas> but... don't you sometimes want to keep data from one frame to another?
11:43:10 <int-e> because you have no references to follow. and then you discard the semispace and use the other one.
11:43:12 <shachaf> A copying GC takes time proportional to your roots or live set or whatever.
11:43:25 <shachaf> It needs to at least look at all of those.
11:43:35 <int-e> shachaf: a sufficiently fancy GC would know that those roots haven't been touched ;-)
11:43:48 <shachaf> b_jonas: Yes, so you don't allocate those things in the per-frame arena.
11:44:23 * int-e is being contrary.
11:44:47 <shachaf> But this isn't even information you can specify to the typical GC because the idea of a GC is that you can be all "idk my bff jill" about it.
11:44:54 <int-e> shachaf: I mean, you're assuming that the programmer can outsmart the GC. You're probably right.
11:44:57 <shachaf> So it has to figure out everything dynamically.
11:45:11 <int-e> shachaf: But there's no fundamental reason for this ;-)
11:45:33 <shachaf> The fundamental reason is that the GC is dynamic.
11:45:44 <b_jonas> but yes, I do agree that in many situations, allocating stuff in a pool that you free at once is a useful strategy
11:45:56 <shachaf> If your language lets you do things more statically, with regions or whatever, then your GC isn't really a GC anymore.
11:46:38 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62715&oldid=62650 * A * (+336)
11:46:42 <int-e> It doesn't have to be. You can do static life-time analysis to improve the GC. I'm sure there's research on that...
11:47:04 <int-e> shachaf: I disagree quite strongly with that premise.
11:47:38 <int-e> shachaf: congratulations, you made the wiki!
11:47:46 <shachaf> I'm pretty mad about that.
11:47:53 <b_jonas> int-e: the problem is that it's a bit hard to mix static and gc, because your gc often has to traverse the data you allocated, even such data that you will deallocate precisely at the right moment because of static considerations
11:48:18 <shachaf> My suggestion to the wiki adminstrators: Edit my name out of the wiki; purge my name out of the wiki history; ban A.
11:48:39 <shachaf> They are obviously acting in bad faith.
11:48:47 <b_jonas> and I do think that in most cases, static lifetimes works well, and refcounting works in the rest of the cases
11:49:26 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62716&oldid=62715 * A * (+406)
11:49:39 <b_jonas> it's just that the static lifetimes are sometimes hard to prove, so you use refcounting or gc to work around programmer mistakes, even if it's technically slower at runtime
11:50:00 <shachaf> I'm feeling pretty distressed about being quoted on the wiki like that.
11:50:09 <int-e> yeah I suspect a ban will happen sooner or later.
11:50:19 <b_jonas> you take the runtime speed penalty because it's better than the potential use after free from a bug
11:50:43 <int-e> or many A ban.
11:50:46 <int-e> *maybe
11:51:06 <b_jonas> A is back? hello
11:51:07 <b_jonas> `hi A
11:51:08 <HackEso> Hi A. HA.
11:51:21 <shachaf> int-e: Well, maybe a static lifetime manager exists that's as good as handwritten code and as easy as never thinking about allocations. But it sounds like even a rough approximation is an open research problem.
11:51:29 -!- LKoen has joined.
11:51:42 <shachaf> Approximately all existing GCs rely on doing most everything at runtime.
11:52:20 <b_jonas> and then there's python
11:53:59 <shachaf> oerjan or whoever: I would appreciate not being mentioned in the wiki in any way, not even in the history, if that can be arranged.
11:54:01 <int-e> just a random find: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.219.1650
11:56:27 <int-e> shachaf: I think the main problem here is this: GC is /good enough/ for so many purposes... except when you have hard memory or time constraints, and then nobody's would be willing to trust any amount of static analysis to get things right; they want to see exactly what happens in the code. So there's no real market for better static analysis in this area.
11:56:50 <shachaf> Hmm, by "not research" I meant "an implementation I can use for something", rather than a paper with a sentence like "we formulate an operational semantics".
11:57:06 <b_jonas> heh
11:57:09 <int-e> shachaf: I was following my own "there's research" remark.
11:57:22 <shachaf> Ah, sure.
11:57:32 <shachaf> int-e: I suspect GC isn't actually good enough for any interactive program.
11:57:47 <b_jonas> oh nice, this command line is almost completed
11:58:00 <shachaf> Today consumer screens run at 120Hz or 144Hz refresh rate.
11:58:13 <shachaf> 144Hz is <7ms update time.
11:58:19 <int-e> shachaf: I think you're proving wrong in practice, by tons of interactive software written in Java and Javascript.
11:58:40 <int-e> 60Hz is still mostly standard, 120Hz is for gamers ;-)
11:58:41 <shachaf> Oh, sorry, I meant "good software".
11:58:42 <b_jonas> my TFT screen runs at 60Hz, but I'm not sure it counts as "today's".
11:58:53 <int-e> (especially since we have this silly UHD trend)
11:59:30 <shachaf> The iPad Pro runs at 120Hz.
11:59:56 <b_jonas> `? resolution
11:59:57 * int-e ionders about that 144Hz figure..
11:59:57 <HackEso> As of 2015, highest resolution commercial computer monitors are 5120x2880 Apple and 3840x2160 other.
12:00:08 <b_jonas> int-e: good point, I think this wisdom has to be updated
12:00:32 <shachaf> My laptop is 3840x2160.
12:00:38 <b_jonas> I think there are now monitors with larger resolution
12:00:39 * int-e should try finding a HD IPS display whiler there's still time...
12:01:25 <b_jonas> um, what does "HD IPS" mean? I'm lost in all these marketing abbreviations
12:01:45 <b_jonas> "HD" is some sort of resolution that they use so we don't know the actual numbers, right?
