←2017-09-06 2017-09-07 2017-09-08→ ↑2017 ↑all
00:07:53 -!- tromp has quit (Remote host closed the connection).
00:43:37 -!- tromp has joined.
00:48:15 -!- tromp has quit (Ping timeout: 248 seconds).
01:13:15 -!- sleffy has quit (Ping timeout: 252 seconds).
02:05:32 -!- grumble has quit (Ping timeout: 615 seconds).
02:05:41 -!- rumble has joined.
02:31:34 -!- tromp has joined.
02:36:31 -!- tromp has quit (Ping timeout: 248 seconds).
02:50:53 -!- SigmundYx has joined.
02:54:06 -!- oerjan has joined.
02:58:56 <oerjan> @messages-loud
02:58:56 <lambdabot> int-e said 16h 37m 50s ago: @hoogle not responding at all is strange indeed, but I have no clue why; no suspicious OOM kills, and it's running locally; maybe an IO bottleneck and resulting timeout...
02:58:56 <lambdabot> how would one diagnose this
02:59:15 <oerjan> @tell it actually did respond with a list of packages, just not the actual functions.
02:59:16 <lambdabot> Consider it noted.
02:59:19 <oerjan> oops
02:59:26 <oerjan> @tell incomprehensibly it actually did respond with a list of packages, just not the actual functions.
02:59:26 <lambdabot> Consider it noted.
02:59:29 <oerjan> argh
02:59:38 <oerjan> @tell int-e it actually did respond with a list of packages, just not the actual functions.
02:59:38 <lambdabot> Consider it noted.
03:00:12 <oerjan> @tell incomprehensibly ...tab completion error.
03:00:12 <lambdabot> Consider it noted.
03:02:39 -!- ATMunn has quit (Quit: See ya! o/).
03:06:06 -!- SigmundYx has quit (Quit: See Ya.).
03:06:22 -!- SigmundYx has joined.
03:09:03 -!- tromp has joined.
03:13:35 -!- tromp has quit (Ping timeout: 240 seconds).
03:41:32 <HackEgo> [wiki] [[Drawkcab]] M https://esolangs.org/w/index.php?diff=53052&oldid=52138 * HereToAnnoy * (+0) C++ ----> ++C
04:01:36 -!- sleffy has joined.
04:04:58 -!- erkin has quit (Quit: Ouch! Got SIGABRT, dying...).
04:11:36 -!- SigmundYx has quit (Remote host closed the connection).
04:13:10 -!- SigmundYx has joined.
04:25:59 -!- ais523 has joined.
04:56:55 -!- tromp has joined.
05:01:35 -!- tromp has quit (Ping timeout: 240 seconds).
05:04:31 -!- SigmundYx has quit (Remote host closed the connection).
05:19:54 -!- newsham has quit (Read error: Connection reset by peer).
05:20:11 -!- newsham has joined.
05:20:16 -!- Gregor has quit (Ping timeout: 248 seconds).
05:20:17 -!- j-bot has quit (Ping timeout: 248 seconds).
05:21:52 -!- Warrigal has quit (Ping timeout: 248 seconds).
05:21:59 -!- Warrigal has joined.
05:24:27 -!- zzo38 has quit (Ping timeout: 260 seconds).
05:27:00 -!- SigmundYx has joined.
05:27:49 -!- Gregor has joined.
05:32:40 -!- FireFly has quit (Quit: Goodbye).
05:44:58 -!- FireFly has joined.
05:51:45 -!- tromp has joined.
05:55:47 -!- tromp has quit (Ping timeout: 246 seconds).
06:13:20 -!- jaboja has quit (Remote host closed the connection).
06:37:29 -!- SigmundYx_ has joined.
06:38:49 -!- Hoolootwo has changed nick to Hooloovo0.
06:39:24 -!- SigmundYx has quit (Ping timeout: 252 seconds).
06:45:36 -!- tromp has joined.
06:47:16 -!- imode has joined.
06:50:05 -!- tromp has quit (Ping timeout: 240 seconds).
07:03:47 -!- FreeFull has quit.
07:34:47 -!- imode has quit (Ping timeout: 255 seconds).
07:40:17 -!- tromp has joined.
07:44:47 -!- tromp has quit (Ping timeout: 248 seconds).
07:58:27 -!- sleffy has quit (Ping timeout: 260 seconds).
08:04:48 -!- sleffy has joined.
08:07:16 <int-e> oerjan: uh, weird
08:07:45 <int-e> am I to conclude that the command line hoogle is not deterministic?
08:11:40 -!- AnotherTest has joined.
08:15:44 -!- SigmundYx_ has quit (Remote host closed the connection).
08:23:09 -!- imode has joined.
08:26:15 -!- tromp has joined.
08:37:10 <oerjan> int-e: well it wasn't on that day, anyway.
08:37:47 <shachaf> It's good enough if it's deterministic some of the time.
08:38:16 <shachaf> (This was acknowledging oerjan's joke, not missing it and then making the same joke.)
08:57:48 <int-e> oerjan: I don't know anything tangible that changed between then and now
08:59:19 <int-e> (nor did I actually observe the behavior you describe, but that's a smaller issue)
09:06:23 <oerjan> hm perhaps i should paste my log
09:07:38 <oerjan> 08:35 =oerjan> @hoogle a->a
09:07:38 <oerjan> 08:35 =lambdabot> package base
09:07:38 <oerjan> 08:35 =lambdabot> package bytestring
09:07:38 <oerjan> 08:35 =lambdabot> package containers
09:08:08 <oerjan> int-e: it gave the same response to every type i tried
09:09:08 <oerjan> this "worked" though:
09:09:10 <oerjan> 08:34 =oerjan> @hoogle f::(a->b)->(a,c)->(b,c)
09:09:10 <oerjan> 08:34 =lambdabot> Formatting.ShortFormatters f :: Real a => Int -> Format r (a -> r)
09:09:13 <oerjan> 08:34 =lambdabot> Turtle.Format f :: Format r (Double -> r)
09:09:16 <oerjan> 08:34 =lambdabot> Debug.SimpleReflect.Vars f :: FromExpr a => a
09:09:24 <int-e> hmm
09:09:29 <int-e> @hoogle a -> a
09:09:29 <lambdabot> Prelude id :: a -> a
09:09:29 <lambdabot> Data.Function id :: a -> a
09:09:29 <lambdabot> GHC.Exts breakpoint :: a -> a
09:09:32 <int-e> @hoogle a->a
09:09:32 <lambdabot> package base
09:09:32 <lambdabot> package bytestring
09:09:32 <lambdabot> package containers
09:09:38 <int-e> some parsing problem
09:09:42 <oerjan> huh
09:09:55 <oerjan> shachaf: false alarm, it was deterministic anyhow
09:10:30 <int-e> I'll try a newer hoogle version soonish, though it may have to wait until october
09:10:35 <oerjan> oh hm
09:11:07 <oerjan> @hoogle a
09:11:07 <lambdabot> Text.Blaze.Html4.FrameSet a :: Html -> Html
09:11:07 <lambdabot> Text.Blaze.Html4.Strict a :: Html -> Html
09:11:07 <lambdabot> Text.Blaze.Html4.Transitional a :: Html -> Html
09:11:25 <oerjan> int-e: maybe it simply treats everything without a space as an ident
09:11:34 <oerjan> @hoogle Int
09:11:35 <lambdabot> Prelude data Int
09:11:35 <lambdabot> module Data.Int
09:11:35 <lambdabot> Data.Int data Int
09:11:54 <int-e> @hoogle a ->a
09:11:55 <lambdabot> Prelude id :: a -> a
09:11:55 <lambdabot> Data.Function id :: a -> a
09:11:55 <lambdabot> GHC.Exts breakpoint :: a -> a
09:11:58 <int-e> (not quite)
09:12:42 <int-e> anyway, this is Hoogle 5.0.9
09:13:40 <int-e> 5.0.13 apparently fixes this
09:13:45 <oerjan> aha
09:14:07 <int-e> https://github.com/ndmitchell/hoogle/issues/219
09:14:29 -!- tromp has quit (Remote host closed the connection).
