00:02:21 -!- tromp has joined.
00:03:02 -!- heroux has quit (Read error: Connection reset by peer).
00:03:02 -!- heroux_ has quit (Write error: Connection reset by peer).
00:03:15 -!- heroux has joined.
00:03:16 -!- heroux_ has joined.
00:04:26 -!- heroux has quit (Read error: Connection reset by peer).
00:04:26 -!- heroux_ has quit (Read error: Connection reset by peer).
00:05:30 <esowiki> [[The Esoteric File Archive]] https://esolangs.org/w/index.php?diff=57509&oldid=42081 * Blacksilver * (+87) I made a mirror! Table to hold it and any future mirrors.
00:06:34 -!- tromp has quit (Ping timeout: 246 seconds).
00:08:16 -!- heroux has joined.
00:08:35 <esowiki> [[User:Blacksilver]] N https://esolangs.org/w/index.php?oldid=57510 * Blacksilver * (+83) Does the IRC hook ignore User pages?
00:11:49 -!- S_Gautam has joined.
00:13:16 -!- heroux_ has joined.
00:17:54 -!- Blacksilver has quit (Quit: Page closed).
00:18:13 -!- Blacksilver has joined.
00:23:01 -!- wob_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
00:26:08 <arseniiv> Blacksilver: I’m quite new here and I haven’t used that archive so I can’t properly appreciate what you did, but as the channel is still right now, someone should say probably this is great!
00:27:26 <Blacksilver> 5-minute virtualhost setup on an existing Apache VM. It's nothing.
00:28:10 <Blacksilver> I wonder if we can get the esofiles stored in one of those nuke-hardened bunkers in Switzerland...
00:31:39 <Blacksilver> More related: I'm trying to do code golf in EXAPUNKS.
00:51:45 -!- oerjan has joined.
01:24:38 -!- Blacksilver has quit (Ping timeout: 245 seconds).
01:32:08 -!- arseniiv has quit (Ping timeout: 245 seconds).
01:45:54 -!- erkin has quit (Remote host closed the connection).
01:59:11 <zzo38> I found that in SQLite, the xBegin method is not called (even though a transaction has started) if a CREATE VIRTUAL TABLE statement is executed. If you are connecting to an existing virtual table then xBegin is called when a transaction with that virtual table is started.
02:39:16 -!- XorSwap_ has quit (Quit: the creeping crawling chaos will return.).
03:31:25 -!- S_Gautam has quit (Quit: Connection closed for inactivity).
03:31:55 -!- bradcomp has joined.
04:02:01 -!- bradcomp has quit (Ping timeout: 244 seconds).
04:10:04 -!- bradcomp has joined.
04:15:11 <zzo38> How to make a phonetic writing that is not limited to human physiology and maybe not even limited to a possible kind?
05:04:47 -!- bradcomp has quit (Ping timeout: 240 seconds).
05:06:48 -!- xkapastel has quit (Quit: Connection closed for inactivity).
05:19:23 -!- tromp has joined.
05:24:12 -!- tromp has quit (Ping timeout: 272 seconds).
05:56:08 -!- nfd9001 has joined.
06:18:36 -!- mynery has left.
06:19:09 -!- myname has joined.
06:24:25 -!- tromp has joined.
06:29:13 -!- tromp has quit (Ping timeout: 245 seconds).
07:06:51 -!- tromp has joined.
07:11:30 -!- tromp has quit (Ping timeout: 252 seconds).
07:12:32 -!- oerjan has quit (Quit: Nite).
07:20:50 -!- LKoen has joined.
07:23:00 -!- tromp has joined.
07:33:30 -!- quintopia has quit (Remote host closed the connection).
08:10:53 -!- XorSwap has quit (Ping timeout: 245 seconds).
08:36:06 -!- heroux_ has quit (Read error: Connection reset by peer).
08:36:06 -!- heroux has quit (Read error: Connection reset by peer).
08:36:20 -!- heroux has joined.
08:36:49 -!- heroux_ has joined.
08:43:15 -!- fungot has quit (Ping timeout: 252 seconds).
08:44:34 -!- heroux_ has quit (Read error: Connection reset by peer).
08:44:34 -!- heroux has quit (Read error: Connection reset by peer).
08:44:34 -!- fungot has joined.
08:44:47 -!- heroux has joined.
08:44:48 -!- heroux_ has joined.
08:45:32 -!- heroux_ has quit (Read error: Connection reset by peer).
08:47:47 -!- heroux has quit (Read error: Connection reset by peer).
08:49:48 -!- heroux has joined.
08:54:48 -!- heroux_ has joined.
09:20:51 -!- vans12 has joined.
09:21:32 -!- vans12 has quit (Remote host closed the connection).
11:01:37 -!- wob_jonas has joined.
11:14:19 <wob_jonas> So, I no longer even constantly feel a nagging sensation that there's a wound in my mouth, but there's still a wound, so now I want to eat meals and want to chew it on both sides, but have to constantly spend conscious effort to only chew in the right hand side of my mouth and don't let food touch the wound.
11:15:47 <Taneb> That sounds frustrating
11:16:13 <wob_jonas> Also, there's an annoying fishing line knot in my mouth that I keep constantly feeling, but I must resist my reflex to try to move it with the inside of my cheek or my tongue. And I can't even use a chewing gum to redirect my mouth movements, because that would be worse, then I'd have to make sure to only chew the chewing gum on the right hand side
11:19:33 <wob_jonas> It's possible that I might not wait the two weeks (starting from Friday which was the day of the operation) for my next visit to the dentist (she's not just a dentist, but I can't type maxillofacial surgeon) to remove the stitch, but ask my brother to remove it early next week, for by that time the wound will have healed enough that the stitch is n
11:31:40 <wob_jonas> zzo38: that sounds like how sqlite delays opening the file that backs up the database to as late as possible, in order that certain pragmas can take effect. This was especially important back when they didn't have the crazy URL syntax to open files with options, but IMO the url syntax is a mistake and options should be provided with pragmas anyway.
11:36:00 <wob_jonas> zzo38: it's possible that this is because the CREATE VIRTUAL TABLE statements are stored in the database schema, and when the schema is parsed, there might be an interdependence between the virtual tables, so it's worth to delay virtual tables to as late as possible.