12:01:55 <b_jonas> but what resolution? I can't follow all these abbreviations
12:01:56 <int-e> HD = 2K = 1920x1080; IPS = in plane switching, which is the TFT technology that gives high angular color invariance.
12:02:24 <b_jonas> ok
12:02:26 <int-e> (and which my current display doesn't have)
12:02:43 <shachaf> Wow, you can even get a 240Hz laptop.
12:02:46 <shachaf> > 1000/240
12:02:48 <lambdabot> 4.166666666666667
12:02:53 <int-e> why?!
12:03:01 <int-e> why waste all that precious memory bandwidth?
12:03:09 <shachaf> To match the responsiveness of an Apple II?
12:03:17 <shachaf> https://danluu.com/input-lag/
12:03:52 <b_jonas> can humans even distinguish 240 Hz and 120 Hz progressive updates?
12:04:31 <shachaf> I'll have to try out the hardware sometime.
12:04:42 <shachaf> People say 60->120 is a big improvement.
12:04:43 <int-e> though tbh I'd be more willing to spend that bandwidth on higher refresh rates than on higher resolution
12:04:51 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62717&oldid=62716 * A * (+135)
12:05:08 <int-e> yeah that 8ms improvement should be noticable.
12:05:09 <b_jonas> int-e: for 120 Hz versus 60 Hz, sure, that makes sense
12:05:43 <shachaf> If you're at 1920x1080 at 60 fps, you can quadruple the bandwidth to go to 4K or 240Hz.
12:05:52 <b_jonas> int-e: not when you run windows software that sometimes makes you wait for seconds before they react
12:06:07 <int-e> > 1/120 -- this is how much ealier you may get to see an update when using 120 instead of 60Hz refresh rate
12:06:09 <lambdabot> 8.333333333333333e-3
12:06:34 <shachaf> > 1000/120
12:06:37 <lambdabot> 8.333333333333334
12:06:40 <shachaf> hth
12:06:56 <int-e> I can add -3 and 3 ;-)
12:06:59 <shachaf> 8ms is a long time.
12:07:17 <b_jonas> int-e: but 120 Hz referesh rate can be better than 60 Hz for reasons other than latency too
12:07:26 <int-e> For 240Hz that can give you another 4ms... that really should be barely noticable.
12:07:56 <shachaf> I am actually confusil about why people went with 144Hz rather than 120Hz.
12:08:03 <int-e> Yeah, 120Hz is a multiple of 24.
12:08:18 <int-e> (24 being a standard movie frame rate)
12:08:35 <int-e> Maybe there are 48Hz movies? That could explain the 144.
12:09:14 <int-e> or just 24fps with interleaving...
12:14:27 <int-e> I still think a lot of this is just the marketing attitude that bigger numbers are better.
12:17:19 <b_jonas> what I'd like is cheap cameras where the sensor has six or seven different color channels, rather than just three. they'd see differences in color that we can't percieve. yes, I do understand that we have almost no software support for such images, and also that it would come at the expense of resolution, but still.
12:17:35 <b_jonas> I'm not saying that all cameras should have that, sure, we can keep the 3 and 1 channel cameras too
12:17:39 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62718&oldid=62717 * A * (+254)
12:18:31 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62719&oldid=62718 * A * (+47)
12:19:00 <int-e> . o O ( can we rate limit edits per user to bursts of 5, 5 per hour? ;-) (I might regret thios...) )
12:19:03 <b_jonas> int-e: yeah. and in a sense that's better when they use words where it's not even clear which one is better, like "full HD" and "QVGA" and "widescreen" and "ultra widescreen" so they can toggle between two options every few years and sell the new option as better than the old option every time
12:19:17 <b_jonas> I think we went from iodized salt to iodine-less salt and back twice each already
12:19:29 <b_jonas> each time they sold it as an advantage over the previous kind of salt
12:19:48 <b_jonas> so now the fancy salt costs 200 forints a kilogram rather than 12 forint per kilogram
12:20:09 <b_jonas> int-e: please don't use such a low limit
12:20:28 <esowiki> [[Talk:Bitch]] M https://esolangs.org/w/index.php?diff=62720&oldid=62619 * A * (+378) Is this the etymology?
12:20:32 <int-e> b_jonas: I know :P
12:20:42 <b_jonas> int-e: or do you mean announcements of the edits on the irc channel? because that could work
12:21:20 <int-e> b_jonas: It's a very flawed idea... rate limits don't model the intent (what matters is the significance of the edits)
12:21:48 <b_jonas> int-e: so just limit to 5 bad edits per day, and an unlimited number of good edits
12:22:04 <esowiki> [[User:A]] M https://esolangs.org/w/index.php?diff=62721&oldid=62719 * A * (-301)
12:22:07 <b_jonas> yeah, let me repeat my question from a few days ago
12:23:13 <b_jonas> in linux, for an ip socket, what setsockopt do I use to change whether the evil bit shall be set in outgoing packets for that socket? I can't find it in http://man7.org/linux/man-pages/man7/ip.7.html
12:23:23 -!- GeekDude has quit (Ping timeout: 245 seconds).
12:24:47 <int-e> @google IP "evil bit"
12:24:49 <lambdabot> https://www.ietf.org/rfc/rfc3514.txt
12:25:21 <int-e> Ah, 1 April 2003 ... maybe you don't.
12:26:48 -!- GeekDude has joined.
12:33:20 <esowiki> [[Jungle]] N https://esolangs.org/w/index.php?oldid=62722 * Madk * (+22095) Add Jungle esolang
12:38:34 <b_jonas> ok, I have to leave now, both for the election and for real-life stuff involving family. see you later.
12:38:36 -!- b_jonas has quit (Quit: leaving).