09:15:52 <int-e> so let's just try installing the new version
09:18:39 -!- oerjan has quit (Quit: Nite).
09:23:39 <int-e> @hoogle a->a
09:23:40 <lambdabot> Prelude id :: a -> a
09:23:40 <lambdabot> Data.Function id :: a -> a
09:23:40 <lambdabot> GHC.Exts breakpoint :: a -> a
09:25:26 <int-e> so what is the significance of those 'package' results (if they are results at all...)?
09:34:36 -!- SigmundYx has joined.
09:34:53 -!- SigmundYx_ has joined.
09:35:47 -!- tromp has joined.
09:39:22 -!- sleffy has quit (Ping timeout: 260 seconds).
09:40:38 -!- SigmundYx has quit (Remote host closed the connection).
09:42:48 <Hooloovo0> a -> a
09:48:27 -!- SigmundYx_ has quit (Remote host closed the connection).
09:58:43 -!- Sgeo has quit (Ping timeout: 246 seconds).
10:19:08 -!- augur has quit (Remote host closed the connection).
10:24:44 -!- Sgeo has joined.
10:30:08 <b_jonas> lol. today's first SMBC cracks me up http://www.smbc-comics.com/comic/pictograms
10:30:24 -!- imode has quit (Ping timeout: 252 seconds).
10:33:39 -!- LKoen has joined.
10:38:01 -!- augur has joined.
10:39:19 <Hooloovo0> that's not even a chimp
10:40:06 <Hooloovo0> I guess it could be a very stylized one but pls
10:41:16 <Taneb> Hooloovo0, are you accusing Zach Weinersmith of being able to draw and choosing not to
10:41:57 <Hooloovo0> um
10:42:02 <Hooloovo0> sure
10:42:16 <Hooloovo0> thought it was a gorilla at first
10:42:26 -!- augur has quit (Ping timeout: 255 seconds).
10:42:48 -!- TheRegulator has joined.
10:47:05 -!- TheRegulator has left.
11:04:43 -!- Melvar` has joined.
11:05:34 -!- Melvar has quit (Ping timeout: 240 seconds).
11:09:30 -!- ais523 has quit.
11:48:06 <b_jonas> Hooloovo0: I think misrepresenting the visual appearance of chimps is a noble tradition from Irregular Webcomic.
11:48:11 <b_jonas> But yes, it doesn't look like a chimp.
11:50:09 <Hooloovo0> us that a meme?
11:50:38 <Hooloovo0> in the more technical sense o the term, that is
11:50:52 <b_jonas> Hooloovo0: maybe just a callback
11:51:01 <Hooloovo0> ah, ok
11:51:24 <b_jonas> but more likely it's not intentional at all, only I'm connecting it, and Zach just doesn't care too much about drawing accurately as long as he can get the meaning through
11:51:47 <b_jonas> they're just funny comic strips, not some hard sci-fi stories
12:19:41 -!- j-bot has joined.
13:02:04 -!- heroux has quit (Ping timeout: 240 seconds).
14:06:51 -!- rumble has changed nick to grumble.
14:14:39 -!- ATMunn has joined.
14:17:16 <HackEgo> [wiki] [[Special:Log/newusers]] create * Therrore * New user account
14:49:23 -!- erkin has joined.
15:01:58 -!- Mayoi has joined.
15:04:05 -!- erkin has quit (Ping timeout: 240 seconds).
15:06:00 -!- Mayoi has changed nick to erkin.
15:18:38 -!- ATMunn has quit (Quit: .).
15:48:34 -!- Bowserinator has quit (Ping timeout: 240 seconds).
15:54:24 -!- Bowserinator has joined.
15:54:47 -!- Bowserinator has changed nick to Guest23223.
16:26:07 -!- xkapastel has joined.
16:51:28 -!- AnotherTest has quit (Ping timeout: 240 seconds).
16:54:34 -!- jix has quit (Ping timeout: 240 seconds).
16:58:08 -!- jix has joined.
17:02:35 -!- jix has quit (Ping timeout: 240 seconds).
17:04:17 -!- jix has joined.
17:09:04 -!- FreeFull has joined.
17:18:47 -!- tromp has quit (Remote host closed the connection).
17:42:54 -!- tromp has joined.
17:53:06 -!- tromp has quit (Remote host closed the connection).
17:54:14 -!- augur has joined.
17:56:16 -!- tromp has joined.
17:59:13 -!- augur has quit (Ping timeout: 248 seconds).
18:02:08 -!- Melvar` has changed nick to Melvar.
18:04:13 -!- ATMunn has joined.
18:13:14 -!- augur has joined.
18:17:41 -!- tromp has quit (Remote host closed the connection).
18:18:19 -!- Phantom_Hoover has joined.
18:18:48 -!- imode has joined.
18:26:20 -!- mroman has joined.
18:26:41 <mroman> evening lads
18:27:09 -!- erkin has quit (Remote host closed the connection).
18:31:43 -!- augur has quit (Remote host closed the connection).
18:32:29 -!- augur has joined.
18:32:36 <shachaf> `5 w
18:32:42 <HackEgo> 1/2:brevity//syn. "shortness" \ iwc//iwc contains puns! Puns galore! Puns after puns after puns! Also science! \ cocoon//Cocoon was built by the fal'Cie, and floats above Gran Pulse. \ it'//It's written with an apostrophe. \ `fetch//`fetch [<output-file>] <URL> downloads files, and is the only web access currently available in HackEgo. It
18:32:46 <shachaf> `n
18:32:46 <HackEgo> 2/2: is a special builtin that cannot be called from other commands. See also `edit.
