←2017-09-04 2017-09-05 2017-09-06→ ↑2017 ↑all
00:00:09 <ais523> I know why the shocklands are good
00:00:53 <wob_jonas> whereas I started playing when Dissension was already out, which is why I don't notice that those weren't first
00:01:12 <wob_jonas> sure, I do have some idea about earlier sets, but often forget
00:01:15 <ais523> I started playing with 9th edition
00:01:19 <ais523> and stopped early into lorwyn
00:01:49 <ais523> (i.e. 9th, ravnica block, coldsnap, time spiral block, 10th, and a bit of lorwyn)
00:03:04 <wob_jonas> (I know it was Dissension, not just Guildpact, because the person who introduced me brought Sky Hussar. I have seen people play M:tG before, but I haven't played.)
00:03:36 <ais523> dissension is an odd place to start, really
00:03:59 <ais523> one "problem" with the original ravnica block is that it only really makes sense as a combined entity, rather than being playable set-by-set
00:04:07 <ais523> although dragon's maze is even worse in that respect
00:04:18 <wob_jonas> Coldsnap may have been out, but only very recently, so I have only seen cards from it a month or two later when I bought my first cards, which was the Coldsnap black-blue theme deck.
00:04:33 <ais523> it did have one of the most intriguing names ever, though, people were asking for dragon's maze spoilers during rtr, rather than asking for the next set as usual
00:04:43 <wob_jonas> In retrospect, the black-blue Coldsnap theme deck is a bad one as a beginner product, but obviously I didn't know that.
00:04:51 <ais523> also I don't hate Coldsnap as much as most people do
00:05:31 <wob_jonas> ais523: I don't hate Coldsnap, I'm just saying it was not good as a first set of cards to buy.
00:05:42 <wob_jonas> It's fine to buy later.
00:06:10 <wob_jonas> But at this point it doesn't matter.
00:06:12 <ais523> I think people don't like it because it sucked to draft
00:06:27 <ais523> although people are bad at figuring out what draft environments are good until Wizards has already stopped supporting them
00:06:30 <wob_jonas> I have so many cards that one theme deck doesn't matter.
00:06:56 <ais523> this is probably what lead to the balance issues, Wizards deciding that trying to balance the sets carefully was a waste of money because they move onto the next set anyway before it's figured out
00:07:08 <wob_jonas> "supporting" as in you can no longer buy boxes cheap, or only that they are no longer played much in tournaments?
00:07:21 <ais523> they both happen around the same time, probably not coincidentally
00:07:54 <ais523> but when I first started playing Magic, I was under the impression that the game had been continuously refined and optimized over time and would always stay much the same
00:07:55 <wob_jonas> wait, it lead to the balance issues which time? I don't think that's happened the last time. It may have happened back ten years ago.
00:08:09 <ais523> seeing the name "ninth edition" made me think "ooh, this has had a lot of work put into it"
00:08:29 <ais523> but it turns out it's actually a game where they randomly change things every few months to keep it fresh, whether the old things worked or not
00:08:29 <wob_jonas> hehe... yes, that's a bit misleading because the second edition is exactly the same as the first
00:09:09 <ais523> beta and unlimited?
00:09:24 <ais523> it starts ABUR4 so I assume alpha and beta are collectively the first edition
00:09:28 <wob_jonas> but then, core sets from ninth to M12 did have a lot of good work put into refining it, perhaps more so then the block expansions
00:09:33 -!- imode has joined.
00:09:41 <wob_jonas> yes, AB are collectively the first edition
00:10:00 <wob_jonas> it's silly but they won't change the numbers now
00:10:07 <ais523> wob_jonas: well tenth was definitely an attempt to mix things up
00:10:09 <wob_jonas> because the numbers are printed on the core sets from fifth to tenth
00:10:11 <ais523> `card-by-name lightning bolt
00:10:12 <HackEgo> Lightning Bolt \ R \ Instant \ Lightning Bolt deals 3 damage to target creature or player. \ A-C, B-C, U-C, RV-C, 4E-C, M10-C, M11-C, MM2-U, MED-C, PD2-C
00:10:27 <ais523> wait, it was in M10 /and/ M11? I can't be reading that correctly
00:10:34 <alercah> nope that's right
00:10:38 <alercah> tenth was garbage though
00:10:40 <ais523> wow
00:10:42 <wob_jonas> yes, and it wasn't tenth
00:10:51 <alercah> M10 was when they shook things up
00:11:01 <ais523> one problem with numerical core sets is that it's hard to remember which ones are which
00:11:04 <wob_jonas> I like tenth, but mostly because of nostalgy. M10 might actually be a better set, even with all the new cards.
00:11:15 <alercah> tenth was the core set before M10
00:11:25 <ais523> right, I know it went tenth, M10
00:12:26 -!- FjordPrefect has joined.
00:13:12 <wob_jonas> what I don't like about the current direction magic is heading to is that they keep making changes that amount to printing more and more distinct cards per year. this started at approximately coldsnap or tenth edition. they always give very nice sugar-coated explanations of why the particular change is good, but the gross number of new cards per ye
00:13:12 <wob_jonas> ar always keeps going up, almost never down.
00:13:41 <wob_jonas> and some of the explanations are even kind of mutually contradictory.
00:14:45 <wob_jonas> it's like that time many years ago when they first raised bus prices to encourage people to travel by train which is cheaper to run, then later they raised train prizes to harmonize mass transport prizes and not disadvantage people who live in places with no train.
00:15:50 <wob_jonas> (this was before the time when they raised the prices of trains that stop at fewer stations, but basically every train stops at fewer stations, so you don't have a real choice. that one was much more obvious because they did it in a single step.)
00:16:03 <ais523> a while back in Birmingham, the bus and train companies went through a phase of repeatedly undercutting each other by 10p
00:16:06 <ais523> alas, it couldn't last
00:16:42 <wob_jonas> this is between cities, not within Budapest, in case that's not clear
00:17:23 <ais523> intercity train travel in the UK is very expensive except when it's randomly much cheaper
00:17:41 <ais523> coach travel can be very reasonable but it's slower
00:18:16 -!- augur has quit (Remote host closed the connection).
00:19:16 <wob_jonas> it's always expensive in Sweden, but there are surprising price differences between different types of trains and buses on the same line, depending on at least whether you forfeit your right to refund your ticket, and whether the price for reservations increases dynamically by time like with international airplane tickets. both of those categories
00:19:16 <wob_jonas> don't exist here.
00:19:48 <wob_jonas> there's also a difference between trains where you need a seat reservation versus no, which is a distinction that exists in Hungary but very often you only have one of those choices
00:19:51 <ais523> most train lines in the UK have four price categories: anytime, off-peak, super off-peak, advance
00:20:13 <ais523> having a reservation is orthogonal to this (although it'd be unusual to get an advance ticket without a reservation)
00:20:15 <wob_jonas> this is partly because south-Sweden is big and has multiple big cities and big airplanes, so there's a lot more train and bus travel than in Hungary
00:20:44 <ais523> although, many train companies, when selling advance tickets, just give you a ticket saying you're guaranteed a seat but don't allocate any specific seat, to save the trouble of putting the "reserved" signs onto the seats
00:21:36 <wob_jonas> ais523: sure, even though you get a seat number, in practice seat reservations work that way on trains in Hungary too (and often in cinemas)
00:21:48 -!- hppavilion[1] has quit (Ping timeout: 240 seconds).