12:46:20 -!- oklopol has joined.
12:56:05 <esowiki> [[Hello world program in esoteric languages]] M https://esolangs.org/w/index.php?diff=62723&oldid=61699 * A * (+2527) /* VerboseFuck */
12:59:43 <esowiki> [[Talk:Volatile]] https://esolangs.org/w/index.php?diff=62724&oldid=62708 * TuxCrafting * (+2490)
13:04:21 -!- FreeFull has joined.
13:16:16 <esowiki> [[Talk:Volatile]] https://esolangs.org/w/index.php?diff=62725&oldid=62724 * TuxCrafting * (+1478)
13:18:14 <esowiki> [[Language list]] https://esolangs.org/w/index.php?diff=62726&oldid=62459 * Madk * (+13) /* J */ Add Jungle to language list
13:20:37 <esowiki> [[Talk:Volatile]] https://esolangs.org/w/index.php?diff=62727&oldid=62725 * TuxCrafting * (+72)
13:23:25 <esowiki> [[Volatile]] https://esolangs.org/w/index.php?diff=62728&oldid=62714 * TuxCrafting * (-369)
13:42:20 -!- xkapastel has joined.
14:04:48 -!- oklopol has quit (Ping timeout: 272 seconds).
14:09:07 -!- LKoen has quit (Remote host closed the connection).
14:14:23 -!- Sgeo__ has quit (Read error: Connection reset by peer).
14:14:47 -!- Sgeo__ has joined.
14:19:14 -!- arseniiv has quit (Read error: Connection reset by peer).
14:19:23 -!- arseniiv has joined.
14:20:12 -!- Soni has left ("http://quassel-irc.org - Chat comfortably. Anywhere.").
14:34:39 -!- adu has joined.
14:38:06 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62729&oldid=62728 * A * (+74)
14:38:15 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62730&oldid=62729 * A * (-1766) /* Buggy Java implementation */
14:41:22 -!- Sgeo_ has joined.
14:44:39 -!- Sgeo__ has quit (Ping timeout: 258 seconds).
15:07:47 <int-e> /ignore -channel #esoteric -regex -pattern " .5\\*. .03A. .5\\*. " esowiki <--- I finally found a pattern that works with irssi... have to include the color codes... silly.
15:31:44 <esowiki> [[Volatile]] https://esolangs.org/w/index.php?diff=62731&oldid=62730 * TuxCrafting * (-4) optimize examples
15:41:20 <int-e> beautiful: VPMADD52HUQ — Packed Multiply of Unsigned 52-bit Unsigned Integers and Add High 52-bit Products to 64-bit Accumulators
15:41:46 <esowiki> [[User:Madk]] https://esolangs.org/w/index.php?diff=62732&oldid=62339 * Madk * (+13) /* Implemented */ Add Jungle to implemented language list
15:42:04 <int-e> (s/H/L/;s/High/Low/ exists as well)
15:45:14 -!- adu has quit (Quit: adu).
15:49:03 <esowiki> [[Jungle]] https://esolangs.org/w/index.php?diff=62733&oldid=62722 * Madk * (-50) /* Reference implementation */ Fix inaccurate info about ///BEGIN/// and ///END///
16:13:03 <esowiki> [[Point]] https://esolangs.org/w/index.php?diff=62734&oldid=62491 * Unlimiter * (+6)
16:14:09 <esowiki> [[Point]] https://esolangs.org/w/index.php?diff=62735&oldid=62734 * Unlimiter * (-284)
17:20:21 -!- MDude has quit (Quit: Going offline, see ya! (www.adiirc.com)).
17:48:30 -!- b_jonas has joined.
17:52:07 -!- MDude has joined.
18:10:03 -!- LKoen has joined.
18:10:42 <b_jonas> `? may
18:10:43 <HackEso> may? ¯\(°​_o)/¯
18:10:45 <b_jonas> `? can
18:10:46 <HackEso> Can cans can cans?
18:11:03 -!- LKoen_ has joined.
18:12:12 * int-e can scan canned cans.
18:13:15 <b_jonas> May is an auxiliary verb, a month, and a prime minister.
18:14:25 <kmc> *was* a prime minister!
18:14:43 -!- LKoen has quit (Ping timeout: 252 seconds).
18:14:52 <b_jonas> still is, I think, but not for very long
18:15:27 -!- LKoen_ has quit (Ping timeout: 252 seconds).
18:18:02 <b_jonas> /topic IOCCC winners are announced; source code release planned for 2019-06-02 | Welcome to the international stock market for esoteric programming language invention, implementation, ignorance, innovation, and integration! | https://esolangs.org | logs: https://esolangs.org/logs/ http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D |
18:18:07 <b_jonas> https://www.dropbox.com/s/fyhqyvy3i8oh25m/wisdom.pdf
18:18:09 <b_jonas> that's too long
18:19:04 <b_jonas> /topic IOCCC winners are announced; source code release planned for 2019-06-02 | Welcome to the international center for esoteric programming languages! | https://esolangs.org | logs: https://esolangs.org/logs/ http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D | https://www.dropbox.com/s/fyhqyvy3i8oh25m/wisdom.pdf
18:19:17 <b_jonas> better, but still slightly too long I think... what's the max length?
18:19:23 <b_jonas> 390
18:19:41 <b_jonas> [ # 'IOCCC winners are announced; source code release planned for 2019-06-02 | Welcome to the international center for esoteric programming languages! | https://esolangs.org | logs: https://esolangs.org/logs/ http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D | https://www.dropbox.com/s/fyhqyvy3i8oh25m/wisdom.pdf'
18:19:42 <j-bot> b_jonas: 335
18:19:46 <b_jonas> should work
18:19:53 -!- b_jonas has set topic: IOCCC winners are announced; source code release planned for 2019-06-02 | Welcome to the international center for esoteric programming languages! | https://esolangs.org | logs: https://esolangs.org/logs/ http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D | https://www.dropbox.com/s/fyhqyvy3i8oh25m/wisdom.pdf.