18:37:12 -!- augur has quit (Ping timeout: 260 seconds).
18:38:44 -!- tromp has joined.
18:44:29 -!- erkin has joined.
18:51:43 -!- AnotherTest has joined.
18:55:03 -!- `^_^v has joined.
18:56:54 -!- tromp has quit (Remote host closed the connection).
18:59:20 -!- zseri has joined.
18:59:51 -!- tromp has joined.
19:01:02 -!- zseri_ has joined.
19:02:35 -!- zseri__ has joined.
19:04:02 -!- zseri has quit (Ping timeout: 260 seconds).
19:05:21 -!- LKoen has quit (Remote host closed the connection).
19:06:18 -!- zseri_ has quit (Ping timeout: 252 seconds).
19:06:35 <\oren\> ok, I think I'm getting to the bottom of this
19:07:29 -!- ATMunn has quit (Quit: ¯\_(ツ)_/¯).
19:10:17 -!- tromp has quit (Remote host closed the connection).
19:13:59 <\oren\> AAAAUAAUAGHH it froze again
19:24:26 -!- jaboja has joined.
19:39:58 -!- tromp has joined.
19:40:26 -!- jaboja has quit (Read error: Connection reset by peer).
19:41:09 -!- jaboja has joined.
19:44:33 -!- tromp has quit (Ping timeout: 240 seconds).
19:53:02 -!- zseri_ has joined.
19:56:33 -!- zseri__ has quit (Ping timeout: 240 seconds).
20:03:52 -!- jaboja has quit (Ping timeout: 246 seconds).
20:11:30 -!- sleffy has joined.
20:13:32 -!- jaboja has joined.
20:15:03 -!- wob_jonas has joined.
20:15:34 <wob_jonas> hi guys
20:16:41 <wob_jonas> \oren\: bottom of what, and what froze? a drink?
20:16:43 -!- Phantom_Hoover has quit (Ping timeout: 255 seconds).
20:17:50 <wob_jonas> how are you all today?
20:18:08 -!- Guest23223 has changed nick to Bowserinator.
20:18:14 -!- Bowserinator has quit (Changing host).
20:18:14 -!- Bowserinator has joined.
20:22:44 -!- Phantom_Hoover has joined.
20:31:11 <mroman> ah rats
20:31:42 <mroman> battey went out during creation of my recovery stick.
20:31:48 <mroman> 3h wasted.
20:32:02 <mroman> and this keyboard can't really tolerate fast writing.
20:32:11 <mroman> *typing
20:32:35 <mroman> but it'll do for travelling I guess
20:34:10 <wob_jonas> mroman: whoa, what keyboard is that? Is it some really old device like a ZX Spectrum? Or some of this fancy modern software like Windows or mobile phone stuff that runs on fast modern hardware but still manages to have reaction times so slow that your keypresses appear on screen only a second later?
20:34:36 <wob_jonas> (Not that I haven't seen that happen on Linux too, but it's still certain bad software that causes that.)
20:34:44 -!- tromp has joined.
20:35:04 <wob_jonas> Anything in between can easily tolerate much faster writing than basically any human can manage.
20:36:14 <wob_jonas> I guess technically there's also old slow modems with a low communication throughput limit, and old typewriter terminals where the mechanics of the printer limit the speed, but even most of those are fast enough for typical human typing.
20:37:58 <mroman> wob_jonas: It's a mini notebook
20:38:03 <mroman> fairly modern.
20:38:18 <wob_jonas> probably a software problem then
20:38:27 <wob_jonas> unless your typing has to go through network without local echo
20:39:07 -!- tromp has quit (Ping timeout: 260 seconds).
20:39:50 <mroman> Im' probably not usde to hit this hard
20:40:07 <mroman> and my timing si probably of because hte distance between keys is'nt exaclty thes ame anymore
20:40:23 <mroman> but this si waht it look like if i try to type fash
20:40:43 <mroman> It's a very narrow keyboard.
20:41:40 <wob_jonas> my typing gets bad when I'm angry and doing heated debate flames on internet. you can sometimes see that on IRC, even on this channel. it doesn't quite look like that though.
20:42:30 <mroman> https://www.notebookcheck.com/Test-Medion-Akoya-E2215T-Convertible.188012.0.html <- a version of this one.
20:42:50 <wob_jonas> also, funnily, I have difficulty typing slowly. I think it's the centipede problem: I can type English just fine when I do it at the usual fast speed, but when I try to slow down to pay attention to the individual keypresses, I get all confused and forget how to type.
20:43:14 <wob_jonas> If I try to think of where each individual letter key is, I can only tell that very slowly, even though I press those letters in normal typing easily.
20:43:24 <mroman> that's normal.
20:43:40 <mroman> it happens with pin codes for debit cards etc. as well
20:43:57 <wob_jonas> And passwords of other sorts, yes.
20:44:22 <wob_jonas> I also have difficulty walking slowly, or also walking anything between my two habitual walking paces, the normal fast one and a slower one.
20:44:42 <wob_jonas> They differ in movement, not only speed, which is why I can't just interpolate between.
20:45:14 <wob_jonas> (The slower pace is also slightly more difficult than the faster pace, at least dexterity wise, not necessarily in energy use.)
20:45:40 -!- xkapastel has quit (Quit: Connection closed for inactivity).
20:45:49 <wob_jonas> wtf. they reskinned the library search webpage of FSzEK.
20:46:22 <wob_jonas> now it has these input boxes with custom borders, not just ordinary browser input controls with the height forced too small
20:47:06 <wob_jonas> I don't understand that stupid fashion. Even Wikipedia bought into it with checkboxes. I have just fine form controls in my fucking browser, there's no need to reimplement them with something worse.
20:48:01 -!- zseri_ has changed nick to zseri.
20:48:04 <wob_jonas> (At least the custom input boxes usually only differ in appearance, they still use the browser's normal text input. It's much worse when a websites reimplements even the behaviour of control, much worse than the browser.)
20:58:43 <mroman> nope
20:59:31 <mroman> Because they don't offer any functionality browsers implemented over the years
20:59:47 <mroman> such as remembering what you entered when you press backwards or F5
21:01:38 <wob_jonas> F5? wtf. you press alt-down to bring up the history dropdown. but the javascript implementations also do remember what you typed and offer suggestions, but often with a worse interface, and sometimes in a way that both the browser's dropdown and the custom one appears at the same time and conflict with each other when you try to choose an entry fro
21:01:38 <wob_jonas> m either dropdown.
21:01:47 -!- xkapastel has joined.