11:43:27 -!- Shragazord has joined.
12:12:47 -!- A_D9 has joined.
12:13:32 -!- A_D9 has quit (Remote host closed the connection).
12:32:08 -!- cats23 has joined.
12:32:42 -!- erkin has joined.
12:32:52 -!- cats23 has quit (Remote host closed the connection).
12:34:27 -!- wodencafe4 has joined.
12:35:28 <wob_jonas> `ehlist http://eheroes.smackjeeves.com/comics/2663524/sorted-sort-of/
12:35:29 <HackEso> /srv/hackeso-code/multibot_cmds/lib/limits: line 5: exec: ehlist: not found
12:37:21 -!- zseri has joined.
12:39:18 -!- wodencafe4 has quit (Remote host closed the connection).
12:39:25 -!- LKoen has quit (Read error: Connection reset by peer).
13:11:15 -!- arseniiv has joined.
13:58:26 -!- ais523 has joined.
14:02:02 -!- heroux has quit (Write error: Connection reset by peer).
14:02:11 -!- heroux has joined.
14:06:04 -!- SopaXorzTaker has joined.
14:10:25 -!- LKoen has joined.
14:15:06 <zzo38> Not all options could use pragmas; ATTACH opens the database and reports any errors immediately, so the VFS to use (and any VFS options needed at time to open) needs to be specified in the ATTACH command. Still some options for the VFS that can be changed after it is opened, can be specified by pragmas (a VFS can only override pragmas after the file is open, anyways).
14:20:19 <wob_jonas> zzo38: wait, why does ATTACH do that? the semantics rule is that a bare table/view name without an attach name before it refers to the table in the earliest database, so a new ATTACH can't cause to reinterpret statements that were already working.
14:23:13 <zzo38> I don't know why, but I have tried it.
14:23:41 <wob_jonas> zzo38: in any case, there are a few things you can only do by opening something as the primary database (namely setting the directory for the master journal for cross-database atomic commits) and I think there are some global settings that you may have set before you act on any database
14:24:53 <zzo38> Still, the VFS can only implement pragmas if the file is open.
14:29:00 <zzo38> Maybe the URL syntax is a mistake, but then other syntax may be needed to specify what VFS to use for a attached database
14:38:06 <wob_jonas> zzo38: technically you could implement global pragmas from a VFS wrapper to set the default options for the next databases to be opened, but that's not a very good idea for a VFS to do
14:39:09 <wob_jonas> zzo38: but the VFS layer could define a pragma to tell which VFS to activate for opening a file
14:39:55 <wob_jonas> I'm not sure how all this should work.
14:42:43 -!- xkapastel has joined.
14:46:26 <zzo38> I don't know either, but for compatibility it can be kept how it is.
15:03:26 -!- wob_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
15:13:36 <arseniiv> zzo38: How to make a phonetic writing that is not limited to human physiology and maybe not even limited to a possible kind? => seems too unconstrained a task
15:14:18 <arseniiv> non-human-like alien phonetics could be almost anything, I think
15:15:50 <arseniiv> then you could try to do something with acoustic side of phonetics, not with articulatory one — at least acoustics is common to us all
15:16:30 <arseniiv> …if the language is an acoustic one
15:17:31 <arseniiv> and also there are still variables like audible frequency range
15:18:47 <arseniiv> but many contemporary human phonetic writing systems are based more on articulation and less on acoustics
15:24:37 <zzo38> I assume it is still with sounds, at least
15:38:09 <arseniiv> ah, also we can try to enumerate various ways in which sound can be produced, and especially inside the specified alien
15:42:21 <arseniiv> then we could envision their articulatory system and finally go the way out phonetics gone — investigate what can and can’t combine with what and get something IPA-like (but IPA is far from completely describing our articulatory abilities, it focuses mainly on those used in known languages, and there is an extension to IPA adding something about “speech impairment”, as they say it)
15:42:49 <arseniiv> this way is not generic, obviously, though
15:43:34 <arseniiv> I bet one can’t systematize all the possible results in a humane fashion
15:43:49 <zzo38> They have the grid of sounds and not all of the combinations are filled in.
15:46:44 -!- XorSwap has joined.
15:48:34 -!- john_metcalf has joined.
15:50:36 <john_metcalf> Hi, has anyone got a copy of the paper "Is it alive, or is it GA?" by Thomas Ray? I can't find a copy online :-(
15:54:55 -!- sprocklem has joined.
16:01:48 -!- MDude has quit (Quit: Going offline, see ya! (www.adiirc.com)).
16:02:27 -!- MDude has joined.
16:09:49 <arseniiv> zzo38: They have the grid of sounds and not all of the combinations are filled in. => ah, this should be not hard then
16:12:11 <arseniiv> let that grid have N dimensions, then you could assemble each glyph from N parts. Then you could mutate these glyphs, for example when some parts occur together, they can be drawn in other way; etc. etc.?
16:12:52 <zzo38> Yes, I have considered that actually and once drawn them on a paper, so then they could be made using METAFONT
16:13:56 -!- ais523 has quit (Quit: quit).
16:27:20 -!- kritixilithos has joined.
16:31:13 <zzo38> I consider because, I play a GURPS game but my character is not human character so how to figure their speech? Presumably due to many languages redundancy you can still understand the speech if you are speeching English even though it is not perfectly the proper sound.
16:33:31 <zzo38> Do you know how to do that though?
16:42:41 -!- ep100 has joined.
16:56:47 -!- ep100 has quit (Ping timeout: 240 seconds).
16:58:05 -!- mist2 has joined.
16:58:37 -!- LKoen has quit (Remote host closed the connection).
16:59:04 -!- LKoen has joined.
17:00:31 <arseniiv> afraid I don’t understand what is meant precisely. How do other characters figure their speech, how should one render it when describing what is happening in English, or something else? For the second, there’s a great deal of freedom: we could write simply “<intended meaning>”, probably adding some alien concepts in that meaning, or write verbatim/transcribe (possibly both at the same time as in “and then he said привет
17:00:31 <arseniiv> <hi>”), and maybe other ways. For the first, it’s a story how people (or other entities) learn/know languages. Also two languages could be very near and often understandable (bot sometimes not really) by speakers of each another, they may enjoy being rendered as e. g. two distant English dialects
17:02:17 <zzo38> It isn't usually relevant, but sometimes you like to do anyways, because sometimes might be relevant. Once there was a Latin inscription, and so I wrote it in Latin as well as English.