00:22:22 <wob_jonas> that is, most passengers and the controllers don't care about the seat numbers unless there's a dispute between passengers
00:23:45 <wob_jonas> this is unlike airplane tickets or theatre tickets, where people only swap places locally among nearby ones in the same row or when the airplane stewards ask, in theatres because they're more full (a movie is cheap to repeat multiple times for the cinema) and in airplanes partly because of security reasons the airlines insist
00:25:11 <wob_jonas> also for some reason many people seem to seem to consider different seats in an airplane to have different values, so much that some airlines even sell the right to choose your seat at an extra price and a varying price depending on the position,
00:25:45 <alercah> seats with extra legroom fetch a premium usually
00:25:50 <wob_jonas> whereas on the trains where you can get a seat reservation, usually all seats are equivalent (this is sometimes not true for some trains with no reservation, because those are lower comfort)
00:26:49 <wob_jonas> alercah: yeah, but then I don't want extra legroom seat on airplanes because those are usually at the front row so you can't keep your luggage at your feet
00:27:11 <wob_jonas> the extra legroom is probably worth more on longer flights though, on which I rarely fly
00:27:50 <wob_jonas> on long bus trips, I'll certainly take the extra legroom any time, but that usually just manifests in the difference between a 40 seat bus versus a 50 or 63 seat bus, which you can't really change
00:29:34 <ais523> in the UK, most long-distance trains have two tiers of seats
00:30:04 <ais523> known as first class and third class for silly historical reasons, although the "third class" name typically isn't used nowadays (they have "first class" and "standard class" or "first class" and "not first class")
00:30:29 <ais523> but the vast majority of people don't buy first class seats, as they often aren't much better, and when they are, they're often very expensive
00:30:32 <shachaf> sounds like a topic for #trains, y'all
00:30:42 <ais523> and it's leading to something of a problem because of trains getting fully packed
00:30:50 <alercah> wob_jonas: the exit row in particular is what I was thinking of
00:30:57 <ais523> sometimes they have to open up the first class section, without first class advantages, to normal tickets
00:31:20 <wob_jonas> alercah: that has the same problem. no luggage at your feet because it obstructs the emergency exit.
00:31:38 <wob_jonas> ais523: yes, of course, but they don't sell that except for very expensively
00:31:42 <alercah> that's usually been allowed when I've been on flights
00:31:46 <ais523> wob_jonas: it depends on the route
00:32:18 <wob_jonas> most of the flights I've been on are cheap ones that don't even have first class seats (they do have middle class seats)
00:32:23 <wob_jonas> (well, some do)
00:33:48 <ais523> wob_jonas: I picked a random off-peak train journey from Birmingham to London to compare, far enough in the future that Advance tickets are still cheap (also on a Saturday so that I don't run into peak time restrictions)
00:34:09 <wob_jonas> the class distinction on trains is totally meaningless too these days by the way. most trains only have second class, because people just don't buy first class tickets, so the company just uses originally first class carriages as second class. and the distinction between different trains is much bigger than the class distinction.
00:35:06 <wob_jonas> I wouldn't be surprised if they removed the distinction entirely a few years into the future, it's just that MÁV is a huge company with lots of bureaucracy plus is regulated heavily by the state because it's mostly a monopoly which means even more bureaucracy, so changes take a lot of time.
00:35:07 <ais523> it costs £17 (London Midland) / £27 (Virgin) first class, or £6-£8 (London Midland) / £11-16 (Virgin) third class
00:35:17 <ais523> that's on Advance tickets, which get more expensive the more of them have been bought
00:35:31 <wob_jonas> ais523: no, I'm saying this for trains in Hungary
00:35:38 <ais523> let me check Chiltern
00:35:39 <wob_jonas> they can totally matter in other countries
00:35:43 <ais523> wob_jonas: I'm just interested to compare
00:35:52 <wob_jonas> ais523: are those on the same train?
00:36:27 <ais523> wob_jonas: same set of trains; Virgin and London Midland use similar routes, but London Midland's is less direct and therefore cheaper because it takes longer
00:36:30 <wob_jonas> different classes of trains are very real, although sometimes you have limited choices, but sometimes you do have a real choice, and the price difference may be zero or it may be bigger than the difference between first and second class.
00:36:57 <ais523> Chiltern is charging £5.50 to £12 on trains during the same set of days
00:37:07 <ais523> they have a notably different route but it's comparable to the other two
00:37:12 <zzo38> Make your own implementation of Magic: the Gathering in computer if you have a better idea.
00:37:32 <ais523> after all, they all go from Birmingham to London, and normally you wouldn't vary the route based on where in London you ended up
00:37:42 <ais523> as you'd probably have to make a journey within London anyway
00:37:51 <wob_jonas> ais523: are there ever three distinct classes of carriages on the same train? "third class" in Hungary is something from before I was born
00:37:58 <ais523> but Chiltern doesn't even seem to have a first class
00:38:10 <ais523> wob_jonas: no, basically what happened was that the government decided that trains were too expensive
00:38:20 <ais523> so asked trains to introduce third class tickets (that were cheaper than the existing second class)
00:38:27 <wob_jonas> expensive to run or expensive to travel on?
00:38:32 <ais523> and it wasn't long before second class had vanished altogether because first and third were just more sensible option
00:38:36 <ais523> *options
00:38:37 <wob_jonas> ah
00:38:39 <ais523> expesnive to travel on
00:38:41 <wob_jonas> so there's first and third class
00:39:00 <wob_jonas> makes sense
00:39:41 -!- hppavilion[1] has joined.
00:40:05 <ais523> that £5.50 to London on Chiltern is crazy, though
00:40:25 <ais523> admittedly it's in the middle of the day on a Saturday, in the heart of super off peak time, and it's booked like a month in advance
00:40:56 <ais523> but I hadn't realised that prices that low were actually achievable except on adverts showing the lowest possible price because that's what adverts do
00:41:33 <oerjan> <ais523> known as first class and third class for silly historical reasons <-- the first class gives you a better chance if the train hits an iceberg hth
00:41:56 <ais523> oerjan: on many services, first class is at one end of the train, often the front
00:42:04 <ais523> so it might give you lower chances in the case of a train hitting something
00:42:18 <ais523> although london midland put first class in the middle of a middle carriage, normally
00:57:29 <fizzie> I think at least on quite a lot of companies, it's now just first class and "standard class".
00:57:51 <wob_jonas> ais523: now as for your game design draft, you don't allow payments of moving a card other than the one that has the mano written on it, right?
00:58:12 <ais523> wob_jonas: wasn't planning to, at least to start with
00:58:41 <ais523> it might be possible to find a nondisruptive way eventually, probably as long as the other card moved doesn't touch the tactics track; you'd probably have to ban the card from being manoeuvred itself
00:58:44 <wob_jonas> ais523: so moving other cards can only be an action, and there are no restrictions on how much they can move a card at the same priority, and you just ignore impossible actions, right?
00:58:53 <ais523> wob_jonas: right
00:59:26 <ais523> I didn't mention it, but the intention is that any action that can't be fully carried out is entirely ignored (also, health can go negative), but with multiple actions, it's possible for one to fail and others to succeed
00:59:27 <wob_jonas> "beyond their current health" is probalby a minor typo
00:59:34 <ais523> err, probably
01:00:04 <fizzie> Virgin is one of those companies. ("Our WiFi is complimentary in First Class, and pay as you go in Standard Class, --")
01:00:28 <wob_jonas> what's missing to me from this document is listing manos that are common among a large class of cards (like all creatures, all lands, all spells, whatever)
01:00:55 <wob_jonas> also I don't quite understand how playing cards from your hand usually works.
01:02:01 <wob_jonas> it seems like they can't just go to the tactic from your hand, because then you couldn't distinguish between the card used from your hand or from the ready; but if they go directly to the busy or ready or attached, then how do you counterspell them?
01:02:08 <fizzie> [[ Trains in Great Britain provide a two-tier class structure, with the higher tier called "first class". The lower tier was re-branded from "third class" to "second class" by British Rail from 3 June 1956, and then to "standard class" from 11 May 1987. ]]
01:02:35 <ais523> aha, so that's what happened
01:02:43 <wob_jonas> obviously you could have cards that are directly spent from your hand or something
01:02:52 <wob_jonas> but cards have to get to the bf somehow
01:03:11 <ais523> wob_jonas: oh, the way it works is that most cards go from hand to tactic 1, but from play to attack / defend / tactic 2
01:03:22 <ais523> at least, that's the plan
01:03:32 <wob_jonas> hmm
01:03:42 <wob_jonas> so they can go directly to tactic 1 even if attack is empty?