18:20:45 <b_jonas> can we remove the old wisdom pdf?
18:21:13 -!- b_jonas has set topic: IOCCC winners are announced; source code release planned for 2019-06-02 | Welcome to the international center for esoteric programming language design, development, and deployment! | https://esolangs.org | logs: https://esolangs.org/logs/ http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D.
18:21:23 <b_jonas> `? wisdom.pdf
18:21:24 <HackEso> Nicely formatted wisdoms and quotes book at https://www.dropbox.com/s/fyhqyvy3i8oh25m/wisdom.pdf
18:22:35 <int-e> `slwd wisdom.pdf//s=wis=classical wis=
18:22:37 <HackEso> wisdom.pdf//Nicely formatted classical wisdoms and quotes book at https://www.dropbox.com/s/fyhqyvy3i8oh25m/wisdom.pdf
18:22:48 <int-e> ;-)
18:23:12 <b_jonas> yeah
18:42:30 <b_jonas> `? x
18:42:31 <HackEso> x? ¯\(°​_o)/¯
18:42:57 <zzo38> I should think only the https://esolangs.org/logs/ is needed for the logs and others isn't (although maybe that web page should link to other one in case you want to look at older logs), and if it is still too long by one byte, change it to "http:" instead of "https:"
18:43:54 <b_jonas> zzo38: I think fizzie or someone suggested that it's better to have all three in the topic, because there have been times when the esolangs.org logs server was down but one of the others were working
18:44:04 <b_jonas> `? logs
18:44:07 <HackEso> Logs: see channel topic.
18:44:15 <b_jonas> seriously?
18:44:18 <b_jonas> `? log
18:44:19 <HackEso> Logs: see channel topic.
18:45:29 <b_jonas> `slashlearn log//#esoteric channel logs: https://esolangs.org/logs/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D http://codu.org/logs/_esoteric/
18:45:31 <HackEso> Relearned 'log': #esoteric channel logs: https://esolangs.org/logs/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D http://codu.org/logs/_esoteric/
18:45:32 <b_jonas> `? logs
18:45:34 <HackEso> ​#esoteric channel logs: https://esolangs.org/logs/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D http://codu.org/logs/_esoteric/
18:45:47 <b_jonas> could also say something about logarithme
18:45:56 <b_jonas> `? logarithm
18:45:57 <HackEso> logarithm? ¯\(°​_o)/¯
18:46:12 <zzo38> Anyways, the codu.org server does not have recent logs (although tunes.org does, so perhaps should be kept in case esolangs.org is down).
18:47:36 <b_jonas> I'd err in favor of advertising the channel logs well, because one of the network-wide policies on freenode is that if a channel has public logs, then those should be advertised
18:47:43 <zzo38> (The wisdom file should still list all of them though)
18:48:07 <b_jonas> ``` grep -ERiw codu wisdom
18:48:10 <HackEso> wisdom/log:#esoteric channel logs: https://esolangs.org/logs/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D http://codu.org/logs/_esoteric/
18:48:12 <b_jonas> ``` grep -ERiw tunes wisdom
18:48:13 <HackEso> wisdom/log:#esoteric channel logs: https://esolangs.org/logs/ http://tunes.org/~nef/logs/esoteric/?C=M;O=D http://codu.org/logs/_esoteric/
18:48:16 <b_jonas> hmm
18:48:18 <zzo38> Yes, but the codu.org does not have recent logs.
18:48:19 <b_jonas> `log
18:48:19 <HackEso> ​/hackenv/bin/log: 2: cd: can't cd to /var/irclogs/_esoteric \ shuf: '????-??-??.txt': No such file or directory \ ????-??-??.txt:
18:48:19 <b_jonas> `logs
18:48:20 <HackEso> ​/srv/hackeso-code/multibot_cmds/lib/limits: line 5: exec: logs: not found
18:48:35 <b_jonas> huh
18:49:22 <b_jonas> oh, I think `log tries to look up logfiles from back when there was a copy of them in hackeso's filesystem
19:14:10 <arseniiv> I tried to advertize #esoteric logs in my circles but it did no discernible effect
19:14:14 <arseniiv> :P
19:15:27 <shachaf> int-e: So they kept posting my name? What's that about?
19:17:51 -!- ais523 has joined.
19:20:04 <ais523> <int-e> . o O ( can we rate limit edits per user to bursts of 5, 5 per hour? ;-) (I might regret thios...) ) ← there are rate limiting settings that can be used, but I don't think any do that precisely; also, it'd be a problem if people were creating articles via repeated edits
19:20:32 <ais523> I know that what's going on on [[Bitch]] is a trainwreck but it'd likely do too much harm to stop it (besides, it /is/ an interesting language…)
19:24:33 <shachaf> hi ais523
19:25:44 <ais523> hi
19:26:06 <shachaf> Is my request (not to be mentioned in the wiki, even in the history) feasible?
19:26:43 <ais523> it's possible to replace history entries with markers showing that something was deleted from them
19:26:46 <shachaf> This person is obviously doing it on purpose, because they're quoting people talking about how annoying they are on the wiki.
19:26:57 <ais523> we do that in cases where people post personal information to the wiki
19:27:09 <ais523> when someone's just copying public information to the wiki, there's no obvious privacy issue
19:27:20 <ais523> however, if someone is intentionally doing it to annoy someone else that'd be a reason to remove it of its own accord
19:27:26 <shachaf> It's only a strong preference.
