00:06:16 -!- erkin has joined.
00:14:53 -!- erkin has quit (Quit: Ouch! Got SIGABRT, dying...).
00:20:11 <ais523> update: I've now configured the server to use the sort of garish coloured prompts that people turn on because they care more about showing off what their terminal looks like than actually using it
00:20:20 <ais523> hopefully I'm unlikely to forget I'm using it then
00:21:02 <shachaf> I think you're reading into people's intent a little more than is reasonable there.
00:21:10 <wob_jonas> ais523: and the purpose for this is to make sure you don't accidentally type something to that prompt instead of the prompt for some other server?
00:21:53 <ais523> wob_jonas: nah, it's to make sure that I don't type something on the server which I expect to be on my own laptop, which I'm connecting from
00:22:10 <ais523> it's hard to actually do damage that way (probably possible), but it can be very confusing
00:22:30 <ais523> because things I expect to be there just aren't (like ssh keys and programs)
00:22:50 <wob_jonas> ais523: the easiest is to shut down. I actually set up custom aliases of the form "foohalt" and "fooreboot" that are like halt and reboot but only on the machine named foo
00:23:25 <wob_jonas> (well, sort of. they're not exactly equivalent to /sbin/halt but almost the same)
00:23:30 <ais523> I've never had trouble with that because it would take very bizarre circumstances to shut down my own laptop from a graphical terminal
00:23:44 <ais523> if it were an emergency, I'd be using the text terminal or REISUO; if it's a routine shutdown, I'd use the GUI
00:24:18 -!- ais523 has quit.
00:24:27 -!- ais523 has joined.
00:25:02 -!- augur has quit (Remote host closed the connection).
00:36:38 -!- augur has joined.
00:47:37 -!- augur has quit (Remote host closed the connection).
01:38:33 -!- jaboja has quit (Read error: Connection reset by peer).
01:38:33 -!- jaboja64 has joined.
01:41:17 -!- wob_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
01:51:08 <imode> is urbit a running gag now.
01:51:15 <imode> because if it isn't it should be.
01:55:03 -!- jaboja64 has quit (Read error: Connection reset by peer).
01:55:57 -!- jaboja has joined.
02:06:15 -!- augur has joined.
02:15:01 -!- boily has quit (Quit: TORSADED CHICKEN).
02:51:37 <HackEgo> 8208:2016-05-29 <b_jonäs> learn Manager FAQ (by seebs) at http://www.seebs.net/faqs/manager.html
02:57:15 <zzo38> wob_jonas: Yes I think they are aligned in a different coordinate system; the constellations are defined for B1875 but the coordinates in those file are J2000.
03:03:41 -!- ais523 has quit (Quit: bedtime).
03:06:26 -!- oerjan has quit (Quit: Niter).
04:15:31 -!- dos has joined.
04:19:19 -!- hppavilion[0] has quit (Ping timeout: 268 seconds).
04:20:57 -!- dos has quit (Ping timeout: 240 seconds).
05:10:02 -!- doesthiswork has quit (Quit: Leaving.).
05:25:33 -!- dos has joined.
05:27:50 -!- augur has quit (Remote host closed the connection).
05:28:36 -!- doesthiswork has joined.
05:34:33 -!- jaboja has quit (Ping timeout: 268 seconds).
05:48:15 -!- augur has joined.
07:34:31 -!- doesthiswork has quit (Quit: Leaving.).
08:02:45 -!- augur has quit (Remote host closed the connection).
08:04:11 -!- imode has quit (Ping timeout: 246 seconds).
08:39:38 -!- sleffy has joined.
09:03:41 -!- sleffy has quit (Ping timeout: 246 seconds).
09:17:08 -!- aloril_ has quit (Ping timeout: 260 seconds).
09:55:00 -!- LKoen has joined.
10:23:58 <HackEgo> [wiki] [[Hi\n]] https://esolangs.org/w/index.php?diff=52428&oldid=52426 * Destructible watermelon * (+1563) Undo revision 52426 by [[Special:Contributions/Xavo|Xavo]] ([[User talk:Xavo|talk]])
10:40:52 -!- erkin has joined.