17:02:36 <zzo38> But Latin is not the only language in use.
17:02:58 -!- mist2 has quit (Remote host closed the connection).
17:03:51 -!- LKoen has quit (Ping timeout: 252 seconds).
17:16:54 -!- Phantom_Hoover has joined.
17:20:07 -!- sprocklem has quit (Ping timeout: 240 seconds).
17:20:43 -!- evilman_work has joined.
17:21:26 -!- evilman_work has quit (Remote host closed the connection).
17:22:27 -!- xkapastel has quit (Quit: Connection closed for inactivity).
17:23:23 -!- LKoen has joined.
17:32:52 -!- ep100 has joined.
17:43:32 -!- kritixilithos has quit (Quit: all).
17:48:28 -!- XorSwap has quit (Ping timeout: 246 seconds).
17:50:36 -!- ep100 has quit (Ping timeout: 252 seconds).
18:10:15 -!- Blacksilver has joined.
18:26:25 -!- S_Gautam has joined.
18:41:32 -!- zseri has quit (Quit: Leaving).
18:58:57 -!- SopaXorzTaker has quit (Remote host closed the connection).
18:59:41 -!- raynold2 has joined.
19:00:48 -!- xkapastel has joined.
19:02:29 -!- raynold2 has quit (Remote host closed the connection).
19:08:36 -!- sebbu2 has joined.
19:10:22 -!- wob_jonas has joined.
19:10:25 -!- ocharles has quit (*.net *.split).
19:10:25 -!- mich181189 has quit (*.net *.split).
19:10:25 -!- diginet has quit (*.net *.split).
19:10:56 -!- mich181189 has joined.
19:11:10 -!- ocharles has joined.
19:11:48 <wob_jonas> Don't you love it when two organizations are implementing the two sides of supposedly the same interface, but they disagree in what the details of the interface are and interpret them in slightly incompatible ways, and each of them insist that they know what the interface is because they use it a lot and I don't?
19:12:12 -!- ocharles has changed nick to Guest39584.
19:12:17 -!- Guest39584 has quit (Changing host).
19:12:17 -!- Guest39584 has joined.
19:12:33 -!- sebbu has quit (Ping timeout: 252 seconds).
19:14:49 -!- diginet has joined.
19:15:20 <zzo38> Do you have the example?
19:16:23 <wob_jonas> zzo38: yes, but this example is a bit complicated to explain, because it's not in informatics, but in bureaucracy with rules mostly invented by government organizations.
19:18:36 <wob_jonas> I buy stuff, the vendor gives me a receipt saying what I bought and how much I payed for it and some other stuff, the format of the receipt is controlled by government rules, then I give this receipt to an organization and then I get a reduction on my income tax,
19:18:48 -!- LKoen has quit (Remote host closed the connection).
19:20:05 <wob_jonas> and the party that prints the bill has some idea on what format the receipt has, which is enforced partly by the computer program they use, partly by the employee sitting at the keyboard, and the organization receiving the bill has an idea on the format of the receipt, and they don't exactly match. I think in this case it's not a hard error, only t
19:20:06 <wob_jonas> he receipt I got doesn't match the best practices according to the receiver, but it's still acceptible, but I'm not sure.
19:21:19 -!- XorSwap has joined.
19:24:28 <wob_jonas> Sometimes you get similar problems between software libraries (or hardware components, but I don't work with those).
19:27:04 <wob_jonas> Also, I realized that the uncommon Hungarian compound word "házhozszállítás" has z+sz in it. Now I wonder what are the most common words that have zs+z, but that's probably rarer.
19:28:33 <wob_jonas> "házszám" also has z+sz, and it's probably more common than "házhozszállítás"
19:31:49 <wob_jonas> z+s and sz+s are in some even more common uncommon words, mostly derivatives of "igazság" and "egészség" resp, but many others. s+zs is in "szemeteszsák", which I usually write is "szemetes-zsák" despite that the academy rules forbid that
19:45:20 -!- sebbu2 has changed nick to sebbu.
20:08:38 <esowiki> [[ESOPUNK]] N https://esolangs.org/w/index.php?oldid=57511 * Blacksilver * (+588) Basically just duplicated the manual...
20:08:51 <esowiki> [[ESOPUNK/Commands]] N https://esolangs.org/w/index.php?oldid=57512 * Blacksilver * (+1992) [Subpage]
20:13:10 <esowiki> [[ESOPUNK/Commands]] M https://esolangs.org/w/index.php?diff=57513&oldid=57512 * Blacksilver * (+0) Typo...
20:16:54 <zzo38> What example might you make up of a word of a alien concepts?
20:18:18 <wob_jonas> zzo38: the style for those words rather depends on the writing style of the author or franchise
20:18:49 <wob_jonas> zzo38: do you want examples of well-known words from fiction?
20:18:53 <zzo38> Ah, I suppose, although I am the author for now
20:19:00 <zzo38> Give whatever examples you know
20:19:06 -!- XorSwap has quit (Ping timeout: 272 seconds).
20:19:40 <wob_jonas> zzo38: do you have a setting in mind? like some sort of sci-fi with extraterrestrial aliens (Star Trek), space fantasy (eg. Star Wars), something else?
20:20:07 <zzo38> It is actually a role playing game
20:20:26 <wob_jonas> zzo38: https://worldbuilding.stackexchange.com/ and https://scifi.stackexchange.com/ and their chatrooms can help a lot in this, the latter if the setting is sci-fi or fantasy
20:21:48 <wob_jonas> zzo38: ok, but role-playing game in what sort of universe? mediaeval magical fantasy? how alien are the aliens? just LotR or D&D elves? LotR or D&D orcs and goblins and trolls and ents? buglike aliens with human mind from a nearby star? modern theme park versions of chthulu mythos?
20:22:13 <Blacksilver> Search up Artifexian on Youtube, he has great conlang and worldbuilding videos.
20:23:33 <wob_jonas> zzo38: if it's similar to D&D settings, and you don't want well-known examples but a random model for potentially new words, then http://www.d20srd.org/fantasy/name/ can help.