19:27:35 <shachaf> But this person is clearly acting in bad faith here.
19:28:04 <ais523> I think A thinks that the wiki works like a nomic, i.e. any rule that isn't explicitly stated doesn't exist
19:28:22 <shachaf> I don't care how they think.
19:29:04 <ais523> in this case I think A's userpage counts as an attack page, because I can't really see a legitimate reason for the way it's being used atm
19:29:44 <ais523> actually doing a revision delete on it is going to be painful, though, because of the way he edits it, with repeated small changes
19:30:21 <shachaf> But not doing it because of that reason is rewarding bad behavior.
19:31:18 <ais523> I'm trying to find the last good revision to revert to now
19:31:18 <shachaf> If repeated small changes are a problem then something should be done about it because they're going to keep doing it.
19:32:08 <ais523> they're not normally a problem, they just make revision deletes slower
19:32:17 <ais523> if the whole thing had been done in one change it'd have been gone by now
19:32:32 <ais523> I'm trying to do the delete at the moment, it just takes a while to do it correctly in this case
19:33:02 <shachaf> OK. Thanks.
19:33:19 -!- Phantom_Hoover has joined.
19:33:32 <esowiki> [[User:A]] https://esolangs.org/w/index.php?diff=62736&oldid=62721 * Ais523 * (-3812) set top revision for revision delete; this is the most recent version of this page that isn't disruptive / an attack on other people
19:34:34 <ais523> shachaf: could you check the history to make sure that there's nothing in the history metadata that you object to? (i.e. not the content of the page, just things like the edit summary, i.e. things directly visible on the "history" page)
19:36:04 <shachaf> It seems fine.
19:36:17 <esowiki> [[Special:Log/delete]] revision * Ais523 * Ais523 changed visibility of 34 revisions on page [[User:A]]: content hidden: attack page / intentional disruption / intentionally antagonizing other people; also off-topic
19:36:18 <shachaf> My direct objection was only to the last few edits, that mention me by name.
19:37:24 <ais523> well, it's been attacking people (both me and others) for a while
19:37:25 -!- ais523 has quit (Remote host closed the connection).
19:37:37 -!- ais523 has joined.
19:37:44 <ais523> <ais523> I should have looked at the history earlier, the most recent revision was toned down a little but some of the earlier ones were obviously unacceptable, it showed intent to antagonise people
19:37:50 <ais523> <ais523> (even the most recent revision was pretty bad though)
19:38:56 <shachaf> I agree.
19:39:24 <ais523> quoting someone because you intend to annoy them is one of the worst reasons to quote…
19:39:40 <ais523> anyway, I think the revision delete worked (although I can't easily check because I'm an admin and can see the deleted text regardless)
19:40:13 <ais523> it's much easier than it used to be, the old method was both confusing to anyone watching and easy to screw up, also it was harder to audit
19:40:37 <Phantom_Hoover> wow, esowiki drama?
19:40:39 <ais523> the newer version lets us mark the history as having been deleted, which is much better for transparency, also the process is a little easier
19:40:55 <ais523> Phantom_Hoover: this happens every now and then
19:41:05 <Phantom_Hoover> i remember when you deleted revisions by deleting the entire page and then only undeleting certain revisions
19:41:09 <ais523> yes, that's the old method
19:41:30 <ais523> it still works but it's easy to screw up and if the page ever gets deleted a second time, all the revisions get mixed up with each other
19:41:45 <ais523> the new method was added a while back because the old method kind-of sucked :-)
19:43:12 <shachaf> The real drama is when it'll turn out A was secretly me all along.
19:44:22 <Phantom_Hoover> and by 'a while back' you mean over ten years ago, right
19:44:35 <Phantom_Hoover> i remember fucking around with revisiondelete when i was a bureaucrat on rationalwiki lol
19:45:06 <kmc> shachaf: https://scontent-sjc3-1.xx.fbcdn.net/v/t1.0-9/61108698_10100593041263622_7774938434182840320_n.jpg?_nc_cat=111&_nc_eui2=AeGvh5fMnqlQ5N0kDXYiXyOUPLFa4QyuR-JWDj2l8sT6urnUDguFAQFLgUKB4mbDweYPObW8r3uqRlhR1yyKhjc928FnroVdeQTQl_wnvoNGzA&_nc_ht=scontent-sjc3-1.xx&oh=2840eb864facd95f2d14a142d5989774&oe=5D5FE019
19:45:15 <ais523> Phantom_Hoover: it's more recent on Esolang because it was a long time before we upgraded and gained the feature
19:45:20 <ais523> it was added to MediaWiki ages ago, though, yes
19:45:22 <shachaf> cat
19:45:37 <shachaf> Do all y'all like cats?
19:46:10 <Phantom_Hoover> people who don't like cats are deeply suspicious
19:46:30 <ais523> <shachaf> The real drama is when it'll turn out A was secretly me all along. ← do you really think A's a sockpuppet of an established #esoteric user? it's technically possible to check, and it would be a major problem if true, but I think it's unlikely on behavioural grounds
19:46:36 <shachaf> I agree.
19:46:56 <shachaf> No, I don't think that, it was a joke.
19:47:07 <shachaf> AlsoI don't know how you'd check whether it actually is me.
19:47:31 <Phantom_Hoover> i feel it was unsporting to delete the drama before giving me a chance to rubberneck :p
19:48:04 <shachaf> Phantom_Hoover: The drama was just a page full of verbatim quotes from IRC of people talking about how annoying A's editing behavior is.