11:31:13 <Vorpal> upgrading debian from jessie to stretch I saw a line saying "deconfiguring udev (broken by systemd)". Scary.
11:46:40 <fizzie> I switched to the systemd persistent network interface names on this box, though "eno1" still looks a little weird.
11:48:24 <lambdabot> [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,...
11:48:57 -!- DHeadshot has joined.
11:49:12 <int-e> > fix$(0:).scanl(+)1
11:49:14 <lambdabot> [0,1,1,2,3,5,8,13,21,34,55,89,144,233,377,610,987,1597,2584,4181,6765,10946,...
11:54:30 <Vorpal> fizzie: I have seen ens33 on some box for the first wired interface
11:54:37 <Vorpal> using ubuntu 16.04 iirc
11:58:29 <fizzie> It will do "enoX" for on-board (== getting an interface index from firmware), "ensX" for pcie hotplug slots (and I think some virtualized things as well), "enpXsY" for PCI device by location, "enpXsYuZuWuQ..." for a chain of USB ports.
11:59:25 <fizzie> Debian installed some stuff in /etc/systemd/network/ to disable the thing for my Xen guest, with some comments about there being some weirdness going on with virtual interfaces.
12:00:07 <fizzie> (And of course it was doubly moot since it also won't do anything if you still have matching /etc/udev/rules.d/70-persistent-net.rules lines.)
12:02:17 <fizzie> Based on "udevadm info /sys/class/net/eth0" output, I think my VPS's network interface would call itself "ens3" if it wasn't for that.
12:12:37 <int-e> I don't have systemd and the interface is still called eth0. I'm not complaining, but why...
12:13:01 <fizzie> It would be more unexpected the other way around.
12:13:19 <fizzie> The names I mentioned were the systemd names, the kernel names are still the ethX they've long been.
12:14:44 <int-e> ah, down there: E: INTERFACE=eth0 ... fine
12:16:20 <int-e> . o O ( but why is this done by systemd rather than udev anyway )
12:17:12 <int-e> . o O ( I mean they could just teach udev to use ID_NET_NAME_PATH for the interface name. It would probably be a simple rule. )
12:17:40 <HackEgo> [wiki] [[Brainfuck constants]] https://esolangs.org/w/index.php?diff=52429&oldid=52010 * Primo * (+0) /* 100-149 */ soft-wrapping is never necessary for 3-cells, other values updated for consistency
12:18:51 <HackEgo> [wiki] [[Brainfuck constants]] https://esolangs.org/w/index.php?diff=52430&oldid=52429 * Primo * (+0) /* 150-199 */ soft-wrapping is never necessary for 3-cells, other values updated for consistency
12:21:05 <fizzie> They might've had an argument for that, though I forget what it was.
12:21:17 <HackEgo> [wiki] [[Brainfuck constants]] https://esolangs.org/w/index.php?diff=52431&oldid=52430 * Primo * (-229) /* 200-255 */ soft-wrapping is never necessary for 3-cells, other values updated for consistency
12:21:21 <fizzie> Might be just that systemd/udev developers don't really like to think of them as separate things.
12:24:30 <int-e> (The sad thing is that you're probably right. But I hate it.)
12:25:48 <fizzie> https://www.freedesktop.org/wiki/Software/systemd/PredictableNetworkInterfaceNames/ calls it "systemd/udev" a whole lot.
12:26:11 <fizzie> "With systemd 197 we have added native support for a number of different naming policies into systemd/udevd proper" "same on all distributions that adopted systemd/udev" and so on.
12:26:55 <int-e> oh, because that's not udevd; they have their own
12:30:38 <int-e> but that's not right either; my "systemd-less" VM is running /lib/systemd/systemd-udevd --daemon
12:31:42 <fizzie> I thought they merged them all into a single codebase.
12:32:05 <int-e> yeah, it appears they did do that
12:32:27 <fizzie> I assume they'll incrementally make it harder and harder to use udev without the rest of systemd.
12:43:22 <int-e> systemd is a cancer, it gets everywhere. I think that's the main reason I hate it. That and I have to relearn things with no (immediate) benefit to me.