20:23:47 <zzo38> It is GURPS and it is mediaeval, including old Lsd money.
20:23:54 <wob_jonas> Reading similar existing books or other fiction always helps of course.
20:24:52 <zzo38> "Lsd" is a Latin abbreviation for "pounds, shillings, pence"
20:25:28 <wob_jonas> that system that's derived from the even worse ancient roman currency system
20:27:15 <wob_jonas> I wasn't familiar with the abbreviation, only the concept. (I know "NIS" and "ILS" are abbreviations for the same currency though.)
20:27:31 <zzo38> O, there are other abbreviations too
20:27:50 <wob_jonas> sure. there's always lots of abbreviations.
20:27:59 <Blacksilver> Fun fact: The latin root for "Salary" is "Sal", meaning Salt. Because roman soldiers were paid in salt.
20:31:15 <wob_jonas> There are some common patterns that make them conflict a lot too: there are a lot of three or four letter English abbreviations with "o" in the middle standing for "of", lots of Hungarian government-institution-related abbreviations starting with "M" for "Magyar" or lately "N" for "Nemzeti" (these are just single letters, so worse than the old syst
20:31:15 <wob_jonas> em of calling everything "K&K" for "Kaiser und König")
20:32:07 <wob_jonas> (there were some socialist system abbreviations in between, but the names for them have been purged since, so I'm familiar with only a few)
20:42:26 <zzo38> I can describe how is my character (named Ziveruskex (sometimes abbreviated as Ziv, sort of like Lempel-Ziv-Welch)) (and his friend, named Strixan) is like, in case they would explain the language to someone who can know such thing
20:43:44 <shachaf> What do Lempel and Welch have to do with it?
20:43:55 <zzo38> Nothing; it is just a similar name
20:44:11 <arseniiv> zzo38: alien concepts. E. g. aliens have usually marriages of three individuals and behavioral roles in these marriages are traditionally distinct and have their names and these are better to not translate lest to make a misunderstanding
20:44:59 <zzo38> Yes, some things are difficult to translate (even among similar languages actually; such alien concepts will therefore be even more difficult)
20:48:31 <zzo38> They have five eyes, can see five colours (rather than three, that human do), antennas, scales, feathers, wings to fly, four fingers instead of five, sharp beak, and also a breath attack, and can eat blood
20:49:41 <zzo38> Yes, therefore the word for the colours will be difference, even some thing that might seem same if you can see three colours, seem to difference instead.
20:50:15 <int-e> 5 acoustic sensors, and 3 vocal organs; language depends on subtle interferences that need 4 functioning sensors to discern reliably
20:52:31 <int-e> You can make this arbitrarily complex. (This idea arose from an earlier question about a universal "IPA"; with wild thoughts like these you may have to record full 3D wave functions to capture all audible signals.)
20:53:21 <Blacksilver> Could be interesting to make a species that is 5-way symmetric.
20:53:40 <zzo38> Blacksilver: Yes, you could try to do that too
20:54:27 <HackEso> Twinkle, twinkle, little star!
20:54:54 <wob_jonas> int-e: 5 acoustic sensors => humans have 5.1 ears too according to the people who sold all those home movie sets that were so fashionable ten years ago (and the period in "5.1" has the same significance as the one in "8.3 char long filenames on FAT"
20:55:02 <int-e> `` ln -s \* wisdom/☆
20:55:43 <int-e> wob_jonas: well, we have two ears and we wander around (which I believe is the real selling point of more than 2 channels)
20:56:23 <wob_jonas> int-e: you don't wander around when you're sitting on a couch watching a movie, except perhaps for getting up for more pop-corn from the kitchen
20:56:29 <int-e> that said... I have two small speakers in front of me :P
20:56:33 <Blacksilver> 2 speakers would work if people didn't turn their heads.
20:57:13 <Blacksilver> If the sound is supposed to be coming from behind you, you don't notice if you're facing the tv.
20:57:27 <Blacksilver> But if you turn your head 90 degrees it breaks the illusion.
20:57:39 <Blacksilver> Hence you need speakers behind the listeners
20:57:43 <int-e> makes sense. but then I can no longer see the monitor
20:58:07 <int-e> in any case it's a better explanation than wandering around, thanks
20:58:15 <Blacksilver> You can detect the forward/back even if you turn you head a little bit. Brains are good at that sort of thing.
20:59:34 <int-e> Hmm... so are people making headphones with location/orientation sensors?
20:59:36 <Blacksilver> You can sometimes do it without moving your head because of how the sound bounces off the walls or something.
20:59:53 <Blacksilver> I wouldn't be surprised. Wouldn't be *too* hard.
21:00:27 <Blacksilver> Gyroscope, acceleromoter, some clever algorithms.
21:00:29 <zzo38> I have two external speakes on my computer but sometimes only one of them works
21:00:38 <wob_jonas> 3 vocal organs is harder, but you could argue that singing plus a guitar is common, then the mouth cavity and the nose cavity are separate resonators, or your vocal cords in your larynx for normal speech and your stomach for barfing and your buttock for farting are separate reeds for starting a vibration, or that people use their hands for clapping
21:00:38 <wob_jonas> and sometimes various beatbox effects, or that drummers play on half a dozen different drums.
21:03:21 <wob_jonas> "language depends on subtle interferences that need 4 functioning sensors to discern reliably" => now that's the kind of complexity we only reached after the second world war, with GPS receivers and mobile phones and astronomy instruments and near field electron microscopes and gravitational wave detectors and other such complicated devices,
21:03:23 <int-e> well if drums count you should consider an organ
21:03:55 <wob_jonas> and in these, the processing of those interferences is done autonomously, much more so than when someone plays a guitar or drum the traditional way.
21:04:37 <wob_jonas> so the human definitely isn't the one who decodes the four functioning sensors together.