21:02:36 <wob_jonas> You press F5 for reloading the page. Or control-R. Either works in all current browsers, the two different shortcuts originate from some difference in two popular competing browsers ages ago.
21:03:59 <wob_jonas> That's also why you enter the address bar with either control-L (which straight up activates it) or F6 (which cycles between the webpage, the address bar, and the sidebar in typical Windows fashion).
21:04:30 <shachaf> hob_jonas
21:04:37 <wob_jonas> hi shachaf
21:04:47 <shachaf> My old laptop's keyboard would send F6 if you pressed three keys simultaneously.
21:04:51 <shachaf> I think it was either hjk or jkl
21:05:07 <wob_jonas> I have a crazy idea, related to esolang community
21:07:05 <wob_jonas> shachaf: strangely, some keyboards or software start autorepeating a sequence of multiple letters if you hold down multiple letter keys long enough, so you get something like eg. "hkjhkjhkjhkjhkjhkjhkjhkjhkjhkjhkjhkjhkj". I could do it with even six letters in some configuration. This seems new to the last five years, and I don't understand why any
21:07:05 <wob_jonas> hw or sw would do that. Why would it autorepeat more than one key?
21:07:36 <shachaf> wob_jonas: Well, touchpads and touchscreens are now "multitouch"
21:07:42 <shachaf> So it makes sense to extend the same to keyboards.
21:08:05 <wob_jonas> Also, some keyboards just omit a key if you press certain combinations of too many keys at the same time and at least two of them aren't shift keys, but this is sort of an understandable limitation of the electronic wiring.
21:08:32 <mroman> f5 is refresh in pretty much all browsers.
21:08:35 <wob_jonas> shachaf: typing all letters once when you press the keys simultaneously does make sense. It's autorepeating more than two that doesn't make sense to me.
21:08:49 <wob_jonas> mroman: yes, that's what I said
21:09:34 <mroman> this notebook actually has touchscreen as well
21:09:51 <wob_jonas> shachaf: in particular, I used a workstation where I couldn't type leftshift+capslock+backtick, which is where "Í" is assigned on my strange keyboard layout. Luckily that's a rare letter.
21:10:43 <mroman> but the keyboard is definitely a negative point.
21:11:07 <wob_jonas> saner keyboards and software, like this home machine I'm using now, doesn't do either. it allows basically any number of keys simultaneously, and doesn't try to autorepeat multiple keys at the same time
21:11:37 <wob_jonas> mroman: um, the keyboard is a negative point in that notebook in what sense?
21:11:49 <mroman> it's not sensitive enough.
21:12:08 <mroman> you can tap certain keys and they go down without registering a keystroke
21:12:14 <wob_jonas> (also, touchscreens for ordinary computers, eww, I hate those)
21:12:28 <mroman> so you need to press firmer and hold the key down a bit longer than on other keyboards.
21:12:42 <mroman> meaning typing at 120WPM really sucks with this keyboard.
21:12:54 <wob_jonas> (there's certain reasons to use it for some special applications, like in publically accessable terminals that anyone can access. even there I don't like them, but I admit they have some advantages.)
21:13:03 <mroman> by which I mean: you can't type over 80WPM on this keyboard.
21:13:24 <wob_jonas> mroman: "can tap certain keys and they go down without registering a keystroke" ouch. a keyboard shouldn't do that.
21:13:37 <wob_jonas> that's like mechanical typewriters
21:14:02 <wob_jonas> plug in an external keyboard as a workaround
21:14:26 <mroman> I think it only registers it if you press down exactly in the middle
21:14:41 <wob_jonas> that's even worse
21:14:48 <mroman> if I press on the corner it won't register anything.
21:15:06 <wob_jonas> how did you get that notebook? did your job lend it to you for work?
21:15:17 <mroman> I bought it in a store.
21:15:24 <mroman> For 200 bucks.
21:15:34 <mroman> so it's a really cheap one.
21:15:38 <wob_jonas> what does the keyboard feel like physically? is it like cheap normal computer keyboards, or horrible rubber keys, or something else?
21:16:12 <mroman> like a regular notebook keyboard.
21:17:54 -!- zseri has quit (Quit: Leaving).
21:18:33 <mroman> And there's like 10GB usable space
21:18:38 <mroman> clerk said there's 20GB
21:18:43 <wob_jonas> ah
21:19:18 <wob_jonas> is the storage device 20 GB but filled with 10 GB of junk that comes with the system installation or later automatic downloads and caches and temporary files and log files?
21:23:24 <mroman> need to analyse that first
21:23:27 -!- augur has joined.
21:23:48 <wob_jonas> fdisk
21:24:16 <wob_jonas> unless it has tricky firmware hiding some of the storage, in which case it's much harder
21:24:34 <mroman> yeah
21:24:42 <mroman> this shit should be regulated
21:26:05 <wob_jonas> what shit? shipping mobile phones with software full of security hole, even when they aren't deliberate holes for bad reasons but just plain bad development, with updates that make the original android OS even worse, and with no support for downstream security updates later, or no support at all half a year after people buy the phone model?
21:26:16 <wob_jonas> or false advertisment?
21:26:28 <mroman> it's a windows
21:26:43 <mroman> but bundling products with crap should be regulated yes
21:26:56 <mroman> like pre-installing useless crapware
21:27:06 -!- augur has quit (Remote host closed the connection).
21:27:12 <mroman> or stuff like anti-virus products that can't be uninstalled
21:28:01 <mroman> I'd actually be in favor of prohibiting pre-installing anything EXCEPT the OS
21:28:43 <mroman> it's just a scam to push useless software to unknowing customers.
21:29:18 <wob_jonas> windows might be the same, I'm just less familiar with it than the android crap
21:30:18 <wob_jonas> luckily mostly only from other people's devices and second hand accounts. so far I've managed to avoid owning any of this smartphone stuff myself, and even my next phone, which I'll probably buy within a year, will be a non-smartphone
21:30:36 <wob_jonas> (I like my current phone, but electronics does age)
21:31:28 <wob_jonas> (Also now I want more realiability, so with how long this one lives, I might actually buy a new one BEFORE it breaks down or gets stolen, to avoid even a day of downtime.
21:31:29 <wob_jonas> )
21:33:01 -!- augur has joined.
21:33:38 <mroman> meh
21:33:48 <mroman> It'll do the job.
21:33:57 <mroman> but I already have a normal notebook
21:34:14 <mroman> so I didn't want to invest 1200CHF to buy a good mini notebook
21:34:23 <mroman> I just bought the cheapest one they had :D
21:35:38 <wob_jonas> sure, I understand not wanting to buy something more expensive. that's why I specifically said external keyboard as workaround, because that's both cheap and simple
21:35:54 <mroman> no space fo external keyboard :)
21:35:55 <wob_jonas> there's a huge market of external keyboards with lots of variation. obviously most of it is crap.