12:44:24 <int-e> (The core dump interception is perhaps the bit that has me bitten the hardest so far. It can be turned off, but first one has to figure out *why* the core dumps are no longer produced the way they used to be.)
12:45:27 <int-e> I'm also no big fan of the system journal; I like perusing text files.
12:46:06 * int-e is an irrational being at heart :P
12:46:30 <fizzie> There's a lot of "something's happening, and I have no idea where it's configured" going on there.
12:47:57 <int-e> (Oh right, builtin policies. I wanted to configure my laptop so that it wouldn't go to sleep when the lid is closed while it's in a docking station... I don't think I managed, but it's been a while since I really tried.)
12:48:23 <int-e> (I *do* want it to sleep when the lid is closed while it's running on battery.)
12:54:37 <Vorpal> fizzie: well, UEFI is bugged after upgrade. No boot device found
12:54:48 <Vorpal> booting from USB stick now to fix it
12:54:58 -!- sdhand has quit (Excess Flood).
12:55:07 -!- sdhand has joined.
12:55:30 -!- sdhand has changed nick to Guest28947.
12:56:03 <Vorpal> great, I only get screen on the right monitor using this on this USB stick. Fuck nvidia maybe
12:56:20 -!- DHeadshot has quit (Ping timeout: 260 seconds).
12:58:10 -!- Guest28947 has quit (Changing host).
12:58:10 -!- Guest28947 has joined.
12:58:10 -!- Guest28947 has changed nick to sdhand.
12:59:06 <Vorpal> okay unplugged that monitor, rebooted. Now I get an "incompatible input timing" message on my main monitor
12:59:45 -!- aloril has joined.
13:00:07 <int-e> . o O ( No boot device found, is that coming from the BIOS or later? I've had trouble telling the BIOS about a new configuration from Linux; had to search for it in the BIOS setup itself. But I don't fully understand this ... basically I'm just glad I got it to work. )
13:00:26 <Vorpal> int-e: it is from the UEFI/BIOS
13:04:03 <int-e> I guess what I'm saying is that going through the BIOS' boot setup *may* be necessary to get things to work, because that has happened to me. This is anecdotal, not scientific.
13:05:16 <fizzie> Computers: not an exact science.
13:05:40 <Jafet> since red hat sells training and consulting, it only makes sense for it to make systemd as complicated as possible
13:06:16 <Vorpal> Well the output from efibootmgr is borked. Didn't list any EFI based debian
13:06:40 <fizzie> The other day I briefly unplugged one monitor, and after plugging it back (until the next reboot), xscreensaver's screen-blanking only blanked about 90% of the screen, leaving about one fifth of the left monitor's left edge showing whatever was there before locking the screen.
13:06:41 <Vorpal> <int-e> I guess what I'm saying is that going through the BIOS' boot setup *may* be necessary to get things to work, because that has happened to me. This is anecdotal, not scientific. <-- can't do it from there, have to do it from live USB
13:07:29 <Vorpal> fizzie: well this display works normally, but from the stripped down system rescue CD apparently.
13:08:23 <Vorpal> looks much better in efibootmgr now, hopefully the info is correct too
13:08:35 <fizzie> Also, when switching to the GLVND variant of the non-free nvidia driver, I had to install both libgl1-glvnd-nvidia-glx and libgl1-nvidia-glvnd-glx, which are two different packages.
13:10:52 <Vorpal> fizzie: my nvidia card is too new to be supported by non-free drivers (at least for 3D acceleration) last I looked
13:11:01 <Vorpal> err by the free I mean
13:11:17 <Vorpal> fizzie: what is the glvnd thing?
13:11:49 <fizzie> https://github.com/NVIDIA/libglvnd -- they make two variants of their binary driver nowadays, with and without that.
13:12:01 <Vorpal> jeez, ACP and LVM errors from super-early kernel/initramfs
13:13:18 <fizzie> I don't know what it's good for, but an "nvidia-vulkan-common" upgrade (on Debian testing) was listed as conflicting with the non-GLVND variant.