21:05:43 <wob_jonas> Sometimes people get information from vision and hearing, or vision and hearing and smells, or vision and touch etc combined, but I don't think that it's ever "subtle interferences that need 4 functioning sensors to discern reliably"
21:07:41 <int-e> Well it was supposed to be an alien idea :P
21:07:47 <wob_jonas> hmm... perhaps if you counted three types of color sensor cells in each of humans' two eyes, sometimes combined with switching some color or polarization filter or lens, to make a visual image of something, and the processing done by the human brain alone
21:09:02 <arseniiv> I like how GPS design needs to consider Relativity (all^W both of them! — effects of gravitational potential and high velocity are comparable in magnitude). Of course it could be designed using empirical time dilation observations, but it’s better to have them some base
21:09:18 <wob_jonas> Like when I noticed that (it's hard to play the rubik cube because two of the stickers appear nearly the same color) iff ((I'm under old yellow halogen street lights) or (in the cheaper LED lighting of the carriages of the M4 metro and I'm wearing my brown color filtering sunglasses))
21:10:05 <int-e> sodium vapor lamps are the worst in that regard :P
21:10:12 <wob_jonas> arseniiv: GPS design needs to consider a lot of things. it's a wonder it works. it needs an accurate time based maintained by clocks on the satellites synchronizing to ridiculously precise atomic clocks on Earth,
21:10:14 <arseniiv> brown color filtering seems something hi-tech
21:10:47 <arseniiv> wob_jonas: indeed a wonder. I marvel at such technological advances
21:11:40 <wob_jonas> and since there are now four different GPS-like satellite positioning systems and they are maintained by four different secretive government organizations, I wonder if they could get into an accidental dependence loop when each of them trusts time or location readings of one of the other and results in a runaway unstable error.
21:13:38 <wob_jonas> they also need to measure all sorts of anomalies in the gravitational fields of Earth indirectly to figure out the trajectories of the satellites, plus tectonic movements to figure out the trajectory of ground-based stations. it's horribly complicated, and another possible source of dependence loop among different secretive organizations.
21:16:07 <wob_jonas> And then of course all the ground stations use modern digital processors and other huge integrated circuits that are horribly complicated to create, and are designed by americans who then manufactured by east asians and both of them can introduce subtle backdoors that are impossible to discover and could cause basically arbitrary behavior changes.
21:17:39 <wob_jonas> Dedicated hackers with lots of free time have reverse engineered basically the entire design of the MOS 6502 chip, but that only has a few thousand transistors and the entire layout was designed by one genius person.
21:20:33 <wob_jonas> But these days, CPUs are much more complicated. Intel and AMD and ARM deliberately provide enough technical info about some parts of the CPU so that some dedicated reverse engineers like Agner Fog still make independent discoveries about the newest models, but it's well known that it's easy to add such subtle backdoors that are impossible to discov
21:20:33 <wob_jonas> er unless you know some hidden information, and with some ingenuity, it's also possible to install backdoors that are plausibly deniable even if some super-intelligence who can reverse-engineered the entire CPU and discovered it as a bug.
21:20:48 <wob_jonas> s/ation, and with/ation; and with/
21:22:26 <wob_jonas> arseniiv: it's expensive, mostly because this type of high refraction index Zeiss glass lens is already expensive and there's lower demand for the tinted sunglass version.
21:24:12 <wob_jonas> arseniiv: but I don't think the brown color filtering itself is very high tech. grinding the glass lenses to an assymetric shape, that's high tech. all the fancy giant optical astronomical telescope mirrors and lenses and all the expensive camera lenses are all ground to spherically symmetric, which is easy.
21:25:23 <wob_jonas> glass isn't easy to shape to an exact shape that is smooth in the ten nanometer sclae.
21:28:46 <wob_jonas> plastic and silicone lenses (used in cheaper eyeglass lenses and all contacts) are easy compared to that, because you can mold them and reshape them after heating, instead of just grinding.
21:28:50 -!- LKoen has joined.
21:30:51 <wob_jonas> though admittedly, contact lenses are made of particularly tricky kind of polymer materials.
21:31:32 <wob_jonas> not those torture instruments that were in use thirty years ago on actors.
21:32:31 <wob_jonas> but women's self-torture fashion accessories don't get better, they just invent some new fashion as soon as a less painful replacement for an older effect is found.
21:33:02 <wob_jonas> I always laugh on stereotypic women's footwear
21:39:38 -!- Blacksilver has quit (Ping timeout: 245 seconds).
21:41:52 -!- Esqapezord has joined.
21:43:47 -!- Shragazord has quit (Ping timeout: 240 seconds).
21:49:06 -!- erkin has quit (Remote host closed the connection).
21:50:50 -!- Blacksilver has joined.
21:52:22 -!- Blacksilver has quit (Client Quit).
21:52:41 -!- Blacksilver has joined.
22:18:57 <arseniiv> wob_jonas: that was a joke, like, “brown color filtering sunglasses” can be parsed as a “((brown color) filtering) sunglasses” i. e. sunglasses that filter brown color (which is semantically funny too) :)
22:19:36 <zzo38> The computer should be made that does not have too much complicated
22:23:57 <rain1> youknow the trusting trust infected compiler idea?
22:24:08 <rain1> there is 'rotten' scheme compiler which is a real example of it
22:24:11 <rain1> are there any other examples?
22:25:14 -!- pierte has joined.
22:25:31 <rain1> and are there any examples of 2 compilers where it infects both compilers
22:25:55 -!- pierte has quit (Remote host closed the connection).
22:30:44 <rain1> you can think of the infected binary as having inside it 2 different infection genetic codes, and it acts as a multi-quine depending on who compiles it
22:33:14 <wob_jonas> arseniiv: nah, sorry. brown tinted sunglasses that only have a color filter, no polarization filter
22:33:56 <wob_jonas> well, they also have an interference-based anti-reflection coating, but so do my normal lenses. but definitely no polarization.
22:34:55 -!- sebbu has quit (Read error: Connection reset by peer).
22:35:24 -!- sebbu has joined.
22:36:33 <wob_jonas> by normal, I mean the non-tinted ones in my glasses that aren't sunglasses. they're still similar expensive Zeiss glass lenses, and these days they're the kind that are produced on-demand (rather than produced in advance and stored until someone orders the lens) with 1.8 reflective index, although I only settled on this type of glass a few years ag
22:36:58 <arseniiv> rain1: I heard about a Delphi (old days, v7 and alike) package virus that inserted itself in package binaries and then if a program compiled with them arrived on a computer with another Delphi instance, the virus copied itself in its packages etc.