21:36:07 <wob_jonas> but keyboards are easier to test in shops than notebooks.
21:36:23 <wob_jonas> no space? like, holding the notebook in your lap on a train?
21:36:36 <mroman> backpack
21:36:39 <mroman> but yes
21:36:45 <mroman> this is going to be my travel notebook
21:36:49 <wob_jonas> I see
21:37:13 <mroman> my backpack is now complete and ready
21:37:27 -!- augur has quit (Ping timeout: 260 seconds).
21:37:34 <mroman> a bit on the heavy side though
21:37:35 <wob_jonas> I remember when I was traveling to the airport in Sweden this year, by train. The train had nice tables between the seats. Three people sat together with me, and all three were busy apparently working on a large notebook each.
21:37:53 <mroman> but temperatures are anywhere from -15 to +35C here
21:37:53 <wob_jonas> They didn't seem to know each other, so it looked as if everyone independently chose to work on the train.
21:38:02 <mroman> so I had to pack winter stuff as well
21:38:16 <wob_jonas> (Only the ones sitting close to me though, not everyone else on the train.)
21:39:07 <mroman> after my MRI I'm probably going to vanish for some time
21:39:10 <wob_jonas> When I travel within Hungary, I don't often see large notebooks. I see people pushing the touchscreen on tiny smartphone displays, with often a broken display.
21:39:18 <mroman> unless the MRI actually shows something.
21:39:38 <wob_jonas> I saw a girl with a nintendo DS on the bus at one point, it stuck in my memory because it was so unusual.
21:39:38 <mroman> but I don't have the will to go from doc to doc anymore.
21:39:46 <wob_jonas> vanish where?
21:39:58 <mroman> I don't know.
21:40:02 <wob_jonas> ...
21:40:04 <wob_jonas> that sounds scary
21:40:12 <mroman> That's why I packed for every temperatue range.
21:40:13 <wob_jonas> you don't want to get yourself committed, do you?
21:40:21 <wob_jonas> is it just some vacation?
21:40:31 <wob_jonas> an extended vacation that is?
21:40:37 <wob_jonas> you want to leave the environment of your normal home?
21:41:35 <mroman> I don't know.
21:41:46 <wob_jonas> even more scary
21:41:49 <wob_jonas> I'm worried
21:41:52 <mroman> Well
21:42:06 <mroman> what are you supposed to do otherwise?
21:42:18 <wob_jonas> I don't know
21:42:25 <mroman> scary.
21:42:27 <wob_jonas> and I don't know your circumstances
21:42:56 <wob_jonas> I can enjoy vanishing for a vacation, but only for two weeks or so
21:43:03 <wob_jonas> some people do it for much longer
21:43:24 <mroman> I'm basically ill 3/4 of the time
21:43:27 <mroman> like influenza ill
21:43:31 <mroman> but it's not influenza
21:43:32 -!- LKoen has joined.
21:43:50 <mroman> so I have very reduced capacity to actually do stuff
21:44:02 <wob_jonas> someone from our school had enough of everything at some point and suddenly vanished into Nepal, herding yaks on a bicycle or something, and returned a year later. or something. that's a bit distorted version of the rumour, but still.
21:44:06 <mroman> which means rather poor quality of life
21:44:16 <wob_jonas> yeah, I also have that, but not that badly
21:44:31 <wob_jonas> slightly ill all the time, seriously ill twice a year, reduced work capacity during
21:45:02 <wob_jonas> just took a four month long leave from work, but didn't really vanish, not from work, not from my family or the internet
21:45:11 <wob_jonas> (not for longer than a week at once at least)
21:45:39 <wob_jonas> I'm too stressed, and also don't do enough sports and healthy eating and all that stuff, and it's a self-reinforcing bad cycle
21:45:56 <mroman> also preinstalled software is usually demo 30 days
21:45:59 <wob_jonas> after a while I get overweight, which makes my health worse, my future health expectations much worse, but also makes it more difficult to break the habits and do more sports
21:46:03 <mroman> but it's not removed after 30 days
21:46:08 <mroman> like mcaffee
21:46:15 <mroman> it'll consume ressources even after that
21:46:18 <wob_jonas> it's seriously worrying me, but hard to change
21:46:21 <mroman> and that should be even more illegal.
21:46:34 <wob_jonas> so be careful
21:46:38 <mroman> I can barely work
21:46:55 <wob_jonas> I hope at least you also do have a family supporting you, and that you don't vanish from them at least
21:46:57 -!- jaboja has quit (Ping timeout: 248 seconds).
21:46:59 <mroman> It's like I only have 20% of my cognitive potential avalable
21:47:05 <mroman> yeah no
21:47:10 <wob_jonas> as in, keep a mobile phone
21:47:12 <mroman> no family suppoting me
21:47:21 <mroman> weird relationship
21:47:46 -!- jaboja has joined.
21:47:53 <wob_jonas> yes, my memory and mind is getting worse. that's the really scary part, from the inside even more so than from the outside I think, because my family and coworkers can see how I'm fat and unhealthy and can't work as much, but I see that and also how I can't think normally
21:48:15 <wob_jonas> I'm not saying my family isn't a "weird relationship", but their support still means a lot for me
21:48:29 <wob_jonas> if you have no family, then you should strive to make one, even if they're not blood relations
21:48:35 <wob_jonas> but friends
21:48:56 <mroman> I can hardly look my parents in the eyes
21:49:02 <mroman> and sometimes I can't talk to them
21:49:11 <mroman> as in: selective mutism can't talk
21:49:26 <wob_jonas> original family is great default in some cases, but I mostly care about them too because they care about me, all they did for me in the past when they were younger and more able to help, but also how they always try support me now as much as they can
21:49:40 -!- ^v^v has changed nick to ^v.
21:49:58 <mroman> and I can't be comitted :D
21:50:09 <mroman> theoretically yes
21:50:12 <mroman> practically no
21:50:23 <wob_jonas> that... can be difficult. I hope they can work it around. my mother always seems to somehow correctly guess most of what happens to me, even if I try to not tell everything. it's hard to prove, but I think he has some secret superpowers.
21:50:31 <wob_jonas> (the rest of my family doesn't have that)
21:51:14 <wob_jonas> I hope you at least have enough money to be able to support yourself for a break from work while you regenerate yourself
21:51:22 <mroman> my last clinic stay gave me PTSD
21:51:23 <mroman> so
21:51:26 <wob_jonas> I simply couldn't keep up working all the time
21:51:40 <mroman> they'd have a _really_ hard time keeping me there.
21:52:09 <wob_jonas> mroman: wait, so you say you can't be committed because staying in the clinic has a terrible effect on you? I don't think that's how institution works, but I don't really know
21:52:18 <mroman> no.