01:03:50 <ais523> in some cases, the effect on being initially played is the same as the effect you get from an ability, so you could just "replay" it as a discount
01:04:06 <ais523> and yes, the same physical slot on the board is used for tactic 1 (when there's no attack) and for defending (when there is an attack)
01:04:18 <ais523> mostly so that 2 continues to have the same meaning, of a slot where you respond to your opponent's action
01:06:05 <wob_jonas> ah right, you even mention this as ""Deploy" is a very common manoeuvre name; it always includes "hand → tactics 1" and no positive resource usage, but the negative resource usages vary."
01:08:36 <wob_jonas> also "This might be able to do with an overspill" sounds odd to me, but maybe it's an Englishism
01:12:55 <ais523> wob_jonas: "overspill" isn't actually a noun, I'm just using it as one
01:13:06 <ais523> "with a consequence for overspilling" would be more correct
01:15:16 <wob_jonas> So can an enchantment add manos to the card it's attached to? Or is that forbidden?
01:18:47 <wob_jonas> (the ugly phrasing is because in M:tG, when you attach a Holy Strength enchantment to a Devoted Hero creature, then the Holy Strength is the "attached permanent" but the Devoted Hero is the "enchanted permanent" even though to attach something and to enchant something is sort of the same; and "equip" works the same as "enchant". the terminology is
01:18:47 <wob_jonas> confusing.)
01:19:14 <wob_jonas> (luckily most of the time in M:tG it's obvious what the words mean because the abilities make sense only for the aura or only for the creature.)
01:21:22 <ais523> wob_jonas: grafting manoeuvres onto cards seems like a reasonable use of an aspect
01:21:52 <wob_jonas> ok
01:21:54 <ais523> (also, writing that document made me much better at spelling "manoeuvre", it's not an easy word to spell)
01:22:13 <wob_jonas> yeah... although you could just choose a different word instead in that case
01:22:14 <ais523> although I think it's spelled "manoeuver" in US English? not sure on that
01:22:55 <wob_jonas> anyway, it's late so goodbye, channel
01:23:29 -!- wob_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
01:23:31 <ais523> night
01:32:50 <shachaf> @time ais523
01:32:52 <lambdabot> Local time for ais523 is Tue Sep 5 01:32:51 2017
01:32:57 <shachaf> I'm on and off at the computer.
01:41:07 -!- jaboja has quit (Ping timeout: 260 seconds).
01:42:19 -!- jaboja has joined.
02:01:09 <oerjan> . o O ( just spell it "manøver" hth )
02:04:07 <ais523> surely ø̈ would be a better vowel there?
02:10:03 -!- erkin has quit (Quit: Ouch! Got SIGABRT, dying...).
02:11:49 <oerjan> ais523: we don't use that in norwegian hth
02:36:18 -!- augur has joined.
02:43:55 <zzo38> ais523: I think the rule of your kind of game, to become more clear, to write the open-source computer program of it, possibly by the literate programming so that they same document can also be all of the same text, as well as the computer program codes.
02:45:15 <shachaf> zzo38: What I want, instead, is for card text to be written in a declarative programming language such that you can generate both behavior and readable English text from it.
02:45:27 <zzo38> I don't really care which word you use, but you should use words that you can write with ASCII and be understandable all of the card, even if you may use some non-ASCII characters in their document they should not be require
02:45:55 <shachaf> I think requiring ø is OK.
02:46:16 <zzo38> shachaf: That is another idea yes, and I have thought about that too. Although, I was talking about the rules rather than the cards.
02:46:58 <shachaf> Well, I'd prefer for most of the rules to be implemented in the same programming language as the cards.
02:47:01 <zzo38> But about requiring a slashed "o", well, I think that is not even the correct spelling anyways, so it does not make sense. And even if it is used in the word, it doesn't cause problem with ASCII-only communications anyways.
02:47:27 <shachaf> Only a small core language needs to be part of the definition of the game itself.
02:47:37 <shachaf> And I don't care whether the core language is specified in English.
02:48:02 <shachaf> zzo38: ASCII-only communication is outdated
02:48:06 <zzo38> shachaf: At least in the case of Magic: the Gathering that does make sense, but for ais523's game it seem difference
02:48:08 <shachaf> It has gone the way of Gopherr.
02:48:29 <shachaf> Oh, I wasn't talking about either of those games, but a hypothetical game I haven't described.
02:48:37 <ais523> shachaf: I like the idea of storing cards as AST and converting it to English and code
02:48:38 <zzo38> O, OK, then it works fine.
02:48:54 <ais523> although the English should convert back to the AST too, otherwise people wouldn't know what a card did by looking at it
02:49:14 <shachaf> ais523: Ideally the language describing card behavior would be simple enough that people could also read it directly.
02:49:17 <pikhq> @metar KAFF
02:49:17 <lambdabot> KAFF 050058Z AUTO 33015G23KT 8SM CLR 24/03 A3029 RMK AO2 PK WND 34026/22 SLP178 T02390030 $
02:49:31 <shachaf> But it would be good if there was a canonical translation back and forth to English.
02:49:32 <zzo38> ais523: Actually that idea I have thought of even for use with Magic: the Gathering too (but not the second part about back conversion, which would make the text a bit messy I think)
02:49:34 <pikhq> 8? I dispute that claim; that is too high.
02:49:35 -!- jaboja has quit (Ping timeout: 240 seconds).
02:49:40 <shachaf> @metar KOAK
02:49:40 <lambdabot> KOAK 050053Z 34006KT 9SM BKN140 BKN200 26/19 A2984 RMK AO2 SLP105 T02560189
02:49:46 <ais523> shachaf: perhaps it should be a subset of English?
02:49:54 <shachaf> ais523: Sure, just like Magic: The Gathering
02:49:54 <pikhq> The air would be nicer if it was breathable.
02:49:57 <ais523> M:tG is sort-of like a "programming language that's a subset of English" already
02:50:01 <ais523> but not that well specified
02:50:07 <shachaf> ais523: It's annoying that Hearthstone card text doesn't describe behavior completely at all.
02:50:19 <shachaf> In fact people have to reverse-engineer card behavior sometimes.
02:50:20 <ais523> hearthstone is trying to deal with screen size problems
02:50:30 <ais523> that's part of the reason why aspects, manoeuvres, actions all have names
02:50:36 <ais523> so that on a small screen, you can display just the name
02:50:54 <ais523> and people know what the names do most of the time, or if they don't, there'd be some way to look up the corresponding action
02:50:55 <pikhq> That 8 mile visibility is from smoke.
02:51:04 <shachaf> ais523: MtG card language is fairly well-specified, but it can also say whatever it wants.
02:51:30 <zzo38> shachaf: Yes, mostly it is well specified enough. Not quite, though.
02:51:50 <ais523> shachaf: I'd argue that it's very poorly specified, although rigid
02:51:50 -!- jaboja has joined.
02:52:00 <ais523> there isn't even a grammar, unless you count Alex Churchill's
02:52:01 <zzo38> And, "almost" isn't good enough.
02:52:03 <ais523> which is unofficial
02:52:14 <shachaf> ais523: I like using keywords and short descriptive words in card names.
02:52:29 <shachaf> But I kind of wish those were part of a "standard library", separate from the core game rules.
02:52:33 <ais523> actually I think one of my favourite innovations is cards referring to themselves in first person
02:52:40 <shachaf> Yes, I liked that.
02:53:25 <zzo38> Magic: the Gathering has text editing effects, which implies a AST anyways, even if nobody has ever written it down.
02:54:01 <shachaf> gain hug until end of turn
02:54:01 <zzo38> ais523: Yes I like that too, although it isn't only you who did that; someone else too has described a game they were making, with the same thing, to me.
02:56:48 <zzo38> My ideas for AST for Magic: the Gathering is one example can be: [:counter [:target :spell]] or [:counter [:target [:and [:not :black], :spell]]] and it can use precise rules (written in open-source computer code) to describe the exact conversion and meaning.