19:48:04 <ais523> well, you could make it very hard to check by consistently using different VPNs, browsers, etc. for different personas
19:48:09 <ais523> but most people who try that screw up at one point
19:48:14 <Phantom_Hoover> lol
19:48:30 <b_jonas> ais523: I tried to get some revisions deleted from hu.wikipedia and wikidata earlier this year. It was about the birthdate of a famous actor who asked to get that removed from the article about her. That actor asked wikipedia to delete it, but
19:48:45 <shachaf> I mean, I don't have a wiki account, and I don't think my IP address is even visible to anyone here.
19:48:48 <ais523> b_jonas: oh, Wikipedia birthday wars are ridiculous
19:49:01 <Phantom_Hoover> i shouldn't be surprised but i am
19:49:09 <shachaf> I guess you could trick me by posting an interesting link in the channel or /msg and hoping that I cilck on it.
19:49:26 <b_jonas> was initially refused, because the birthday was mentioned in some cvs online that were apparently published with her agreement. She got two of those deleted, so the birthday was available only from wikipedia and some archive.org pages.
19:49:53 <ais523> shachaf: I've done that before now, although not to you
19:50:07 <ais523> and not in this channel
19:50:25 <b_jonas> There was some stupid long policy debate where some people tried to defend the idiotic policy that we can list the birthday because it can still be sourced from archive.org. That's obvious nonsense, because then any webpage could refuse to delete the birthday _first_ because it's still on other public webpages.
19:50:26 <ais523> was trying to catch someone who was being disruptive in ##nomic
19:50:36 <Phantom_Hoover> i vaguely remember reading about tactics like that being used on eve online to catch moles
19:51:03 <b_jonas> Eventually after some edit war, the birthday was removed from the article (the month stayed). But it still remained in revision history on hu.wikipedia and wikidata. So I tried to get it deleted.
19:52:33 <b_jonas> That's complicated, because by policy hiding revisions would be the job of oversighters (a privilaged user group), but hu.wikipedia doesn't have any of them, so the fallback is to ask wikimedia stewards.
19:53:00 <ais523> there's two levels of hiding
19:53:03 <b_jonas> I asked wikimedia stewards, they said it's not their job, ask wikimedia legal. I asked wikimedia legal, they told me that they'll delete if the actor asks them directly. At that point I gave that up.
19:53:11 <ais523> I don't think this would warrant oversight, revision delete is surely enough for something like this?
19:54:00 <Phantom_Hoover> maybe he means revisiondelete was lumped into the oversighters in their permissions structure?
19:54:07 <ais523> the only difference is that an oversighting can't be undone by a normal admin, so it's secure against more people
19:54:27 <b_jonas> ais523: https://meta.wikimedia.org/wiki/Oversight_policy suggests that this should be oversighted, but the details of the policy are not clear to me:
19:54:41 <b_jonas> that global policy mentions deleting private data from userpages, not from articles
19:55:45 <b_jonas> but this is also a legal issue, because the laws require that we scrub the personal data if the relevant person asks, and https://foundation.wikimedia.org/wiki/Terms_of_Use says that we have to follow the laws
19:55:54 <b_jonas> which is why the legal dept is relevant
19:56:08 <ais523> aha, was wondering if laws applied to this
19:56:13 <b_jonas> they certainly do
19:56:20 <b_jonas> because this is in europe
19:56:22 <ais523> I guess exact birthdays do count as private personal information
19:57:15 <b_jonas> I can send a link about the discussion in private if you want, but I don't want to have that in the channel logs
19:57:32 <ais523> no, I don't want it
19:57:54 <b_jonas> it also turned out that hu.wikipedia has a local policy saying that we shouldn't list the birthdate if it's not publicly known
19:57:56 <ais523> I've seen enough birthday wars already
19:57:57 <b_jonas> but I found this rather late
19:59:12 <b_jonas> and of course OTRS was involved too, because they're the ones who got the request from the actor, with a claim that she's really that actor
20:00:14 <ais523> <int-e> As long as that works for a large enough proportion of the wikis you find... you won't spend any more effort than that. If it stops working you'll start looking for common countermeasures, like easy captchas. ← there are spamming frameworks that use humans to break captchas, I know this because I've seen the same spam framework hit multiple wikis I admin where user signup worked in totally different ways
20:00:43 <ais523> the whole "introduce yourself" thing was set up because the spam frameworks think "oh, I created an account, I can take over myself rather than using up the human captcha-breaker's time"
20:00:49 <b_jonas> ais523: right, forwarding attacks for captchas
20:01:01 <shachaf> Wait, what? I just discovered Unicode has two code points with the same name.
20:01:08 <shachaf> I always thought code point names were unique.
20:01:16 <ais523> b_jonas: at least one I don't think it was a forwarding attack, because the account creation process was to create an account on a different (related) site
20:01:17 <b_jonas> shachaf: which ones? it's not "BELL" is it?
20:01:21 <shachaf> It's BELL.
20:01:24 <ais523> I don't see how you can do that with a forwarding attack
20:01:50 <b_jonas> ais523: yes, the forwarding is for the generic captchas
20:01:57 <shachaf> Oh, maybe BELL is just labeled "<control>" and the name BELL is an alias?
20:02:26 <b_jonas> shachaf: I think the name of the control character isn't BELL, but I don't know what it is
20:02:28 <ais523> b_jonas: this is the same spammer framework that attacked esowiki
20:02:45 <ais523> IMO the name of the control character is BEL with one L, but I think Unicode disagrees
20:03:03 <b_jonas> isn't the name of the control character something like ALERT ?
20:03:30 <b_jonas> http://unicode.scarfboy.com/?s=U%2B0007 says "Name not known"
20:03:38 <b_jonas> um no, it says "(name not known)"
20:03:47 <b_jonas> maybe it doesn't have a name then?
20:04:04 <shachaf> b_jonas: I think my script extracting data from the official Unicode file took some shortcuts.