13:17:23 <Vorpal> interesting, the new version of debian auto-mounted a data disk with NTFS as rw under /media/<user>/
13:17:36 <Vorpal> not the system (C:) partition though
13:18:27 <fizzie> I use "udisksctl mount -b /dev/..." to mount removable media, and can never tell how it decides between /media/usb0 and /media/<user>/<label>.
13:18:44 <Vorpal> it always seem to auto-mount for me
13:18:56 <Vorpal> anyway why did it mount ntfs (using some fuse driver) as rw
13:19:05 <Vorpal> I don't want it doing that. I don't trust it
13:19:39 <Vorpal> oh so mint menu no longer works in new debian
13:23:37 <Vorpal> also the chromium failed to update, some dependency mess
13:31:46 <Jafet> @ask ais523 the x := (j + x)/(k + x) device in trajedy should also work in conedy
13:34:39 <Jafet> @ask ais523 my guess is that the x := (x + j)/k device is not enough to be universal, but using both might be
13:59:00 -!- boily has joined.
14:00:40 -!- doesthiswork has joined.
14:04:29 <Vorpal> Going to see if cinnamon is okay
14:04:50 <Vorpal> can't find the right version of mintmenu to make it work with the mate version on debian
14:13:00 -!- DHeadshot has joined.
14:19:03 -!- AnotherTest has joined.
14:21:46 <int-e> oh, no mirrors, hmm
14:23:02 <int-e> (That is, there are no mirrors in Conedy, in contrast to Trajedy.)
14:26:27 <int-e> Otoh, a ray may pass through a beacon, so perhaps that's not so bad.
14:33:05 <Jafet> I had thought of changing to orthogonal mirrors to remove the tunnelling artifact, but conedy's design is better
14:45:33 -!- DHeadshot has quit (Ping timeout: 260 seconds).
14:49:23 -!- Rem|Sleep has changed nick to Remavas.
14:52:21 <int-e> but the state is effectively reduced from 2 to 1 rational number, which may explain why Conedy looks very much Turing incomplete to me, though that will be impossible to prove. ;-)
14:54:06 <Jafet> the state is one rational number in both languages
14:54:23 <Vorpal> cinnamon seems okay. don't like the semi-transparent menu and taskbar popups. Apparently that is tied to the theme though, so I'd have to switch to another theme, which may be worse in some other way
14:54:27 <Vorpal> or make a custom theme
14:54:57 <int-e> Jafet: no, in Trajedy, if the ray originates from a mirror, you have a direction *and* an offset on the mirror
14:55:12 <int-e> Jafet: which is very much exploited in the tube device.
14:55:48 -!- LKoen has quit (Remote host closed the connection).
14:56:01 <int-e> Jafet: in Conedy, every ray passes through some beacon, so only the direction is left (all this is claiming some additional finite amount of state which I think of as flow control)
14:57:31 <int-e> claiming? not sure what I wanted to use there... "ignoring" fits
15:00:54 <Jafet> mirrors can rearrange the state into two parameters, but that's not really new state
15:01:36 <int-e> Vorpal: to make this idea a bit more formal: If I wanted to translate a Trajedy program into a programming language that has variables that are rational numbers, manipulated by rational Möbius transformations, and conditional jumps based on comparison with a rational constant, then I'd need two variables. With Conedy, one variable is enough.
15:02:05 -!- boily has quit (Quit: CATHODE CHICKEN).
15:02:38 <Jafet> anyway, fractran has half a rational number as state so that doesn't, by itself, disprove anything
15:02:58 <int-e> Jafet: note the restriction on the conditional, it's very important
15:03:48 <Jafet> huh, there is such a thing as a Möbius transformation
15:08:20 <int-e> I guess I need addition as well for the reduction from Trajedy (u += v), otherwise the mirror-mirror transition cannot be encoded.
15:13:51 <int-e> Jafet: Knowing about Möbius transformations (in particular the fact that they form a group) gave me the confidence to attempt building the x -> (1-x)/(1+x) transition that is fundamental for the corrected zig-zag test.
15:18:55 <int-e> (you can also think of them as perspective transformations in a single dimension; both views can be helpful)
15:27:58 -!- jaboja has joined.