02:58:17 <zzo38> ("Counter" in Magic: the Gathering has another meaning too, which refers to a one-shot property, although in this AST the two kind of "counter" are two different words.)
02:58:42 <ais523> zzo38: I think RoboRosewater uses "unspell" internally?
03:00:32 <zzo38> OK, maybe, and yes that will work too. (Or maybe it uses "uncast"; I don't know, but that works too)
03:00:37 <shachaf> zzo38: Do you like the chain rule?
03:01:18 -!- ATMunn has quit (Quit: See ya! o/).
03:02:22 <alercah> zzo38: text editing doesn't require an AST
03:02:31 <shachaf> It just requires S
03:02:55 <alercah> oh wait
03:03:07 <alercah> I forgot that it refers to types
03:03:09 <alercah> so yeah it does
03:03:15 <alercah> nvm
03:03:34 <zzo38> What chain rule and what about chain rule?
03:03:46 <shachaf> The derivative one.
03:04:55 <zzo38> O, yes, OK
03:05:57 <ais523> the chain rule makes intuitive sense if you see derivatives as limits of deltas
03:05:58 <shachaf> The chain rule makes a lot more sense when you add types.
03:06:03 <shachaf> It's the only thing that type-checks.
03:06:13 <shachaf> ais523: How do you mean?
03:06:23 <ais523> shachaf: there's two ways to write it
03:06:33 <ais523> dy/dx = dy/du × du/dx is the one I'm thinking of
03:06:58 <ais523> the other version is less symmetrical, shorter, and comes to the same thing
03:07:01 <shachaf> That's the best way to write it, though I haven't figured out quite what it means.
03:07:26 <ais523> shachaf: basically, y is a function of x, but can also be written as a function of u, where u is a function of x
03:07:32 <shachaf> Yes, I know that much.
03:07:47 <ais523> so you can differentiate y with respect to u, because y is a function of u
03:07:49 <shachaf> I like Leibniz notation a lot. But I don't really know how to formalize it.
03:07:58 <ais523> and you can differentiate u with respect to x, obviously
03:08:36 <shachaf> Hmm, do people often write things like "d(x^3)/d(x^2)"?
03:08:37 <ais523> for a silly example, let's say y=u², u=sin x; then we have dy/du=2u, du/dx = cos x
03:08:58 <ais523> then we just multiply them together and get 2u cos x = 2(sin x) cos x
03:09:08 <ais523> which is dy/dx in this case
03:09:11 <shachaf> Yes, I know that.
03:09:40 <ais523> the notation only works if you have one specific independent variable identified (in this case, x)
03:10:04 <ais523> d(x³)/d(x²) is meaningful if you treat x as the variable in question (it's d(x³)/dx ÷ d(x²)/dx)
03:10:35 <shachaf> I think it works in some other cases as well.
03:10:48 <ais523> and the way to think about dy (where y isn't your independent variable) is that it means "the value of y when x has a slightly higher value, minus the value of y right now"
03:11:04 <ais523> (where the "slightly higher value" would be x+dx)
03:11:37 <ais523> in the case where the derivative isn't continuous, "slightly higher" and "slightly lower" may give different results, but that's a case that's typically ignored at lower levels of maths
03:12:03 <ais523> anyway, the point is that you take limits, so you take the limit of the derivative as dx shrinks to 0
03:12:22 <shachaf> You know the thing people do -- I think we've talked about this before -- where they have "x^2 + y^2 = 1", and then "2 x dx + 2 y dy = 0", and then "dy/dx = -x/y"?
03:12:29 <ais523> but clearly, you have dy/du×du/dx=dy/dx while dx is positive, so you wouldn't expect it to change in the limit
03:12:50 <ais523> shachaf: yes, there are a bunch of rules for whether that's safe to do, and I'm not sure I've ever seen them written down
03:12:52 <shachaf> It's true that you can parameterize x and y in terms of some other variable, but it doesn't matter which one you use.
03:13:10 <ais523> in this case, I'd probably define x = cos t and y = sin t
03:13:51 <ais523> but the point is that the exact parameterization can matter, especially when multiple x correspond to a single y
03:13:54 <ais523> or vice versa
03:14:19 <shachaf> Can it?
03:14:22 <shachaf> What's an example where it matters?
03:14:48 <shachaf> Anyway, I was thinking of the chain rule in multidimensional cases, since the single-dimensional version loses some important content.
03:15:02 <ais523> basically, think of x²+y²=1 as a curve (in this case, a circle); in order to calculate things like dy/dx, you have to imagine a point moving along that circle, and dy/dx becomes the ratio of its y velocity to its x velocity
03:15:03 <shachaf> (dy/du * du/dx is really composition of linear maps, or multiplication of matrices; it's not commutative)
03:15:22 <ais523> obviously, it doesn't matter how fast it's moving (as long as it has nonzero speed) in the case of the circle
03:15:31 <ais523> but if you can imagine a more complex curve, say one that crosses itself
03:15:49 <ais523> the speed of the point is going to depend on which path it takes
03:16:03 <shachaf> So you're parameterizing everything in terms of time to make it work.
03:16:09 <ais523> a really degenerate example where the theory breaks down is 0xy=0
03:16:17 <ais523> you can find the derivative at any point to be anything there
03:16:51 <ais523> and sure, dividing by 0 is normally cheating, but you need a rigid set of rules to make sure you aren't doing it in any given case
03:17:13 <shachaf> One thing you can say is, the curve is defined by the zero set of f(x,y) = x^2 + y^2 - 1
03:17:23 <ais523> yep
03:17:40 <ais523> I think this sort of derivative shenanigans only works at points on the curve which have a unique tangent
03:17:41 <shachaf> And the derivative is defined by the zero set of Df(x,y)(dx,dy) = 2x dx + 2y dy
03:17:53 <shachaf> Or rather the derivative at each point
03:17:59 <ais523> and even then, you do have to uniquely define the point (which may involve specifying both x /and/ y, or using a third variable to parameterize)
03:18:16 <shachaf> I'm quite happy using both x and y
03:18:46 <shachaf> For example, even for the half-circle sqrt(1-x^2), "-y/x" is much nicer than "-sqrt(1-x^2)/x"
03:19:05 <shachaf> And also it works for a circle of any radius.
03:19:37 <ais523> right
03:19:53 <zzo38> Is easy what it mean, if you multiply dy/du by du/dx then the du are cancel out, it look like.
03:20:04 <ais523> actually I think what typically goes wrong with this approach isn't necessarily in the dx/dy manipulation itself, but in trying to write x or y in terms of each other when it isn't unique
03:20:12 <shachaf> zzo38: It look like, but is it actually?
03:20:43 <ais523> zzo38: right, the dx and dy symbols don't necessarily obey all the normal rules of mathematics, so you often have to prove any specific thing you can do with them separately
03:20:48 <ais523> but in this case they do follow normal rules
03:22:00 <shachaf> ais523: I'd like them to be something other than symbols.
03:22:08 <shachaf> For example sometimes people define them to be differential forms.
03:22:13 <pikhq> @metar kcos
03:22:14 <lambdabot> KCOS 050154Z 31008KT 4SM HZ OVC050 24/06 A3032 RMK AO2 SLP172 T02440056
03:22:17 <pikhq> :/
03:22:23 <shachaf> But that doesn't really work that well with the typical uses of Leibniz notation, I think.
03:22:34 <zzo38> ais523: I have read that the notations like d^2x and dx^2 don't follow the rules in the common way, but that it is possible to use the notation to mean something else to make it work, but then the second derivative is written in a more complicated way (although it is still d(dy/dx)/dx)
03:26:16 <zzo38> (I have made the figuring out of d(dy/dx)/dx by myself and confirmed what Penrose mentioned.)
03:26:18 <shachaf> zzo38: I worked it out in this channel once.