21:52:23 <wob_jonas> not that I don't believe you can't be committed, only that reason seems odd
21:52:48 <mroman> any doctor can legally committ any random person for 24h no questions asked
21:53:37 <mroman> the questios is whether any random person would comply.
21:53:51 <wob_jonas> also, thanks for telling about your situation, it (and other stories on #esoteric ) shows me I'm not alone in some sense
21:54:03 <mroman> theoreticaly the police can enforce admission of course
21:54:19 <mroman> it's just
21:54:24 <wob_jonas> as for that, I don't know how they keep people in against their will, but they seem good at it. maybe that only works for certain types of people.
21:54:42 <alercah> mroman: are you open to a discussion by PM?
21:54:59 <mroman> oh.
21:55:24 <wob_jonas> well, technically, I do have some ideas from certain terrible stories my mother told about people who really had to be kept in for a good reason, but I mostly blocked them from my mind
21:55:41 <wob_jonas> I try not to pay too much attention to most details of my mother's work, it has a bad effect
21:56:08 <wob_jonas> I only listen to the part of how it affects her outside her works, like a certain workplace accident when he got his hand bitten, strongly, by a kid
21:56:33 <wob_jonas> and how overloaded she is with her work because she just can't stop accepting more and more work all the time
21:56:41 <wob_jonas> she's so dedicated
22:02:41 <wob_jonas> http://dlib.net/ => oh wow. this sounds nice. if it actually lives up to those promises in the front page, then I'll like this library. it reminds me to other well maintained software that I also like.
22:03:47 <wob_jonas> (some other such software are http://eigen.tuxfamily.org/ and http://sqlite.org/ . if you already know this dlib thing, and like it, check out those.)
22:07:54 <wob_jonas> wtf, did they change the shortcut for the bookmark manager too? I swear firefox keeps scrambling keyboard shortcuts lately. they've changed the shortcut for the download list twice, between control-J, control-Y, and control-shift-something, and now the bookmarks
22:10:52 -!- ATMunn has joined.
22:11:11 -!- tromp has joined.
22:15:57 -!- tromp has quit (Ping timeout: 260 seconds).
22:17:46 -!- tromp has joined.
22:30:30 -!- Mayoi has joined.
22:33:19 -!- erkin has quit (Ping timeout: 248 seconds).
22:35:13 -!- `^_^v has quit (Quit: This computer has gone to sleep).
22:40:27 -!- ais523 has joined.
22:40:38 -!- AnotherTest has quit (Ping timeout: 246 seconds).
22:43:02 -!- ais523 has quit (Client Quit).
22:43:12 -!- ais523 has joined.
22:44:13 -!- AnotherTest has joined.
22:55:23 -!- augur has joined.
22:56:24 -!- ais523 has quit (Remote host closed the connection).
22:56:45 -!- ais523 has joined.
22:59:05 -!- AnotherTest has quit (Ping timeout: 240 seconds).
23:00:29 <wob_jonas> shachaf: wait, I wanted to tell about my stupid esolang idea
23:00:35 <wob_jonas> I got distracted
23:01:16 -!- tromp has quit (Remote host closed the connection).
23:01:45 <wob_jonas> basically I was thinking of creating an esolang parody of bad esolangs that some people create and write about on the wiki.
23:02:56 <ais523> wob_jonas: Statistical Brainfuck is basically that, but I never got round to writing it
23:03:11 <ais523> it's basically BF only it infers the commands in use via analysing the program
23:03:18 <wob_jonas> I'll try to use a lot of the bad tropes, such as no really eso features, missing computation power such as not enough control structures or not enough memory access, and also a bad attempt of an implementation that supports only the easier half of the language features, crippling it even more, sometimes fails to match the documentation, and has bug
23:03:18 <wob_jonas> s and bad coding that clearly shows the creator doesn't know how to implement interpreters (this part might be hard for me, I'm not saying I write good code, but it's hard to deliberately write code bad in that way)
23:03:19 <ais523> and if it can't figure it out, prints "Hello, world!".
23:03:32 <ais523> wob_jonas: ah, right
23:03:52 <ais523> the easiest way to write a "language from someone who can't write interpreters" is to have no form of control flow at all
23:04:01 <ais523> all the commands run in sequence, then the program ends
23:05:07 <wob_jonas> Also I won't learn Java or C# for just this, even though it would be realistic to write the interpreter in that, and I also won't unlearn C++ to imitate the horrible C++ styles found online, but I'll still try to do justice to using an unsuitable programming language and the wrong features of that langugae.
23:05:30 <ais523> wob_jonas: write it in C with a .cpp extension
23:05:39 <ais523> that should be close enough
23:05:56 <ais523> and is fairly easy to do even if you're a good programmer
23:06:01 <wob_jonas> ais523: yes, that can work, but if I do that, then either I can't write an interpreter at all, or I'll have to write an interpreter that's not obviously lacking important features of the language. that could be more realstic, but also more boring
23:06:18 <ais523> Python could also work, if you know it
23:06:49 <wob_jonas> ais523: yes, those are options. I was also thinking of using C or pascal with some vague label like "Pascal 5.0" that suggests some old DOS interpreter.
23:07:02 <wob_jonas> I haven't really decided yet.
23:08:01 <ais523> what about some proprietary language which is very similar to a more widely used free language?
23:08:09 <ais523> only I don't know many examples of those nowadays
23:08:51 <wob_jonas> ais523: I do know a little mathematica and maple and matlab, which may be enough to write bad code, but I'm not sure how realistic it would be for a newbie to use those to write interpreters
23:08:57 <wob_jonas> even bad interpreters
23:09:19 <Taneb> wob_jonas, how about visual basic
23:09:22 <ais523> wob_jonas: right
23:09:23 <wob_jonas> I can also write C or C++ that only works in microsoft compilers
23:09:29 <ais523> oh, yes, visual basic works perfectly for this
23:09:40 <ais523> and you can learn the relevant parts in like 5 minutes if you don't know it already
23:09:44 <ais523> (spending any more time would defeat the purpose)
23:09:55 <ais523> C using conio.h would also make sense
23:10:06 <wob_jonas> Taneb: eww. I mean, I can write bad basic, I used to, even in visual basic, but I would feel so dirty from that as if I wrote php
23:10:21 <Taneb> wob_jonas, :)
23:11:28 <wob_jonas> ais523: pascal with turbo/borland pascal's crt library is sort of the same as conio from turbo/borland c, only older and that much simpler (turbo pascal is actually a non-optimizing compiler), and was quite popular in the days when people used turbo pascal in DOS for education, and those days haven't completely ended yet
23:12:06 <ais523> <wob_jonas> (turbo pascal is actually a non-optimizing compiler) ← now I'm even more confused and worried about how when, years ago, I rewrote a Turbo Pascal program in C and it got slower
23:12:19 <wob_jonas> mind you, it was a good IDE and library for the context it was made for
23:12:26 <ais523> was there something massively wrong with the library I was using, or something massively wrong with my code?