15:36:28 <Vorpal> int-e: interesting, but why did you highlight me about it?
15:43:39 -!- sebbu has quit (Read error: Connection reset by peer).
15:44:10 -!- sebbu has joined.
15:58:49 <int-e> Vorpal: no particular reason, I have a strong tendency to pick the wrong name from the recent history to reply to
16:09:27 -!- DHeadshot has joined.
16:26:50 <HackEgo> [wiki] [[Comefrom0x10]] https://esolangs.org/w/index.php?diff=52432&oldid=52424 * Julfers * (+84)
16:30:33 -!- DHeadshot has quit (Ping timeout: 260 seconds).
16:38:09 -!- jaboja has quit (Ping timeout: 248 seconds).
16:41:44 <Vorpal> fizzie: there doesn't seem to be support for nft in stretch, unlike what you suggested?
16:42:10 <Vorpal> or hm, the "install this package to get the command thing" may be broken
16:49:39 <fizzie> Should be there as 'nftables'.
16:55:40 <int-e> . o O ( not fucking terrible?! )
16:56:43 <int-e> (it came up yesterday; I still remember)
17:19:01 -!- imode has joined.
17:19:11 <HackEgo> [wiki] [[Special:Log/newusers]] create * Christopher * New user account
17:23:48 -!- erkin has quit (Quit: Ouch! Got SIGABRT, dying...).
17:38:58 -!- imode has quit (Quit: WeeChat 1.9).
17:42:21 -!- Jafet has quit (Read error: Connection reset by peer).
17:47:43 -!- imode has joined.
17:54:11 -!- erkin has joined.
18:12:37 -!- LKoen has joined.
18:13:15 -!- Jafet has joined.
18:21:18 <Jafet> interesting, the gnome package depends on the apache2-bin package
18:28:53 <fizzie> I once had some HP printer drivers (on Windows) that bundled Apache Tomcat in them.
18:29:05 <fizzie> Because the printer control dialog was actually a JSP thing.
18:30:55 <int-e> it's things like those that make me wish I'd learned carpentry
18:31:48 <fizzie> I'm sure there's something equivalent to complain about in carpentry.
18:31:53 <Jafet> the apache foundation probably has something for your cnc router, too
18:32:57 <int-e> fizzie: ikea furniture? :P
18:33:23 -!- LKoen has quit (Remote host closed the connection).
18:33:58 -!- jaboja has joined.
18:37:52 -!- DHeadshot has joined.
19:04:05 -!- ais523 has joined.
19:08:28 -!- erkin has quit (Quit: Ouch! Got SIGABRT, dying...).
19:26:31 -!- erkin has joined.
19:31:14 -!- ais523 has quit (Remote host closed the connection).
19:31:23 -!- ais523 has joined.
19:31:46 -!- LKoen has joined.
19:34:34 -!- LKoen has quit (Client Quit).
19:36:17 <zzo38> Do you know if GCC can detect type mismatch in printf format strings even if the format argument is using a ?: operator?
19:38:53 <fizzie> I was real confident it wouldn't, but apparently it just might.
19:39:05 <fizzie> 19:38 <fizzie> ,cc int main(int argc, char **argv) { printf(argc ? "%d" : "%05d", 12.3); }
19:39:08 <fizzie> 19:38 <candide> fizzie: [warning: format '%d' expects argument of type 'int', but argument 2 has type 'double' [-Wformat=]
19:42:24 <alercah> you can't have different types on both sides, so it's easy enouhg
19:42:44 -!- jaboja64 has joined.
19:42:56 <fizzie> I think the question was about checking the *other* parameters against the format specifiers.
19:43:08 -!- jaboja has quit (Read error: Connection reset by peer).
19:43:16 -!- erkin has quit (Quit: Ouch! Got SIGABRT, dying...).
19:47:45 <fizzie> With the format string provided as foo ? "format 1" : "format 2".
19:48:12 <fizzie> See the example I tried.
19:48:47 <Vorpal> <alercah> you can't have different types on both sides, so it's easy enouhg <-- that annoyed me in C++, where I wanted to use it to select different overloads.