03:26:57 <shachaf> d(dy/dx)/dx = d^2y/dx^2 + dy/dx d^2x/dx^2
03:27:55 <zzo38> I worked it out on paper once but forget now what it is
03:28:15 <shachaf> What I said is what you get if you parameterize x and y in terms of t
03:28:34 <shachaf> And then treat "d(...)" as meaning "the derivative of ... with respect to t"
03:31:57 -!- MDude has quit (Ping timeout: 260 seconds).
03:39:01 -!- augur has quit (Remote host closed the connection).
04:04:16 -!- relrod has quit (Read error: Connection reset by peer).
04:04:51 -!- relrod has joined.
04:04:53 -!- relrod has quit (Changing host).
04:04:53 -!- relrod has joined.
04:05:22 <shachaf> relrod: helrod
04:14:38 <relrod> shachaf: hello
04:14:45 <relrod> my server just shat itself and I don't know why
04:15:10 <relrod> load average went through the roof, but it wasn't a DDoS or anything (no spike in net traffic, nothing useful in logs) :/
04:16:21 <ais523> relrod: it's likely either a scheduled maintenance task that rarely runs, or a trojan/virus
04:16:38 <ais523> you want something like top in order to see what's responsible for a high load average, not logs
04:16:52 <ais523> or it could be a legitimate program that got stuck in an infinite loop for some reason
04:18:33 <relrod> ais523: It's a Linux box, so I'd assume not trojan/virus. And I run the box... I haven't installed any new cronjobs recently or anything.
04:18:59 <ais523> relrod: the most common malware on Linux is from people brute-forcing your password, then installing it manually when they learn what it is
04:19:00 <relrod> So yeah it's likely a loop of some sort...but I was hoping something would have spit out something useful to a logfile
04:19:19 <ais523> Linux boxes are much more valuable targets to hackers than Windows boxed because Linux is a much more powerful OS
04:19:26 <shachaf> Was there disk I/O? High memory usage?
04:19:28 <ais523> but if it's a loop, the program responsible will show at the top of top
04:19:29 <shachaf> ais523: Huh?
04:19:30 <ais523> at 100% CPU usage
04:19:44 <shachaf> How is Linux more powerful?
04:20:01 <relrod> ais523: I only root login with key auth. So if it were a password attempt, they'd also need to know my username, and the (non-standard) port I run sshd on.
04:20:07 <ais523> shachaf: if you can run as root on Linux, it's fairly easy to do things like make a highly scalable spam cannon
04:20:09 <relrod> in which case it would be a *very* targeted attack
04:20:20 <relrod> only allow root login with key auth*
04:20:20 <ais523> that's harder on Windows, which is further from the metal
04:20:27 <shachaf> How do you mean?
04:20:38 <shachaf> Why do you need metal to make a cannon?
04:21:30 <relrod> shachaf: there was a spike in mem usage, but it didn't peak. https://elrod.me/collectd/bin/index.cgi?plugin=memory&timespan=86400&action=show_selection&ok_button=OK
04:21:45 <ais523> Linux is better for writing servers than home editions of Windows
04:22:05 <relrod> 23:19:28 < ais523> but if it's a loop, the program responsible will show at the top of top
04:22:08 <relrod> if I could've ssh'd into it, yes
04:22:11 <relrod> I had to force a reboot
04:22:14 <ais523> which have arbitrary restrictions that make them very bad at serving as servers
04:22:20 <relrod> I couldn't get into the box at all, on console or ssh
04:22:25 <shachaf> I don't know much about home editions of Windows. But I assume relrod's server isn't running that anyway.
04:23:03 <ais523> Linux has system calls like sendfile(2) which are designed for highly efficient network operations, whereas IIRC Windows can't do more than 4 network connections at once unless you pay Microsoft money to increase the limit
04:23:47 <ais523> relrod: kernel panic, perhaps? that would normally show in the logs, but sometimes it's something bad enough that the logs can't be saved to disk
04:23:54 <shachaf> relrod: you should give up and set up prometheus like fizzie did hth
04:24:00 <ais523> although that can't be the case if the graphs were collected by a usermode process on the machine
04:24:10 <relrod> ais523: well it was responding to pings (so net stack was still up), but that was about it.
04:24:19 -!- FjordPrefect has quit (Quit: WeeChat 1.9).
04:24:56 <relrod> ais523: I'm thinking some process just went crazy
04:25:34 <ais523> that shouldn't make it impossible to ssh in, and definitely shouldn't make it impossible to get in via the serial console
04:26:18 <HackEgo> [wiki] [[Language list]] https://esolangs.org/w/index.php?diff=53034&oldid=53023 * Dlosc * (+15) /* C */ Added Charcoal
04:26:49 <ais523> ooh, is someone finally documenting Charcoal?
04:27:01 <ais523> I figured out bits by reading the source, but it's still a really hard language to learn
04:27:18 <HackEgo> [wiki] [[User:Dlosc]] https://esolangs.org/w/index.php?diff=53035&oldid=50903 * Dlosc * (+151) Added note about Charcoal contributions
04:27:38 <pikhq> How many people actually have Windows 10 S though?
04:27:50 <pikhq> 10 S is the one with that limit, and it *also* has the limit you can only run Store apps.
04:28:02 <ais523> pikhq: oh, I thought it applied to regular Windows * Home too
04:28:10 <ais523> maybe that got removed at some point
04:29:06 <relrod> ais523: I mean, if a process was hammering both cores, everything else would get queued up, waiting for CPU time, right?
04:29:36 <relrod> which would explain SSH hanging, and the console not being able to throw me a login prompt
04:31:02 <pikhq> I think it might have been also in some of the "developing nations only" editions too.
04:31:40 <pikhq> Home omits some features, like Hyper-V, but doesn't have arbitrary limitations on the features it does have like that.
04:31:45 <relrod> I think it depends though. I've had work servers that nagios-alerted with ~400 load average before that still let me ssh in and kill the process that went to shit. *shrug*
04:32:09 <ais523> relrod: schedulers don't work like that, if a process is really hammering the CPUs, the kernel will make sure that other processes can get timeslices
04:32:23 <ais523> now, if a process is really hammering the /disk/, that can give Linux trouble, for reasons I don't fully understand
04:33:49 <ais523> pikhq: I looked it up, there's a limit of 10 or 20 (depending on edition) devices that can be connected to a single Windows machine, but that wording is vague and probably doesn't apply to all TCP connections
04:34:39 <pikhq> Oh, huh...
04:34:46 <pikhq> That's the limit on the *SMB server*.
04:35:10 <ais523> right, looking around various discussions on this
04:35:24 <pikhq> Which, humorously, is trivial to work around now.
04:35:32 <pikhq> Use Samba in WSL.
04:35:40 <ais523> I've gathered that a) that limit is related to SMB; b) inbound connections that don't use SMB or IIS are sometimes banned by license agreements but there's no technological block on them
04:36:15 <ais523> pikhq: OK, that's brilliant
04:36:41 <HackEgo> [wiki] [[User:Dlosc]] https://esolangs.org/w/index.php?diff=53036&oldid=53035 * Dlosc * (+66)
04:36:49 <pikhq> Which is provided on all versions of Windows that will run 64-bit Windows binaries.
04:37:05 <pikhq> (of up-to-date Windows 10, I should say)
04:39:32 -!- augur has joined.
04:40:56 <oerjan> ais523: your polyglot challenge is up to 127 now
04:41:23 <ais523> oerjan: hmm, has it slowed down a bit then?
04:41:37 <ais523> or have I just failed to track the typical rate of growth mentally?
04:41:59 <ais523> any interesting new languages, or is it all just BF derivatives?
04:42:16 <oerjan> it seems to come in bursts, with several people posting shortly after each other, i assume they coordinate.
04:43:26 <oerjan> latest are: deltaplex, nhonhehr, gammaplex, C(clang), mycelium, monkeys, braincopter
04:43:42 <oerjan> *nhohnhehr
04:44:27 <oerjan> piet wasn't too long ago
04:44:52 -!- augur has quit (Ping timeout: 260 seconds).