20:04:08 <b_jonas> strange
20:04:14 <shachaf> ais523: They also call U+0 "NULL"!
20:05:00 <ais523> ugh, now how are we going to teach people the difference between "\0" and "" and Segmentation fault (core dumped)
20:05:01 <b_jonas> shachaf: yeah, for the full thing you have to use libicu or something, which also dynamically generates the data for the regular korean characters, because there's so many that they didn't want to list it in the source table
20:05:34 <shachaf> My table is also missing a bunch of CJK code points, I'm pretty sure.
20:06:09 <shachaf> https://en.wikipedia.org/wiki/International_Components_for_Unicode says it's "an open-source project of mature C/C++ and Java libraries for Unicode support"
20:06:16 <b_jonas> shachaf: common ones or rare ones?
20:06:34 <shachaf> I don't remember but I believe some common ones are missing.
20:06:38 <shachaf> What's that supposed to mean? Is it C or C++?
20:06:50 <shachaf> I'd like to have a policy of not using C++ libraries.
20:06:52 <ais523> I think they're talking about what languages it binds to
20:06:58 <ais523> it is probably written in the common subset of C and C++, though
20:07:03 <ais523> given that description
20:07:03 <b_jonas> shachaf: yes, and I like it because when I reported a doc feature request, they actually replied to the ticket and fixed it.
20:07:14 <ais523> that's normally a ridiculous language choice but it can make sense for libraries
20:07:19 <shachaf> i,i the common subset of C, C++, and Java
20:07:20 <b_jonas> developers who fix things I report always makes me much more confident about the quality of a library
20:08:10 <b_jonas> shachaf: it has a complete interface for C++, a partial interface for C (some functionality isn't available from the C API, but the most important ones are), and something something Java, I don't follow that
20:08:12 <shachaf> ais523: Apparently it's written in C++ but exposes a C API.
20:08:59 <shachaf> harfbuzz is also written in C++, among other reasons not to use it.
20:09:25 <shachaf> Unfortunately there's a big reason to use it, which is that it's the only cross-platform library that exists that does the thing it does.
20:09:41 <shachaf> And the thing it does is poorly documented so it's quite hard to reimplement.
20:11:32 -!- xkapastel has quit (Quit: Connection closed for inactivity).
20:12:14 <shachaf> Maybe that's also true of ICU?
20:12:29 <shachaf> Golly, handling text correctly is so hard.
20:12:41 <b_jonas> shachaf: most of the times, you don't need the parts of ICU that you can't get easier from smaller libraries
20:12:49 <b_jonas> but for some unicode-related stuff, you do
20:12:58 <b_jonas> there are other libraries handling some of that too, I believe
20:13:19 <b_jonas> because eg. pango somehow knows how to break text into lines at work boundaries, and it doesn't depend on ICU
20:15:50 <int-e> ais523: I know that what's going on on [[Bitch]] is a trainwreck [...] <-- I think the train has moved on to [[Volatile]].
20:16:38 <arseniiv> (I’m slow at puns, but do they call the DEL character DELL also?)
20:18:25 -!- john_metcalf has joined.
20:19:06 <ais523> arseniiv: DEL is a weird character, its original purpose was so that you could delete characters from punched tape by punching out all the remaining holes in that column
20:19:13 <ais523> (the convention being that programs would ignore DEL in their input)
20:19:33 <ais523> likewise, NUL was a no-op so that you could pre-place blocks of NULs in your tape and later punch on them to insert characters
20:19:49 <ais523> but then DEL started to be used as an actual character, e.g. as a representation of the Delete key on the keyboard
20:20:07 <ais523> even though its original purpose means that it has no actual meaning as a terminal control code, and it should logically be treated as a NOP
20:20:30 <b_jonas> yeah, they ignore both NUL (all bits clear) and DEL (all bits set), the latter useful to fix errors (can be done even through a teletype with tape puncher by backspacing and then punching a DEL), the former to leave empty space at the edge of paper tapes so they can be spliced more easily
20:20:31 <arseniiv> didn’t know about Delete representation :o
20:20:41 <ais523> fwiw, most terminals ignore it, which is probably the correct option
20:20:48 <b_jonas> that DEL exists is a main advantage of 7-bit paper tape over 5-bit one
20:20:56 <b_jonas> 5-bit paper tape only has NUL
20:21:12 <ais523> what coding does 5-bit tape use? the main 5-bit code is Baudot but I think it needs 00 for something else
20:21:14 <b_jonas> (to be clear, a bit set means the hole is punched open, a bit cleared means the paper is intact)
20:21:24 <b_jonas> ais523: baudot, yes
20:21:29 <b_jonas> it was used for telex for a long time
20:21:46 <b_jonas> and I don't think it needs code 0 for something else
20:21:55 <b_jonas> it uses code 0 for NUL
20:22:19 <ais523> huh, it doesn't use 0 for something else, I just checked
20:22:31 -!- adu has joined.
20:22:50 <b_jonas> well, some versions do
20:22:53 <ais523> <docs of https://esolangs.org/wiki/7> C<00> encodes NUL,
20:22:55 <b_jonas> there are tons of variants of baudot
20:22:59 <b_jonas> some use it as a space
20:23:06 <ais523> space is 04
20:23:20 <b_jonas> but most don't, exactly so that you can leave empty space at the edge of paper tape
20:23:25 <ais523> (sorry about the leading zeroes, I tend to write Baudot in base 6 out of habit)
20:24:09 <b_jonas> telex machines can transmit and receive much faster than a single operator can type on the keyboard, so for stations that send lots of messages, people punch paper tapes on separate keyboards, and just feed it to the fast tape reader of the telex machine
20:24:10 <arseniiv> hm also it seems a pun about DEL was more of a punch
20:24:50 <b_jonas> a machine with keyboard and paper punch but without the modem is a significantly cheaper than the full telex machine
20:25:02 <b_jonas> arseniiv: (groan)
20:25:31 <b_jonas> ais523: base SIX?