19:49:27 <Vorpal> forgot the exact situation, but it cropped up a couple of times
19:49:59 <Vorpal> Or doing stuff like something ? boost::none : value, needing extra casts to boost::optional<T> in both cases
19:55:52 <fizzie> Vorpal: Random question -- do you happen to know what I could use if I had a program that supported a gamepad in general, but had hardcoded keyboard shortcuts for some actions, and I wanted to bind an (unused) gamepad button to one of those keys.
19:55:58 <zzo38> fizzie: Yes, it is about checking other parameters against format specifiers, and the example you gave is like how I mean.
19:55:58 <fizzie> I've found a couple of things for generally binding gamepad to keyboard/mouse keys, but I don't know if using those would allow the program's native gamepad support to still work.
19:56:03 <fizzie> (Guess I could bind *everything* to keys, and configure the program to just use a keyboard.)
20:00:10 <zzo38> That can work although my idea would have been that the program might be implemented to use one layer to translate input into MIDI (and you can replace this layer then if needed), and the other layer interprets the MIDI messages that are received in this way.
20:01:22 <zzo38> (You can then, in future, to easily support gamepads that can detect how hard you pushed the buttons, if needed, without needing to replace everything.)
20:10:29 <Vorpal> fizzie: Windows or Linux?
20:11:36 <Hoolootwo> hmm, is there a good one of those for linux? I would be interested in that
20:11:44 <Vorpal> fizzie: also, if windows, 1) native xinput pad (e.g. xbox pad) or dualshock 4? 2) is the program using xinput or directx input?
20:12:11 <Hoolootwo> it'd be nice if there were a cross-platform scriptable way of doing it
20:12:40 <Vorpal> zzo38: why midi? Sounds needlessly complicated
20:13:09 -!- imode has quit (Ping timeout: 255 seconds).
20:13:47 <Vorpal> fizzie: anyway for the dualshock 4 case I believe there is a windows app for it, ds4windows that I think can do that. Don't have one of those though.
20:14:26 <Vorpal> fizzie: on android the sixaxis controller root-requring program can do that as well (at least with PS3 controllers)
20:15:07 <Vorpal> fizzie: for the linux case I assume it isn't complicated. I have been wondering about the xinput case myself a couple of times
20:15:18 <Vorpal> xinput on windows that is
20:15:43 <Vorpal> I would also like to be able to remap buttons and such on the controller
20:16:09 <Vorpal> hm that I can actually do on windows, but only stuff like a to b and such, which is not as useful
20:16:40 <zzo38> Vorpal: You do not actually need to use the protocol, although the format of the messages will help at least.
20:16:59 <Vorpal> I have an xbox one elite controller, the windows store app for that lets you set up profiles, but nothing like mapping to keyboard
20:18:47 <zzo38> For Windows, there is a program GlovePIE (with a license that prohibits using in some countries, using it to cheat at multiplayer games (although it is OK to use it to cheat at singleplayer games), and earning money using it for baseball simulation).
20:19:28 <Vorpal> that is a really really strange license
20:19:38 <zzo38> Yes, it is strange
20:33:57 -!- jaboja64 has quit (Ping timeout: 240 seconds).
20:39:38 -!- dos has quit (Ping timeout: 260 seconds).
20:41:30 <Vorpal> fizzie: I can only imagine there exists some program for it on Linux, but I don't know off hand what it would be
20:45:05 <fizzie> I've found a few -- https://github.com/Ryochan7/antimicro seems pretty popular though not packaged in Debian, or xboxdrv which is in the repo. (The latter was originally just a very configurable userspace driver for Xbox pads, but can also read from a /dev/input/eventX so it can do the keymapping stuff for any device.)
20:45:37 <fizzie> Should probably just try out if they'd play well with the program.
20:57:16 -!- hppavilion[1] has joined.
20:57:51 -!- ais523 has quit (Remote host closed the connection).
20:59:02 -!- ais523 has joined.
21:05:38 <zzo38> Should the esolang wiki article about MIX (Knuth) maybe should mention in the instruction listing, which ones are extensions and which one are standards? (I don't know which one they are, but I have added the character codes)
21:09:33 -!- S1 has joined.