04:46:42 <HackEgo> [wiki] [[Brian and Chuck]] N https://esolangs.org/w/index.php?oldid=53037 * Dlosc * (+27) Redirected page to [[Brian & Chuck]]
04:47:17 <oerjan> and before that, they had a bunch of emoji languages
04:47:53 <\oren\> Hmm, I wonder how eso kOS is. Maybe it merits an wiki page?
04:48:12 <ais523> I'm reading the latest additions now
04:57:32 -!- sleffy has joined.
05:00:47 -!- augur has joined.
05:02:00 -!- tromp has quit (Remote host closed the connection).
05:05:52 -!- augur has quit (Ping timeout: 260 seconds).
05:17:20 -!- MDude has joined.
05:35:02 -!- jaboja has quit (Ping timeout: 260 seconds).
06:16:40 -!- Antoxyde has joined.
06:18:08 -!- Antoxyde has quit (Client Quit).
06:40:10 <zzo38> I have the book of Pitman shorthand (from 1971). But, can we to make up a new system of shorthand can be theatre shorthand?
06:43:03 <zzo38> The intention of my idea of the theatre shorthand will be (among other things): [1] To write without having to look at the paper so much. [2] In addition to words, can also be record music.
06:52:51 <zzo38> Do you like this?
07:00:20 -!- augur has joined.
07:03:30 -!- FreeFull has quit.
07:29:30 <HackEgo> [wiki] [[User:Dlosc]] https://esolangs.org/w/index.php?diff=53038&oldid=53036 * Dlosc * (+137)
07:30:12 <HackEgo> [wiki] [[User:Dlosc]] https://esolangs.org/w/index.php?diff=53039&oldid=53038 * Dlosc * (+0) Alphabetized
07:46:17 -!- sleffy has quit (Ping timeout: 260 seconds).
07:47:07 -!- augur has quit (Remote host closed the connection).
07:51:18 -!- tromp has joined.
07:52:34 -!- hppavilion[1] has quit (Quit: HRII'FHALMA MNAHN'K'YARNAK NGAH NILGH'RI'BTHNKNYTH).
07:54:43 -!- augur has joined.
07:54:54 -!- sleffy has joined.
07:57:09 -!- Lymia has quit (Quit: Hugs~ <3).
07:59:04 -!- Lymia has joined.
07:59:11 -!- augur has quit (Ping timeout: 248 seconds).
08:02:48 -!- oerjan has quit (Quit: Nite).
08:20:18 -!- tromp has quit (Remote host closed the connection).
08:23:23 -!- augur has joined.
08:27:59 -!- tromp has joined.
08:27:59 -!- augur has quit (Ping timeout: 248 seconds).
08:34:14 -!- augur has joined.
08:38:21 -!- augur has quit (Ping timeout: 240 seconds).
08:41:11 -!- AnotherTest has joined.
08:45:15 -!- jaboja has joined.
09:05:24 -!- sebbu2 has quit (Read error: Connection reset by peer).
09:05:53 -!- sebbu has joined.
09:23:41 -!- sleffy has quit (Ping timeout: 255 seconds).
09:59:45 -!- Sgeo_ has quit (Ping timeout: 248 seconds).
10:11:46 -!- imode has quit (Ping timeout: 264 seconds).
10:18:36 -!- jaboja has quit (Ping timeout: 240 seconds).
10:25:33 -!- Sgeo has joined.
10:28:20 <b_jonas_> "<shachaf> ais523: It's annoying that Hearthstone card text doesn't describe behavior completely at all." => um sure. M:tG text doesn't either. that's why we have a separate Comprehensive Rules and Oracle Text document, each hundreds of pages long.
10:28:43 <b_jonas_> The rules are so big you can't reasonably fit them on the front of the cards.
10:30:48 <shachaf> For almost all cards, the card text and general rules understanding describes the behavior precisely.
10:30:58 <shachaf> For some special cases where you need extra rulings, those are mostly to disambiguate.
10:31:27 <b_jonas_> "<ais523> actually I think one of my favourite innovations is cards referring to themselves in first person" => why is that useful? couldn't it cause confusion because players also refer to themselves in first person when they describe an action. I'm not saying you should use full card names (or the part before comma) all the time, but M:tG also uses "this <property>" in indirect rules text and "this" in reminder text.
10:32:34 <shachaf> In Hearthstone there are cards with effects like "summon a random creature token of type X", where you don't even know what tokens of type X there are and what abilities they have without trying it out.
10:32:42 <shachaf> Or something like that, I don't know the details.
10:33:08 <b_jonas_> "<zzo38> Magic: the Gathering has text editing effects, which implies a AST anyways, even if nobody has ever written it down." => no, I totally disagree with that. The text editing effects are very limited, they only care about mentions of mana symbols and color words and subtypes (land and creature) in the card text, which just requires a few simple parameters of those types rather than a full AST.
10:37:56 <b_jonas_> zzo38: in particular, I'm not really convinced that it would be worth to specify M:tG in some abstract domain specific language tailored to M:tG. It could work, but I think it might be better to use a good generic programming language for implementing the rules, and not have it automatically linked to the English version of the rules.
10:38:30 <b_jonas_> (The English Comp Rules and Oracle versus the computer-readable code version could be maintained together, or separately one following the other as they are now.)
10:41:15 <shachaf> b_jonas_: I'd rather design a new game meant for this sort of system than use MTG
10:42:01 <b_jonas_> "<ais523> the chain rule [in calculus] makes intuitive sense if you see derivatives as limits of deltas" => it's easy to confuse yourself to think that the chain rule makes intuitive sense, but it's much harder to actually get an intuition that helps you figure out answers to more complicated applications, such as those involving multiple derivatives or more some crazier integral-fubini transforms
10:42:13 <b_jonas_> shachaf: same to you
10:42:51 <b_jonas_> also, there's like five different ways how people say the chain rule makes intuitive sense, and I've no idea how to connect those different intuitions.
10:43:35 <b_jonas_> sure, for simple problems, any of the intuitions work.
10:44:17 <shachaf> b_jonas_: All I was saying was, if you add types, there's only one way the chain rule can work.
10:44:42 <shachaf> If you say that for f : A -> B, Df : A -> (A -o B), where -o means a linear map
10:44:48 <shachaf> Then you just fit it all together
10:45:14 <shachaf> I guess some people would say something about the product of the Jacobian matrices or something.
10:45:51 -!- zseri has joined.
10:51:33 <b_jonas_> "<ais523> [home editions of Windows] have arbitrary restrictions that make them very bad at serving as servers" => and no good ports of C or C++ compiler toolkits or perl interpreters. it took some working on windows at my job to realize how REALLY TERRIBLE the C compiler situation is.
10:51:51 <ais523> strawberry perl is pretty good
10:53:04 <b_jonas_> (if Rust ever takes over, it won't be because it's a better language or has a better compiler, even though both of those are true, but because they're doing a lot of extra work to make sure all the infrastructure works on windows. and this might be a factor of the popularity of python and other high-level languages over C++, though I'm not sure about that.)
10:53:55 <b_jonas_> seriously. NOBODY HAS A GOOD FREE SOFTWARE PORT OF ANY GOOD C COMPILER TOOLKIT FOR WINDOWS X86_64. There are tons of ports, but each of them suck.
10:56:41 <b_jonas_> The root problem isn't that your existing C or C++ programs aren't portable, because libraries are missing. That's just a consequence of having no good C compiler toolkits, which is why existing mostly portable C or C++ libraries already don't get ported, and it's hard to develop entirely new C or C++ programs too.
10:57:45 <b_jonas_> Even libraries that get ported to BSD and Hurd and IBM's non-ASCII systems and VMS all sorts of crazy systems don't have a windows port, because it's such a suffering to continuously maintain C code for windows.
10:58:12 <APic> That is good.
10:58:13 <b_jonas_> (Windows does have some other advantages for a user, I know.)