20:25:33 <b_jonas> what
20:26:10 <b_jonas> base 8 I could understand
20:26:13 <ais523> b_jonas: https://esolangs.org/wiki/7 has twelve commands, but only eight of them have names, and only six of them are easy to put in a literal
20:26:35 <ais523> it uses Baudot as its main encoding for "string constants" but it's a variant that only uses the digits 0-5 because 6 and 7 and the four unnamed commands are a pain to use
20:26:47 <ais523> so most of the actual Baudot writing I do nowadays is done in base 6
20:28:00 <arseniiv> b_jonas: (groan) => hehe thanks I knew it would be a quality content
20:29:01 <ais523> or, I should say: 6, 7, and the four unnamed commands aren't bad to use, it's just mixing them with 0-5 that's awkward
20:29:06 <ais523> so literals tend to use one set or the other
20:29:26 * ais523 vaguely wonders why they didn't name the unnamed commands as 8, 9, 10, and 11, there's an obvious order for them
20:30:32 <b_jonas> ais523: I see
20:31:49 <b_jonas> I guess I forgot about the baudot encoding part of that esolang
20:32:16 <ais523> with my current codegolf.stackexchange account, my two highest-scoring posts are both posts in 7 that just print string literals
20:33:13 <ais523> both had [tag:restricted-source] restrictions related to digits, which was part of the insipiration for using 7 in the first place
20:38:19 -!- AnotherTest has quit (Ping timeout: 248 seconds).
20:58:30 -!- ais523 has quit (Quit: quit).
21:03:23 -!- Sgeo_ has quit (Read error: Connection reset by peer).
21:03:50 -!- Sgeo_ has joined.
21:17:10 <zzo38> I suppose an implementation can use numbers 8 9 10 11 for the unnamed commands, which is not exposed to the program.
21:17:32 -!- Sgeo_ has quit (Read error: Connection reset by peer).
21:17:47 -!- Sgeo_ has joined.
21:25:43 -!- LKoen has joined.
21:42:29 <arseniiv> consider placing N points on a sphere such that the sum of all pairwise (Euclidean) distances between them is maximal; for N = 7 numerical optimization gives a somewhat strange arrangement: two opposite points and a “wobbly pentagon” orthogonal to that diameter
21:43:48 <arseniiv> also either it’s a flaw of Mathematica 8 or the solution for N = 20 is not an icosahedron (two quadruplets of points are close to being squares)
21:44:45 <arseniiv> also it’s obvious in retrospect but I was surprised N = 8 gives no cube, but a square antiprism
21:45:51 <b_jonas> arseniiv: you'll have to ask chemistry people about that problem for small N
21:45:57 <arseniiv> these results may be suboptimal, I’ve used standard NMaximize function for that
21:46:24 <b_jonas> it's perhaps not quite equivalent to the natural problem in chemistry, but similar enough
21:48:04 <arseniiv> b_jonas: yeah, I heard about it, there should be an equivalent “a potential field and some charges” formulation
21:48:47 <arseniiv> also I’ve seen a link today regarding IIRC N = 5
21:49:12 <arseniiv> an article which I didn’t read, but they say it’s nontrivial
21:50:22 <arseniiv> with a proof of optimality of the triangle bipyramid arrangement
21:51:16 <arseniiv> found it: https://arxiv.org/abs/0906.0937
21:52:00 <zzo38> You can find the JSON of a c2 wiki page by appending the page name to http://c2.com/wiki/remodel/pages/ in case you want to write your own parser for it and do not want to use theirs.
21:52:12 <arseniiv> ( chemical N = 5 also: https://arxiv.org/abs/1001.3702 )
22:12:55 -!- 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.”).
22:14:53 -!- Sgeo__ has joined.
22:18:10 -!- Sgeo_ has quit (Ping timeout: 272 seconds).
22:37:41 -!- tromp has quit (Remote host closed the connection).
22:38:12 <esowiki> [[Volatile]] M https://esolangs.org/w/index.php?diff=62737&oldid=62731 * JonoCode9374 * (+49) /* Python Interpreter */ Added ability to output as characters
22:54:51 -!- b_jonas has quit (Quit: leaving).
23:07:46 -!- zzo38 has quit (Ping timeout: 246 seconds).
23:09:26 -!- zzo38 has joined.
23:12:51 -!- tromp has joined.
23:17:33 -!- tromp has quit (Ping timeout: 258 seconds).
23:32:16 <esowiki> [[Point]] https://esolangs.org/w/index.php?diff=62738&oldid=62735 * Unlimiter * (+30) /* In-depth */
23:33:09 <esowiki> [[Point]] https://esolangs.org/w/index.php?diff=62739&oldid=62738 * Unlimiter * (+3) /* In-depth */
23:34:11 <esowiki> [[Point]] https://esolangs.org/w/index.php?diff=62740&oldid=62739 * Unlimiter * (+3)
23:36:53 <esowiki> [[Point]] https://esolangs.org/w/index.php?diff=62741&oldid=62740 * Unlimiter * (+13) /* In-depth */
23:39:24 <esowiki> [[Point]] https://esolangs.org/w/index.php?diff=62742&oldid=62741 * Unlimiter * (+0) /* Countdown */
23:39:42 <esowiki> [[Point]] https://esolangs.org/w/index.php?diff=62743&oldid=62742 * Unlimiter * (+17) /* Counting up */
23:50:46 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
←2019-05-25 2019-05-26 2019-05-27→ ↑2019 ↑all