21:11:11 -!- hppavilion[1] has quit (Remote host closed the connection).
21:13:53 -!- LKoen has joined.
21:23:49 -!- DHeadshot has quit (Ping timeout: 246 seconds).
21:35:43 -!- augur has joined.
21:39:23 -!- augur has quit (Remote host closed the connection).
21:41:58 <zzo38> When writing bindings for other program languages such as SQL and JavaScript, usually the call to create a function can use a user-defined pointer value for whatever purpose you need; I find I usually do not need it but the times I have used it, it is almost always a function pointer. Is that how you do too?
21:42:20 -!- ais523 has quit.
22:03:03 -!- Sprocklem has joined.
22:08:49 -!- augur has joined.
22:13:29 -!- jaboja has joined.
22:27:08 -!- wob_jonas has joined.
22:27:24 <wob_jonas> "<zzo38> Yes I think they are aligned in a different coordinate system; the constellations are defined for B1875 but the coordinates in those file are J2000." ok
22:30:32 <wob_jonas> "<fizzie> The other day I briefly unplugged one monitor, and after plugging it back (until the next reboot), xscreensaver's screen-blanking only blanked about 90% of the screen, leaving about one fifth of the left monitor's left edge showing whatever was there before locking the screen." lol, that's an amusing bug
22:31:53 -!- DHeadshot has joined.
22:33:26 -!- imode has joined.
22:34:20 <wob_jonas> "<zzo38> Do you know if GCC can detect type mismatch in printf format strings even if the format argument is using a ?: operator?" => no idea, you might want to TIAS (possibly both with and without optimizations) and/or ask on #gcc
22:35:32 <wob_jonas> "<fizzie> Vorpal: Random question -- do you happen to know what I could use if I had a program that supported a gamepad in general, but had hardcoded keyboard shortcuts for some actions, and I wanted to bind an (unused) gamepad button to one of those keys." => I think there are programs that bind gamepad buttons to keyboard keys
22:35:48 <wob_jonas> "<fizzie> I've found a couple of things for generally binding gamepad to keyboard/mouse keys, but I don't know if using those would allow the program's native gamepad support to still work." ah
22:35:49 <fizzie> wob_jonas: Yeah, I ended up using this "antimicro" one.
22:36:24 <fizzie> Yeah, just fine. I guess it's okay to have multiple readers on an input device.
22:39:37 <wob_jonas> "<zzo38> Should the esolang wiki article about MIX (Knuth) maybe should mention in the instruction listing, which ones are extensions and which one are standards? (I don't know which one they are, but I have added the character codes)" => perhaps. though sometimes it's not clear what even counts as standard and extension, since the binary and float
22:39:37 <wob_jonas> ing point extensions are explained in the main text, not in exercises.
22:40:28 <wob_jonas> also, about MIX, question, where does TAOCP say that the IN/OUT instructions for disk drive use the rX register for the sector number? is that in chapter 1 somewhere, or only later in chapter 4 or something?
22:41:33 <wob_jonas> zzo38: but basically the instructions I mention in the main text plus the system instructions HLT, JRED, JBUS, IOC, IN, OUT are the standard ones.
22:43:40 <wob_jonas> And even then I guess a cheap machine might not implement IOC if it doesn't have any I/O devices that support it, and might have stub implementations for JBUS/JRED if its I/O devices are always synchronous.
22:48:29 <wob_jonas> What would definitely make sense is to define the JBUS, JRED, IN, OUT, IOC, HLT instructions for the cheaper I/O devices in the text, so that the text becomes enough reference for implementing a simple MIX machine.
22:49:31 <int-e> wow, what a bug. https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=849342 <-- xserver-xorg-legacy break xrdp (on Debian stretch)
22:54:14 <wob_jonas> zzo38: that character code doesn't seem right
22:55:11 <wob_jonas> in the Hungarian translation (which is based on the older editions), the greek characters in the code are Delta, Sigma and Pi. I cant' check in the new English originals, because the greek letters are messed up in the PDF e-book.