10:58:33 <APic> That makes good Programmers go away from Winblows. 😎
11:02:16 <zseri> yes
11:03:29 <b_jonas_> "<zzo38> The intention of my idea of the theatre shorthand will be (among other things): [1] To write without having to look at the paper so much." => people just use stenography (shorthand typewriters) for that, I think
11:05:04 <b_jonas_> "<shachaf> I'd rather design a new game meant for this sort of system than use MTG" => ok
11:05:51 <b_jonas_> "<shachaf> b_jonas_: All I was saying was, if you add types, there's only one way the chain rule can work." => I don't think so. when you do third derivatives, you get small constant factors in your expressions that you can't figure out from just the types.
11:06:55 <b_jonas_> APic: good in what sense? it's not good for a programmer like me who has to suffer with stupid windows software because of this
11:07:22 <APic> Just stop working for a Company that forces You to use Windows?
11:07:54 <b_jonas_> APic: technically they only force me to develop products that will work on windows, but it's difficult either way
11:09:34 <APic> Life is generally not easy. ☺
11:11:02 <b_jonas_> yeah
11:11:30 <APic> And:
11:11:34 <APic> All Generalizations are false. 😉
11:25:43 -!- PinealGlandOptic has joined.
11:33:01 <shachaf> b_jonas_: Third derivatives? How do you mean?
11:42:42 <HackEgo> [wiki] [[TEWNLSWAC]] https://esolangs.org/w/index.php?diff=53040&oldid=53031 * Zseri * (+19) +new cast op equivalent
11:51:53 -!- sftp has quit (Excess Flood).
11:52:13 -!- sftp has joined.
12:01:35 -!- ais523 has quit.
12:17:39 -!- erkin has joined.
12:35:22 <HackEgo> [wiki] [[TEWNLSWAC]] https://esolangs.org/w/index.php?diff=53041&oldid=53040 * Zseri * (+250) Operator Compat Levels
12:48:23 -!- augur has joined.
12:53:03 -!- augur has quit (Ping timeout: 252 seconds).
13:03:32 -!- LKoen has joined.
13:11:54 <b_jonas_> shachaf: uh... let me look up the formulas online, I don't remember how they work. might involve some partial derivatives in multiple dimensions on a totally differentiable function where the symmetry of the order of derivatives causes the constants, or maybe something else was involved
13:12:12 <b_jonas_> shachaf: you already mentioned chain rule and second derivatives
13:12:17 -!- b_jonas_ has changed nick to b_jonas.
13:13:26 <b_jonas> shachaf: look at https://en.wikipedia.org/wiki/Chain_rule#Higher_derivatives and the constant factors of terms on the right hand side of equations
13:14:15 <b_jonas> as always, it's possible that you or ais are already good at calculus and all that stuff is still intuitively obvious to you, and it's just me who has the problem
13:21:53 <HackEgo> [wiki] [[TEWNLSWAC]] https://esolangs.org/w/index.php?diff=53042&oldid=53041 * Zseri * (-30) improve formatting of operator tables
13:23:43 <b_jonas> I still couldn't get used to some of this strange new interface in MS Word these days. So find and spellcheck aren't controlled by dialog boxes now, but instead by dockers. The main action buttons in those dialogs (find next, replace, replace all; skip, skip all, ignore, replace with selected word) do have underlined letters in them,
13:24:59 <b_jonas> but it wasn't obvious to figure out how you actually press the button with those shortcuts. The answer is, you have to first F6 to the docker (clicking on a buttom with the mouse doesn't activate that docker), then press the letter without modifiers.
13:25:40 <b_jonas> Alt+letter or alt+shift letter doesn't work even if that docker is selected, because it's not a dialog. Alt+letter will just activate the menu entry, which sort of makes sense, but alt+shift+letter doesn't seem to do anything,
13:26:25 <b_jonas> which goes against what I learned in all word, in which macro debugging was controlled by such a persistently active non-modal "docker" (actually an extra toolbar thingy) with underlined buttons and you had to press alt+shift+letter for its buttons (because alt+letter clashed with the menu).
13:28:03 <b_jonas> And I still think there's no way to figure out how the shit any of this keyboard contorl works from just the interface and help, unless you already have a lot of fucking experience-based intuition on how older more intuitive MS software works and try pressing a lot of random buttons (and make sure you have gigabytes of free RAM and make backups because trying random buttons has strange side effects).
13:32:15 <b_jonas> And even if you do figure it out, the interface is sometimes unintuitive. Example: you can toggle the ribbon bar as always visible or hidden, have ribbon bars with functions that you commonly use together, only one ribbon bar is shown at any one time, and if a ribbon bar was visible and you activate a new one, the new one will stay visible. Good so far.
13:33:38 <b_jonas> The buttons and other controls on ribon bars have the equivalent of menu shortcuts, these shortcuts are one or two letters, and they're per ribbon bar. Still good. The shortcuts within a ribbon bar are distinct form the shortcuts of different ribbon bars (top-level menu entries), which implies you get a lot of really strange and too long shortcuts, BUT
13:34:35 <b_jonas> you can't just press alt plus the shortcut of a ribbon bar button, despite that it seems like you can, no, each time you have to press the shortcut of the ribbon bar (with alt or f10) and then the shortcut of the particular control, even if the ribbon bar is already fucking active and visible.
13:35:26 <b_jonas> This is just one example of lots of basically good innovated design messed up by stupid small things like that.
13:36:41 <b_jonas> (I also hate the Hungarian localized interface with a passion no matter when I use it, continuiously ever since those eixst. Just stop localizing any of the fucking interface to Hungarian.)
13:38:46 <b_jonas> (But luckily these days it's easier to get software with the original English localized interface than it used to when you'd have Hungarian Winword on two floppies and that was it. I even looked up the magical incantations for telling Firefox to use the English interface. Its default is to use the default localization of Windows, and I did download the English language pack for Firefox, but even then it wouldn't automatically choose that.)
13:44:05 <b_jonas> (sorry for all the rant)
13:56:58 -!- jaboja has joined.
14:07:45 <b_jonas> Oh by the way, in case this is new to other people. Windows 10 does have a multiple virtual desktop thing built in, wherein you can group windows to multiple "virtual desktop" and quickly switch between desktop to show those windows and hide most others. In windows 7 you need extra software for that, although that might still be useful for windows 10 because of a better interface.
14:08:03 <b_jonas> I hadn't known this before, because the feature is somewhat well hidden in the windows 10 interface.
14:16:02 -!- `^_^v has joined.
14:16:58 <HackEgo> [wiki] [[TEWNLSWAC]] https://esolangs.org/w/index.php?diff=53043&oldid=53042 * Zseri * (+251) Computational class
14:32:21 -!- catern has quit (Ping timeout: 240 seconds).
14:33:59 -!- catern has joined.
14:34:16 -!- jaboja has quit (Remote host closed the connection).
14:42:37 -!- ATMunn has joined.
14:44:50 -!- Antoxyde has joined.
14:56:24 -!- 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.”).
15:04:15 -!- augur has joined.
15:07:05 -!- picoder_ has joined.
15:08:13 <picoder_> Hi, has anyone tried to optimize the graham scan by modifying the actual algorith?
15:09:02 -!- augur has quit (Ping timeout: 260 seconds).
15:22:09 -!- picoder_ has left.
15:30:33 -!- picoder_ has joined.
15:45:50 -!- fractal has changed nick to _fractal_.
15:59:04 -!- picoder_ has left.
16:08:05 <HackEgo> [wiki] [[Noid]] https://esolangs.org/w/index.php?diff=53044&oldid=52875 * Zayne * (+38) /* Examples */
16:13:28 <HackEgo> [wiki] [[Noid]] https://esolangs.org/w/index.php?diff=53045&oldid=53044 * Zayne * (+123)
16:41:19 <int-e> this session will be full... (will include ICFP contest ceremony...)
16:43:05 <int-e> @tell oerjan please define "broken"
16:43:05 <lambdabot> Consider it noted.
16:48:49 <int-e> @tell oerjan in general I find it very hard to not simply ignore reports that tell me that something is broken without explaining how.
16:48:49 <lambdabot> Consider it noted.
16:49:15 <b_jonas> int-e: nice! will Endo be present?