22:38:39 <wob_jonas> arseniiv: are you sure the compilation itself was involved? didn't it just infect files using the standard library that programs compiled by that compiler were normally linked to?
22:39:24 <wob_jonas> involving the compiler is possible, but seems unnecessary. attaching to a library, on the other hand, can make the virus simpler and less prone of errors
22:39:39 <rain1> i like the idea for the simpler virus
22:40:51 -!- 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.”).
22:41:25 <wob_jonas> The marketing name for the type of these lenses, in case it matters, is "ZEISS SV Sph Min 1.8 with Super Filter ET" for the normal glasses, and "ZEISS SV Sph Min 1.8 with Umbra SE 85%" for the sunglasses.
22:43:05 <wob_jonas> I tried lenses with even higher reflective index, but they have a problem with breaking up colors everywhere except the center of the lens, which is still present with this type but to a lesser amount.
22:44:56 <wob_jonas> Also some of the lenses have a shape distortion near their sides that makes it very hard to synchronize the image from the two eyes and get proper binocular vision (not even seeing distances, just seeing the same object from both my eyes),
22:45:26 <wob_jonas> but it's not clear if this is influenced by the higher reflective index or more the distance and angle of the lens from my eyes which depends on the shape of the frame.
22:46:18 <wob_jonas> In any case, I'll probably try contact lenses (again for the third time after two failed attempts lots of years ago) next year. Contact lenses give much better visuals than eyeglasses:
22:47:56 <wob_jonas> they're very close to my cornea and lens so they don't shrink the size of objects by ten or twenty percents, so everything is bigger and so better, they have much fewer shape or color distortion, and the center of the lens, which always has fewer distortions, moves as I rotate my eyes.
22:49:26 <wob_jonas> Back when I tried the first and second time, they already gave an arguably nicer image for objects close to me, and a clearly sharper image for objects far from me, and that was back when contact lenses with cylindrical distortion weren't available.
22:50:09 -!- Blacksilver has quit (Read error: Connection reset by peer).
22:51:38 <wob_jonas> So for people like me with an almost strong myopia and some astigmatism, contact lenses should give a much better vision.
22:54:05 <wob_jonas> The problem is that I found contact lenses uncomfortable to wear. That has three potential causes. One is that contact lenses back then genuinely were worse than they are now, so that will be partly fixed by modern contact lenses. The second is that I have dry and irritated eyes due to spending lots of time in front of a computer monitor, and
22:54:10 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
22:54:27 -!- Blacksilver has joined.
22:55:38 <wob_jonas> that will be partly fixed by that I will now try daily contact lenses (that is, ones you throw away after one use, rather than cleaning) and wear them only occasionally, not all the time, so mostly not when I'm sitting in front of the computer. The third is that my myopia is caused by too long eyeballs, and because of that, the curvature of my eyeb
22:55:39 <wob_jonas> all and especially the part with the iris is strong, so contact lenses don't fit on it easily.
23:03:10 <arseniiv> wob_jonas: yeah, it injected itself into standard packages
23:07:40 <wob_jonas> arseniiv: admittedly hooking into the operating system kernel or the on-disk OS part of the boot loader is even easier, since those were often even more standard than some compiler libraries.
23:08:14 <wob_jonas> or hooking into something in 16-bit Windows
23:08:46 <wob_jonas> something that gets ran in practically every installation of the target version of Windows or versions close to it
23:10:26 <wob_jonas> although the Windows ones need some other component to spread, because people rarely copy their windows installations, so that's not a good idea
23:10:51 <wob_jonas> it works as a worm that survives reboots, but not as a virus
23:12:46 -!- ais523 has joined.
23:13:22 -!- wob_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
23:14:22 -!- wob_jonas has joined.
23:15:23 <wob_jonas> Hi ais523. I was just ranting about my vision problems and corrective eyeglasses and contacts, which I probably understand somewhat; plus butted in a little about old PC viruses, which I really don't know much about.
23:15:47 <ais523> I probably know more about old PC viruses than your vision problems
23:15:57 <ais523> I remember, many many years ago, reading the virus descriptions in a virus database for fun
23:17:42 <ais523> actual computer viruses are incredibly rare nowadays
23:18:06 <Blacksilver> Until a security hole comes out, then we get a bunch at once.
23:18:09 <ais523> because modifying executables is a) very difficult, b) pretty much always unnecessary given modern computers' complexities
23:18:13 <wob_jonas> ais523: yeah, although the one that was mentioned above was an honest to goodness virus, supposedly
23:18:24 <ais523> Blacksilver: you get hacks, worms, malware generally
23:18:32 <ais523> but a virus is a specific sort of that, one that's almost entirely died out
23:18:34 <wob_jonas> ais523: these days it's much easier to install trojans through all sorts of internet services
23:18:55 <ais523> a virus is malware that works by modifying executables to run the malware code when executed (instead of or in addition to their own code)
23:19:01 <Blacksilver> Oh, viruses specifically, not including trojans and worms and stuff.
23:20:11 <wob_jonas> rain1: Andy Weir's web-original short story "Twarrior" http://www.galactanet.com/oneoff/twarrior.html has a passing mention of a compiler hack
23:20:41 <wob_jonas> it's a piece of fiction short story
23:21:00 <rain1> https://www.teamten.com/lawrence/writings/coding-machines/
23:21:04 <rain1> il read it now and
23:21:07 <rain1> this was also fun..
23:21:09 <ais523> re: the trusting trust discussion earlier, it struck me a while ago that an interesting esoproject would be to create a compiler chain starting from first principles, and starting on a system that's not complex enough to contain AI-complete exploits in the hardware
23:22:15 <rain1> yeah! some fellows are trying to do that for GNU Guix
23:22:21 <rain1> https://bootstrapping.miraheze.org/wiki/Mes
23:22:30 <rain1> i like these bootstrapping projects
23:22:36 <rain1> there are a few , but they dont all go so far
23:22:51 <wob_jonas> ais523: I was also complaining about a case when a government-related agency defines one side of an interface, and a store with its computer software and training of the workers defines the other side of an interface, they're supposed to be compatible but there seems to be a subtle difference (hopefully it's not a hard error, just not following bes
23:23:26 <wob_jonas> and I can't do anything because both of them are rightly convinced that they've been using this interface for years in thousands of cases and I'm new to it so they know the interface better and don't have to believe me.