23:13:04 <wob_jonas> ais523: turbo C is an optimizing compiler, but that's basically like a non-optimizing compiler today. sort of like how C is a low-level language now, but was a high level language forty years ago
23:13:29 <wob_jonas> serious projects used assembly for all the performance-critical code
23:14:25 -!- tromp has joined.
23:14:27 <wob_jonas> it was easier to mix with C than even now, because there was good support for assembly in the IDEs and debuggers, good support for inline assembly, the compilers had simple ABIs, and the CPU and OS was simpler so optimized assembly prorgamming was easier to learn
23:15:02 <wob_jonas> which of course means that most of the assembly programs written back then are worse than what you'd write in C++ today, with some exceptions of master programmers, but still, it was useful at the time
23:15:32 <ais523> wob_jonas: I'm trying and failing to remember the precise C compiler I used
23:16:03 <wob_jonas> ais523: what target platform? MS-DOS? win16 on x86? win32 on x86?
23:16:05 <ais523> also, ABIs are /still/ simple when the signature of the function is simple
23:16:11 <wob_jonas> or even OS-2?
23:16:31 <ais523> and I know the computer it was on was running either Windows 95 or Windows 98, but I don't know the platform of the executable, although it was probably a DOS executable
23:17:26 <wob_jonas> turbo C and borland C did support win16 as a target, but few people learned that because then you have to learn a lot about win16 system programming first and get familiar with the libraries
23:17:48 <ais523> win16 was my "main" target platform for several years
23:18:01 <wob_jonas> good GUIs are still hard, and it was even harder when you HAD to support cooperative multitasking with no memory protection
23:18:02 <ais523> when it stopped working, I ended up switching to Linux rather than trying to get win32 to work
23:18:20 <ais523> (part of the reason behind this is that my compiler had problems making win32 code, possibly due to bitrot)
23:18:35 <\oren\> Cringe! https://scontent.fwaw3-1.fna.fbcdn.net/v/t1.0-9/21317927_870196236467487_2502978163715573443_n.jpg?oh=0036232e6f91b8a0c522171488267087&oe=5A1901C7
23:18:53 <ais523> win16 + GDI wasn't too bad when you were used to it, anyway, although my programming style was shocking back then
23:18:59 <ais523> or maybe not shocking, just unsuitable for Windows
23:19:26 <ais523> I typically put the entire program logic into the repaint handler, then had a timer to repaint the window at short intervals
23:19:29 <wob_jonas> actually, when did it stop working? I stopped using win16 programs before windows dropped support, but I know win95 OSR2 runs the non-unicode win16 programs winword 2 and 6 and excel 5 fine, and I seem to remember they still ran on win32
23:19:40 <ais523> which is actually how most old games consoles work (although I didn't know that at the time), but is very out of place on PC
23:20:05 <ais523> wob_jonas: even in windows 98 some of the APIs didn't work
23:20:09 <wob_jonas> or do you mean compilers and libraries stopped supporting win16 programming, as opposed to windows stopped supporting running windows programs?
23:20:10 <ais523> like the one to play music through the PC speaker
23:20:19 <pikhq> Technically, it never did stop working -- Windows 10 *still* supports Win16, if you're on x86, not x86_64.
23:20:38 <wob_jonas> ais523: I know some of the old APIs didn't work, but those came up in very old win16 programs from before windows 3 (seriously) or older DOS programs
23:20:40 <pikhq> That said, I imagine stuff only *kinda* works these days.
23:20:44 <ais523> I had to move onto MDI instead which is a pain because on our Windows 98 configuration, attempting to load a MIDI file with MDI caused a freeze of tens of seconds
23:21:08 <ais523> (although oddly, if you played one through Windows Media Player, the freeze happened when the file looped for the first time rather than when it was initially loaded)
23:21:23 <wob_jonas> pikhq: yes, but it's sometimes easier to just run those programs in a full virtual machine
23:21:31 <ais523> I worked around this by preloading all the MIDI files during a loading screen at the start but it meant that the program took ages to load
23:21:56 <ais523> Windows XP reduced this to about a second per file, so it was much more reasonable (although still seems ridiculously large given how simple a MIDI file should be to load)
23:22:15 <ais523> also, IIRC MDI used a "command-line" API where you called a function and passed a string to it describing what you want to do
23:22:34 <ais523> rather than using separate functions which each had their own arguments
23:22:44 <ais523> and I may have the name of the library wrong, it was ages ago
23:23:01 <wob_jonas> ais523: yes, I can imagine the graphics and sound stuff used mostly by games stopped working earlier. that wasn't really MS's target to support. although some of that is because the DOS (and even nominally win16) programs used direct hardware stuff, not documented APIs, for speed and flexibility, and those are much harder to emulate than APIs that
23:23:01 <wob_jonas> reasonably designed wrt future compatibility
23:23:53 <wob_jonas> "I had to move onto MDI instead which is a pain because on our Windows 98 configuration, attempting to load a MIDI file with MDI caused a freeze of tens of seconds" => hehe, that totally sounds believable. for that era, and for today also
23:24:29 <wob_jonas> "by preloading all the MIDI files during a loading screen" => wait, it froze for loading EACH midi file? that's even worse
23:24:45 <wob_jonas> couldn't you work it around by loading just one big midi file and playing sections?
23:25:33 -!- mroman has quit (Quit: need sleep).
23:25:52 <wob_jonas> "also, IIRC MDI used a "command-line" API where you called a function and passed a string to it describing what you want to do" => we use the euphemism "domain-specific language" rather than "command line" for APIs like that
23:26:13 <wob_jonas> they still exist these days, and are sometimes good but sometimes terrible
23:26:49 <\oren\> i like sqlite
23:27:02 <wob_jonas> a reasonably good example is SQLite, where most of the API is SQL statements, and people keep asking for a more direct C api for many functions, but the devs wisely don't add such a thing, because that would make future compatibility much harder
23:28:31 <wob_jonas> a bad example is G'MIC's api, although I don't know enough about it to be sure. it involves syntax horror with impossible quoting of the like cmd and powershell could envy.
23:28:39 <wob_jonas> and it can access files from the file system.
23:30:02 <wob_jonas> you might also count cases where an interpreter was first and an API was added as an afterthought but the API wasn't designed, it just evolved from trying to hack the interpreter. perl is the classical example, but I seem to remember there was a few more like that.