16:49:20 * APic grins magically.
16:49:39 <int-e> b_jonas: I don't know.
16:50:05 <int-e> there's a couple of talks first
16:51:07 -!- AnotherTest has quit (Ping timeout: 260 seconds).
16:52:05 <\oren\> b_jonas: hmm, now I'm wondering how different countries compare, if you asked people there "do you prefer to use most software in English or in your native language?"
16:52:44 <int-e> b_jonas: did Endo's planet even have trees and rivers?
16:55:29 -!- FreeFull has joined.
16:55:49 <\oren\> i guess you'd have to somehow correct for differences in English knowledge between countries
16:59:20 <\oren\> but there might be a crossover point where a country's population knows enough english that they will prefer the English interface to a badly localized version
16:59:40 <b_jonas> int-e: dunno, but I think there his transformed form has definitely affected a tree and river.
17:00:05 <b_jonas> (recall that endo's DNS affects its environment more directly than ours)
17:04:09 <shachaf> b_jonas: I see.
17:32:20 <zzo38> b_jonas: My idea specifies the cards (and some of, but not all of, the rules) in a domain specific language, although it does not have to be one specific to M:tG, and might be usable with other kind of card games too. But it should be one close enough to be able to do alterable AST (whether it is an interpreter and alters it directly, or is compiled and figures out what it needs to compile in order to allow the alterations to be implemented to do t
17:33:12 <zzo38> Does this make sense? Or maybe I missed something
17:43:33 -!- Sgeo_ has joined.
17:45:51 -!- Sgeo has quit (Ping timeout: 248 seconds).
18:21:24 <shachaf> `olist 1096
18:21:25 <HackEgo> olist 1096: shachaf oerjan Sgeo FireFly boily nortti b_jonas
18:41:29 -!- mroman has joined.
18:41:35 <mroman> https://en.wikipedia.org/wiki/Biproportional_apportionment <- how the fuck can this work
18:41:38 <mroman> if I have 6 seats
18:41:51 <mroman> hm. no
18:41:59 <mroman> if I have 2 seats to give
18:42:05 <mroman> and I have the votes 200, 200, 200
18:42:18 <mroman> how can there be a divisor such that the rounded sum yields 2?
18:42:25 <mroman> either 200 / divisor is 0 or it is 0
18:42:31 <mroman> meaning the sum of seats will either be 3 or 0
18:42:34 <mroman> but it can never be 2?
18:48:54 -!- imode has joined.
18:52:26 -!- doesthiswork has joined.
18:52:45 <doesthiswork> I see we're a shrub now
19:09:25 -!- PinealGlandOptic has quit (Quit: leaving).
19:20:42 -!- sleffy has joined.
19:23:22 -!- impomatic has quit (Ping timeout: 260 seconds).
19:28:04 <HackEgo> [wiki] [[Special:Log/newusers]] create * Raumaankidwai * New user account
19:29:26 <HackEgo> [wiki] [[Esolang:Introduce yourself]] https://esolangs.org/w/index.php?diff=53046&oldid=53022 * Raumaankidwai * (+275)
19:45:41 -!- AnotherTest has joined.
19:51:51 -!- jaboja has joined.
19:57:54 -!- doesthiswork has quit (Quit: Page closed).
20:06:41 -!- sleffy has quit (Ping timeout: 248 seconds).
20:08:39 -!- mroman has quit (Ping timeout: 260 seconds).
20:12:47 -!- xkapastel has joined.
20:14:38 -!- sebbu2 has joined.
20:18:00 -!- sebbu has quit (Ping timeout: 252 seconds).
20:18:47 <HackEgo> [wiki] [[Cubestate]] N https://esolangs.org/w/index.php?oldid=53047 * Raumaankidwai * (+5268) Created page with "'''Cubestate''' is an esoteric programming language made by [[User:Raumaankidwai]]. '''Cubestate''' (or '''cubestate''') programs are based on sequences of moves on Rubik's Cu..."
20:19:20 <HackEgo> [wiki] [[Cubestate]] https://esolangs.org/w/index.php?diff=53048&oldid=53047 * Raumaankidwai * (+6)
20:19:37 -!- sleffy has joined.
20:20:46 <HackEgo> [wiki] [[Cubestate]] https://esolangs.org/w/index.php?diff=53049&oldid=53048 * Raumaankidwai * (+202)
20:21:31 -!- FreeFull has quit (Quit: New kernel).
20:44:43 -!- FreeFull has joined.
20:49:34 -!- Phantom_Hoover has joined.
20:49:40 -!- Phantom_Hoover has quit (Changing host).
20:49:40 -!- Phantom_Hoover has joined.
20:55:06 <HackEgo> [wiki] [[Cubestate]] https://esolangs.org/w/index.php?diff=53050&oldid=53049 * Zseri * (-254) improve formatting
20:55:43 -!- ais523 has joined.
20:59:31 -!- Antoxyde has quit (Quit: Leaving).
21:34:35 -!- idris-bot has quit (Ping timeout: 240 seconds).
21:38:06 -!- jaboja has quit (Ping timeout: 240 seconds).
21:58:39 -!- jaboja has joined.
22:06:10 -!- Mayoi has joined.
22:09:21 -!- erkin has quit (Ping timeout: 248 seconds).
22:20:28 <zseri> bye
22:20:31 -!- zseri has quit (Quit: Leaving).
22:31:49 <shachaf> zzo38: Do you like pumpernickel bread?
22:36:46 -!- Guest53736 has changed nick to moonythedwarf.
22:36:57 -!- moonythedwarf has quit (Changing host).
22:36:57 -!- moonythedwarf has joined.
22:42:45 -!- jaboja has quit (Remote host closed the connection).
22:47:35 -!- AnotherTest has quit (Ping timeout: 240 seconds).
22:48:38 -!- AnotherTest has joined.
22:51:33 -!- augur has joined.
22:52:11 -!- BooK has joined.
22:53:05 -!- AnotherTest has quit (Ping timeout: 240 seconds).
23:03:38 -!- `^_^v has quit (Quit: This computer has gone to sleep).
23:04:37 -!- `^_^v has joined.
23:06:04 -!- LKoen has joined.
23:06:04 -!- LKoen has quit (Client Quit).
23:06:07 -!- `^_^v has quit (Client Quit).
23:32:21 -!- xkapastel has quit (Quit: Connection closed for inactivity).
23:34:12 -!- MrBismuth has quit (Read error: Connection reset by peer).
23:51:27 -!- tromp has quit (Remote host closed the connection).
23:51:28 -!- oerjan has joined.
23:52:53 <oerjan> @wii int-e
23:52:53 <lambdabot> https://wiki.haskell.org/int-e_
23:52:57 <oerjan> argh
23:53:19 <oerjan> @messages-loud
23:53:19 <lambdabot> int-e said 7h 10m 13s ago: please define "broken"
23:53:19 <lambdabot> int-e said 7h 4m 29s ago: in general I find it very hard to not simply ignore reports that tell me that something is broken without explaining how.
23:53:29 <oerjan> @hoogle a -> a
23:53:29 <lambdabot> Prelude id :: a -> a
23:53:29 <lambdabot> Data.Function id :: a -> a
23:53:29 <lambdabot> GHC.Exts breakpoint :: a -> a
23:55:18 <oerjan> @tell int-e Broken as in even @hoogle a -> a wasn't working yesterday.
23:55:18 <lambdabot> Consider it noted.
23:55:37 -!- wob_jonas has joined.
23:56:04 <wob_jonas> fuck! is there some simple way to remove a rust stain from a cheap polypropilene box with rough surface?
23:56:19 <oerjan> dynamite hth
23:56:39 <wob_jonas> I stored a throwaway standalone scraper blade in it, and I didn't realize the blade was made from a non-rustproof iron
23:56:39 <oerjan> may also remove other things.
23:57:28 <oerjan> i'm sorry, my chemistry is rusty
23:58:31 -!- MrBusiness has joined.
←2017-09-04 2017-09-05 2017-09-06→ ↑2017 ↑all