23:23:42 <ais523> well, my router's understanding of USB is apparently so far from standard that if I try to use it over USB, the Linux kernel resets it repeatedly because of all the standards violations
23:24:03 <ais523> and yet it's still somehow being sold for the general public
23:24:12 <wob_jonas> At least there's a power unbalance, because the government side has the right to define the interface, so in the worst case I can convince the store by getting proof from the government side.
23:24:15 <ais523> (luckily it works over wi-fi too, ethernet-over-USB is pretty niche as technology goes anyway)
23:24:38 <wob_jonas> In some other situations, especially in software, it's not clear which side gets to define the interface, in which case there's not much you can do.
23:25:14 <wob_jonas> ais523: um, why is your router connected by USB?
23:25:38 <ais523> it's too physically small to fit an ethernet port on it
23:25:41 <ais523> so it uses micro-USB instead
23:25:42 <wob_jonas> ais523: which generation of USB? USB3/3.1 or USB1/2?
23:25:51 <ais523> not sure, it doesn't get that far :-P
23:26:01 <ais523> probably 2 though based on how old it is
23:27:05 <wob_jonas> ais523: pity. if it were 3, you could try force-downgrading to USB 2 and possibly not lose bandwidth if you only need the bandwidth of an internet connection, although these days I think internet connections in the UK can be pretty fast.
23:27:36 <wob_jonas> "it doesn't get that far" -- that means it's USB 2, because I think USB 3 is bootstrapped through USB 1/2
23:27:49 <ais523> re: reverse-engineering the 6502, it was decapped, which is more like reading the source code than reverse-engineering
23:27:58 <ais523> you can look at the traces on the chip and see where the transistors are
23:28:10 <wob_jonas> ais523: have you tried plugging it into another USB port of the computer? that works better in practice than it should according to theory.
23:28:26 <ais523> and there are emulators that emulate the individual transistors
23:28:30 <ais523> wob_jonas: I've tried it on multiple different computers
23:28:58 <ais523> not only that, but a Linux kernel dev inspected a usbtrace and said that what the device was doing is completely broken
23:29:07 -!- sebbu2 has joined.
23:29:09 <wob_jonas> ais523: yes. but then they actually figured out most part what each transistor did, with the help of the lot of earlier information about the behavior discovered earlier.
23:30:03 * ais523 suddenly realises that physical hardware normally doesn't have comments
23:30:38 -!- tromp has quit (Remote host closed the connection).
23:30:47 <wob_jonas> ais523: it's still possible to decap the 6502 and look at some of the low-level electronics, you just need to separate more layers and a view with more expensive microscopes (but microscope technology keeps improving, we now have near-field microscopes to overcome the barrier of the frequency of the scanning wave, and stuff like electron-microscope
23:31:00 -!- ais523 has quit (Quit: sorry for my connection).
23:31:13 -!- ais523 has joined.
23:31:21 <wob_jonas> ais523: it's still possible to decap the 6502 and look at some of the low-level electronics, you just need to separate more layers and a view with more expensive microscopes (but microscope technology keeps improving, we now have near-field microscopes to overcome the barrier of the frequency of the scanning wave, and stuff like electron-microscope
23:31:41 <ais523> nowadays I normally have the logs open in stalker mode while I'm on IRC
23:31:41 <wob_jonas> only the electronics is so huge that it'd be very hard to scan it all, and almost impossible to understand it all.
23:31:45 <ais523> so if I have to reconnect I see what I've missed
23:31:57 <ais523> (and also what failed to get through)
23:32:13 <ais523> it's like a bouncer except with many fewer moving parts
23:32:36 -!- sebbu has quit (Ping timeout: 252 seconds).
23:32:51 <wob_jonas> ais523: larger pieces or more amateur physical HW often have comments, sometimes in the form of debugging connections, sometimes readable letters printed in ink or drawn in wires on the circuit board. a CPU probably has very few of those.
23:33:53 <ais523> motherboards still often have the components labelled, don't they?
23:33:53 <wob_jonas> even larger pieces can have all sorts of written tags attached, stickers and little plastic number tags, plus lots of useful coding conventions like physical arrangements and conventional colors of wires
23:33:58 <Blacksilver> There's nothing physically stopping the PCB from having lettering written in lines that aren't connected to anything. The manufacturer doesn't charge for the amount of stuff etched.
23:34:03 <ais523> presumably to help reduce errors in assembly
23:34:05 <wob_jonas> just don't trust the conventions too much or you can get misled
23:34:29 <wob_jonas> ais523: some components are so small it's hard to fit a label that people can see.
23:34:33 <ais523> Blacksilver: technically speaking the more copper there is on the board the cheaper it is, because you need less acid to dissolve the unused copper from the blanks
23:35:00 <ais523> but that's balanced by the fact that putting unused areas of copper too close to used areas of copper tends to cause unwanted electromagnetic effects
23:35:31 <ais523> however, most circuit boards have a "silkscreen" layer which is basically just white ink that's printed onto the board, and I doubt most manufacturers charge for that by the pixel
23:35:47 <ais523> (and you'd want a silkscreen layer to put things like the manufacturer's name, copyright, etc.)
23:36:31 <ais523> some letters like K would also be hard to put down in copper, because the etching process works least well with acute angles at the edge of the non-copper region
23:36:48 <zzo38> What things might seem like same colours if you can see three colours but if you can see five colours (including near UV) then it seem like different colours?
23:37:02 <wob_jonas> ais523: Yeah, more than half of the time even photocopiers charge for the page rather than the ink, to keep accounting simpler, even though the ink costs might be a few times larger than the per-page maintenance cost.
23:37:10 <ais523> (in fact it's fairly common for circuit boards to use only 45 degree turns if they can help it, presumably to help reduce this effect further)
23:37:24 <wob_jonas> I have abused this by printing stuff on paper with large black areas.
23:37:27 <zzo38> Also, why does Firefox use the title of a webpage as the default filename instead of using the filename in the URL?
23:37:53 <ais523> zzo38: well, one example would be that most people can't distinguish between some greenish blue and some cyan because they both have the same effect on all three color channels, but if there were more color channels they might be distinguishable
23:38:12 -!- Blacksilver has quit (Ping timeout: 252 seconds).