22:57:24 <HackEgo> [wiki] [[Talk:MIX (Knuth)]] N https://esolangs.org/w/index.php?oldid=52433 * B jonas * (+260) Created page with "The character code doesn't seem right. In the Hungarian translation (which is based on the older editions), the three greek letters are Delta, Sigma and Pi. I cant' check in..."
22:57:48 <wob_jonas> zzo38: could you check the character code?
23:05:33 -!- S1 has quit (Quit: Leaving).
23:12:44 -!- 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.”).
23:15:14 <wob_jonas> Re sector number for disk I/O is rX: it's defined right there in 1.3 where it should be, above the definition of the I/O instructions.
23:15:21 <wob_jonas> I was just blind and didn't see it.
23:17:37 -!- boily has joined.
23:20:44 <zzo38> wob_jonas: I have a book from library, which is Art of Computer Programming vol. 3, and the page in the back gives Theta, Phi, Pi. That is where I got them
23:21:17 <wob_jonas> zzo38: what edition of TAOCP vol 3? first or second?
23:23:24 <wob_jonas> Doesn't it say somewhere near the beginning?
23:24:27 <zzo38> I checked and it does not seem to say
23:25:45 -!- AnotherTest has quit (Ping timeout: 255 seconds).
23:26:30 <wob_jonas> I mean, there's one stupid change in the Hungarian translation: they changed the letters used for the assembly syntax of local labels. So it's not impossible that they changed the greek letters too.
23:26:35 <zzo38> Maybe it can be mentioned that some versions are using different Greek alphabets
23:27:24 <wob_jonas> Ideally I'd like to know what the latest paper version in the original English says. Maybe someone here will volunteer that info.
23:27:43 <wob_jonas> But eventually it's not really important.
23:27:44 <zzo38> Yes, if someone can figure out, it can help.
23:27:53 <wob_jonas> It doesn't change anything really.
23:28:37 <wob_jonas> Also, it's totally in character for a 70s computer for the teletype to use slightly different character codes in different countries.
23:30:25 <wob_jonas> Also I should try to figure out what font encoding or simliar problem is why the greek letters are messed up in the e-book, maybe I could reverse engineer the greek letters from that.
23:32:10 -!- DHeadshot has quit (Ping timeout: 240 seconds).
23:37:30 <HackEgo> [wiki] [[MIX (Knuth)]] https://esolangs.org/w/index.php?diff=52434&oldid=52427 * B jonas * (+1427)
23:40:18 <HackEgo> [wiki] [[MIX (Knuth)]] https://esolangs.org/w/index.php?diff=52435&oldid=52434 * B jonas * (+196)
23:42:19 <wob_jonas> I think what's messed up is that in the vanilla computer modern fonts, cmtt doesn't have the double acute accent, and has some other character there instead, so in some LaTeX fonts they added the double acute accent to some other code where the greek letters used to be.
23:42:44 -!- jaboja has quit (Remote host closed the connection).
23:42:53 <wob_jonas> But I'm not sure where I can easily look up the character encoding for those LaTeX fonts.
23:43:23 -!- jaboja has joined.
23:45:18 -!- oerjan has joined.
23:46:31 <wob_jonas> The characters that show up in the TAOCP pdf are: 10 acute accent, 20 ring accent, 21 double acute accent.
23:49:40 <wob_jonas> In the original computer modern encoding, in cmr, acute is on 0x13, ring is on 0x17, double acute is on 0x7D; in cmtt, acute and ring are still in the same place, and double acute isn't present.
23:54:58 <wob_jonas> But those codes are less relevants. In the original encoding, the greek characters in cmtt are at: 0x01 Delta, 0x02 Theta, 0x03 Lambda, 0x05 Pi, 0x08 Phi.
23:58:42 <wob_jonas> And indeed in the LaTeX 8t encoding, the accents are at: 0x01 grave, 0x01 acute, 0x02 circumflex, 0x03 tilde, 0x04 diaresis, 0x05 double acute, 0x06 ring (wtf), 0x08 brevis.
23:59:48 <wob_jonas> That would make the three greek letters in the MIX character code: 10 Delta, 20 Sigma, 21 Pi.