23:31:00 <ais523> wob_jonas: I'm not sure what controlled the duration of the freeze, it might have been that a large file froze for logner
23:31:13 <wob_jonas> (perl 5 obviously
23:31:34 <ais523> wob_jonas: SQLite does have a direct C API for many of its functions
23:32:02 <wob_jonas> perl 6 has an overdesigned impossible to implement API with second system effect, completely unlike the rest of the language, because the designer for that is a different person from Larry who has the vision for the language proper)
23:32:43 <wob_jonas> ais523: direct C api for many of the auxiliary functions, but not for the database table accesses, which would be sort of the main point
23:34:27 <wob_jonas> also while it tries hard to pretend that it has a complete API for writing new SQL functions, there are actually a few things that the builtin functions can do but that are impossible with the public API. I asked about one or two of this, and it's by design, because they couldn't figure out a good enough API for that functionality yet, so they'll r
23:34:27 <wob_jonas> ather not add bad APIs that will constrain future compat.
23:37:09 <wob_jonas> in particular, MIN and MAX magically compare text values (character strings, they just use a strange term) with the collating function appropriate for the column, when the documented API makes that impossible
23:37:49 <wob_jonas> there's also a few builtin optimizations for MIN and MAX, but then those are transparent optimizations so that's OK
23:38:54 <wob_jonas> anyway, if I create such a parody, I'll try to create a new user account on the esowiki for it, with a little background personality story, and perhaps call it fungot
23:38:55 <fungot> wob_jonas: this, my friends, is the malaise of the glutton at life's buffet, building complicaters? domino frustraters? wobbley times u.s.a.? um, maybe if i told his jokes
23:39:52 <Phantom_Hoover> wob_jonas, isn't perl 6 basically an elaborate esolang in the intercal tradition by now
23:39:52 <wob_jonas> fungot: yes, and some of the things they promise are actually physically impossible. it is starting to sound like one of those pseudoscience products actually
23:39:52 <fungot> wob_jonas: this is a very common fantasy among children played hopscotch! maybe! i've never been to the bottom of a bottle. they'll just go somewhere else.
23:40:08 <wob_jonas> Phantom_Hoover: I don't know
23:40:44 <Phantom_Hoover> i mean
23:40:44 <Phantom_Hoover> http://glyphic.s3.amazonaws.com/ozone/mark/periodic/Periodic%20Table%20of%20the%20Operators%20A4%20300dpi.jpg
23:40:49 <wob_jonas> And I should try to hide some specific community in-jokes or references in the text too
23:41:04 -!- 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.”).
23:41:14 <wob_jonas> Phantom_Hoover: yes, I can see why it's like an esolang, it's just the Intercal part that I don't understand
23:41:35 <Phantom_Hoover> well intercal's philosophy was 'be a bizarre reflection of a serious language'
23:42:02 <wob_jonas> oh, and some of the perl6 features it has are very impractical to interpret even programs that sparesly use them, like computed come from
23:42:22 <Phantom_Hoover> whereas the vast majority of esolangs are along the brainfuck/befunge tradition of 'be some utterly weird model of computation'
23:42:55 <wob_jonas> Phantom_Hoover: but it doesn't look like a reflection of any language to me. quite the opposite, it seems a very unique language in its feature set and design, different from everything else I've seen, it's just that the feature set and design is terrible
23:43:22 <Phantom_Hoover> that's true of intercal too man
23:45:08 -!- oerjan has joined.
23:45:48 <wob_jonas> perl6 has some parts of utterly weird model of computation, such as how it stores data. you can imagine it as a parody of perl5, where in perl5 it occured mostly from historical compatibility and different features trying to have different models utterly incompatible with each other, as in DWIM scalars a la perl/bash/tcl versus scalars that know th
23:45:48 <wob_jonas> eir dynamical type a la python/ruby; in perl6 it's a straight up design without much historical compatibility reasons that ends up looking like a bad parody.
23:45:52 <Phantom_Hoover> i mean i've never looked into perl 6 too deeply b/c it was always too incomprehensible in my 'learn all the languages' phase
23:46:13 <wob_jonas> sure, I didn't look at it deeply either
23:46:36 <wob_jonas> and I'm actively avoiding it now, sort of like how I avoid PHP
23:47:41 <wob_jonas> the difference is that PHP has a bad history like C++ and has actually improved a lot in recent versions, slowly turning into a good language, but with most of the code monkeys using it still using all the bad programming practices they learn from past bad code examples and old books;
23:47:56 <wob_jonas> whereas perl6 isn't like that.
23:49:04 <Phantom_Hoover> what's so weird about perl 6's scalars
23:49:29 <wob_jonas> I'm not avoiding PHP because it's such a bad language, but because I have a superstition that the four big p languages (perl, ruby, python and php) are mutually incompatible in that any one person must choose at most three unless they want to tempt fate, and I already know too much about perl, ruby, and python.
23:50:05 <wob_jonas> Phantom_Hoover: I don't remember. I managed to deliberately forget most of the specific knowledge I acquired about perl 6. I'm happier not knowing now.
23:50:16 <wob_jonas> It's possible that the scalars aren't the problem.
23:50:45 <wob_jonas> I only retained the general warning sign reasons so that I know why I shouldn't revisit it ever.
23:50:56 <wob_jonas> You know, to avoid the "how bad can it be" thing
23:51:48 <wob_jonas> but it's not easy to suppress stuff, which is why I have to completely avoid the language or else the suppressed memories might resurface
23:55:07 <wob_jonas> wait wait. how can a modern library provide a portable thread abstraction and at the same time have structures with non-threadsafe reference counting. that makes no sense.
23:55:44 <wob_jonas> atomic refcounts are not hard to implement these days, there are tons of other code you can copy from if you want the portability, and they're almost always just as fast as non-threadsafe refcounts if there's no contention of multiple threads accessing the same refcount close to the same time, with the unusual exception when there's page tear
23:56:10 <wob_jonas> s/page tear/cache line tear/
23:56:27 <wob_jonas> cache line tear as in different threads accessing different parts of a cache line with at least one writing
23:56:59 <wob_jonas> and even with cache line tear you don't lose much performance compared to non-threadsafe code
23:57:08 <wob_jonas> maybe you don't lose any, I'm not sure
23:57:19 <wob_jonas> it's just that cache line tear already loses you performance, no matter what
23:57:38 <wob_jonas> you just have to avoid it, just like you avoid unaligned objects that pass through a cache line boundary
23:58:07 <wob_jonas> both are simply technologically impossible to support efficiently, there's nothing to do but avoid them whenever performance matters
23:59:51 <wob_jonas> but most of the time they're easy to avoid: just align all objects, and use a memory allocator that's reasonably threading aware. if you do that, then you usually have to do specifically stupid things as in disabling the safety that's already there to get misalignment, and typically have to do design that already has bad performance to get cache li
23:59:51 <wob_jonas> ne tear
←2017-09-06 2017-09-07 2017-09-08→ ↑2017 ↑all