23:38:25 <ais523> zzo38: probably because some websites have a URL-filename scheme that's incredibly verbose and contains large amounts of junk data
23:38:32 <wob_jonas> zzo38: probably because there are so many server-side dynamic pages when the base filename part of the url is the same in lots of similar pages you want to save together. the title is sometimes the same too, but it might still be better.
23:38:42 <ais523> curl uses the URL-filename by default and it sometimes contains hundreds of bytes of hex or base64
23:39:05 <ais523> err, I meant wget, but it's probably the same for curl too
23:39:41 <wob_jonas> ais523: curl uses the URL filename if you use the -O command-line switch. it writes content to stdout by default. close enough. correct for wget.
23:39:50 <zzo38> Actually I find the filename in the URL is usually better, at least if it ends in .htm or .html it is usually better; if it ends in .php then neither the title nor the URL filename are suitable
23:40:20 <ais523> what if it ends in &.jpg?
23:40:45 <wob_jonas> ais523: for wget, it's actually hard to override this per url if you want to download multiple pages from the same server, which was my main motivation for making the wgetas script.
23:40:48 <ais523> (implying that the filename is made up of lots of query parameters, and then one was stuck on the end to give an extension)
23:41:09 <wob_jonas> ais523: browsers usually append a file extension if the normal one wouldn't be right
23:41:39 <ais523> sometimes people do that in URLs too though
23:42:21 <wob_jonas> ais523: the Stack Exchange chat parser is the only thing I've seen that pays attention to the "extension" at the end of a URL to guess whether it should try to download the URL and display it as an image. But the Stack Exchange parsers handling URLs is particularly stupid and annoying in multiple other ways.
23:42:44 <ais523> isn't that true about just about everything related to Stack Exchange's technology?
23:43:06 <wob_jonas> That extension thing can be considered a feature since it lets you override the default guess by appending a query parameter that the server ignores.
23:43:30 <wob_jonas> But it's only a "feature" as in a bug that lets you work around a worse bug.
23:43:46 <ais523> now I'm reminded of Wikipedia's use of text/css for serving text files
23:44:00 <ais523> (because IE interprets text/plain as HTML if it looks like it might be)
23:44:50 <wob_jonas> I rant about wikis and their formatting languages regularly here.
23:45:07 <ais523> there was a successful XSS attack against the #esoteric logs once using that, all the <nick> notation made it look HTMLy enough to fit a <script>alert("test")</script> in there
23:45:23 <ais523> and IE actually ran it
23:45:33 <ais523> that was an older set of logs, though, I think the current set are HTML already
23:46:14 -!- tromp has joined.
23:46:18 <wob_jonas> ais523: oh, is that related to the attack where some browser ran apparent javascript found in what's supposed to be an image file and ran it in local context so let it access local files?
23:46:38 <wob_jonas> ais523: the current log has three formats, two of which are text
23:47:25 <wob_jonas> anyway, Firefox has some such format recognition magic too. they spread quickly to support broken webpages
23:47:45 <wob_jonas> yeah, https://esolangs.org/logs/2018-08.html was my previous rant about wiki formatts
23:48:04 <ais523> well, Markdown is terrible for the purpose :-)
23:48:20 <zzo38> You can add a mention that the formatted text logs are not for use with Internet Explorer.
23:49:19 <wob_jonas> ais523: Markdown is the base. SE adds some additional levels of changes to the format.
23:50:06 <ais523> but SE Markdown doesn't even seem to have a general escaping rule
23:50:27 <ais523> that said, Github Flavored Markdown is just as bad
23:50:46 <zzo38> What format recognition stuff does Firefox do? In some cases the header of the file tell the difference for example it might treat image/gif as image/png if the header says it is PNG, but it should not treat text/plain as text/html or vice-versa. Treating text/plain as text/css might be helpful if the HTML elements specify that the file is CSS.
23:50:47 <ais523> I'm not sure if there's a way to produce the output <code>```</code> other than writing the code tags out manually
23:50:51 -!- tromp has quit (Ping timeout: 252 seconds).
23:51:06 <ais523> which means that you can't even express how to write Github Flavoured Markdown in Github Flavore Markdown
23:51:36 <zzo38> I know that in Markdown you can use plain text by starting with several tildes on one line; at least npm supports that.
23:52:22 <ais523> Markdown isn't really standardised at all
23:52:28 <ais523> there's Commonmark but I'm not sure anyone actually uses it
23:52:35 <zzo38> I think the MediaWiki format is better than Markdown
23:53:02 <ais523> and IIRC the original, markdown(1), is just a set of regexps or some similar architecture (i.e. there isn't even a parser)
23:54:50 <wob_jonas> ais523: SE format does have a general escaping rule AFAIK. ampersand-escaping key characters and representing line breaks as <br> works for all plain text modulo the HTML horizontal whitespace collapsing problem,
23:55:18 <ais523> now I remember the whole  thing
23:55:32 <ais523> and I'm still not entirely sure on whether that's useful or not
23:55:32 <wob_jonas> which is familiar to many users because it also works on some other wikis
23:57:36 <ais523> that said, hex-escapes aren't very usable as most people don't have the ASCII code of characters like ` memorised
23:58:03 <wob_jonas> and in posts (questions or answers) and tag wikis (but not in comments or chat), for making hyperlinks, the [link text][a] with later [a]: http://example.example/ in separate line seems to work most of the time (I admit I haven't tried the abomination IPV6 address HTTP urls), and HTML tags work for most of the other formatting available
23:58:29 <zzo38> If you are on Linux (possibly Mac OS X too, although I have not tried it) you can also use "man ascii" for the ASCII table.
23:58:37 <ais523> zzo38: yes, I do that frequently
23:58:54 <ais523> ` is just one of the characters I happen to have memorised, because I know it comes just before a
23:58:54 <HackEso> /srv/hackeso-code/multibot_cmds/lib/limits: line 5: exec: : not found
23:58:58 <wob_jonas> ais523: meh, those most people should ask the ascii or unicode code of characters from their computers, they're good at memorizing those.
23:59:35 <wob_jonas> I don't know the code of all characters ever either, but know the code for many, and know all printable ascii if pressed.