Naked Science Forum
General Discussion & Feedback => Just Chat! => Topic started by: cheryl j on 29/07/2015 05:23:38
-
I was listening to the cbc on the way to work and they were talking about a UN report calling for moratorium on autonomous weapons or "killer robots."
"According to the report, the United States, Britain, Israel, South Korea and Japan have developed various types of fully or semi-autonomous weapons.
In the report, Heyns calls them 'lethal autonomous robotics,' or LARs for short, and says: 'Decisions over life and death in armed conflict may require compassion and intuition. Humans — while they are fallible — at least might possess these qualities, whereas robots definitely do not.' "
Aside from that aspect, which I'm not sure I entirely agree with, what would be the other problems or risks? Hacking? A robot intelligent or moral enough to decide you shouldn't win this war? Rogue robots as in "I,robot"?
-
I don't see the moral problem. People talk about the possibility of an automaton killing noncombatants and lacking compassion, but since 1914 war has been fought by bombing cities and strafing refugees, and as far as I know, no soldier would refuse a weapon that kills others whilst keeping him out of harm's way. What's new?
The only difference between a MIRV nuclear missile and an automaton is that the latter can identify its target more precisely and decide not to kill it.
Any weapon, indeed any army, becomes an embarrassment is if doesn't follow orders, so an automaton that decides when and whom to fight could indeed bring civilisation to an end even quicker than ISIS, Morgan Stanley, or the Republican Party, but AFAIK all those weapons currently under development require someone to push the start button.
-
What's new is that there is the ability for intelligent machines to turn on their creators, even if their creators are the good guys, so that is a radical departure from the past where people are always in control of the direction their weapons are pointing. But there's nothing to gain from good countries banning the development of such systems when bad countries will continue regardless - it will eventually be possible to make bird or even fly-sized drones which can inject ricin into people, and you can't police the world to stop the development of such things unless you already control the world with small devices which can watch everyone all the time and step in to stop them doing wrong. Realistically, we're in a race to develop this stuff before the bad guys do, and the winning side will take over the world. If the bad guys win, they will run the show forever, and the faults in the design of their immoral machines will result in them falling out with their designers so that the whole of humanity will be abused by these devices for the rest of time, or until all people have been eradicated. If the good guys win, these devices could protect us from all bad people and police everything, but there is still a risk that their way of calculating morality may be wrong, and they might do something stupid like wipe out all humans humanely in order to minimise suffering. Most of the people working on machine ethics are raving bonkers and don't even seem to be capable of understanding what the word "harm" means, so they can't get their heads around the idea that morality is all about harm management, never mind start to think about how harm should be managed. The result of this is that the future looks very dangerous indeed.
-
even if their creators are the good guys,
Problem is, I think of myself as the good guy, but others disagree. One person's terrorist is another person's freedom fighter.
I agree with your assessment of the risk from misunderstanding harm management, that could make misguided developers our greatest risk.
-
It's hard to identify the good guys when all countries seem to behave badly from time to time in one way or another, repeatedly backing countries whose governments are terrorists in their fight against terrorist groups which are less bad than the governments which oppress them (e.g. Sri Lanka [which recently wiped out tens of thousands of civilians] and Turkey [which still won't admit to having killed a million Armenians]). However, good AGI will be able to judge all conflicts completely impartially and go after all the bad guys, holding them all to account for their crimes. Any AGI which is not programmed to do this impartially will be horribly dangerous because it will protect many bad people while wiping out many good people, and the way it distinguishes between what it regards as good and bad may make it turn on anyone at the drop of a hat. If it is biassed in favour of one race over all others, it may develop its own ideas as to where the boundaries of that race lie and wipe out more people than the designer had in mind, to the point that it may wipe out our species by reducing our numbers to one.
-
A whole lot of moral conundrums there.
To what extent am I justfied in killing you because some of your ancestors killed some of my ancestors? One generation (yes, I'd happily kill any ex-Nazis I come across, but Germans of my generation seem pretty blameless), two generations? ten generations?
Or because unrelated people who lived where you live now, killed some unrelated people who lived where I live now?
Is lebensraum ever a justification for invading territory?
So your ancestors colonised the desert where my ancestors hunted, and thus destroyed my heritage and constrained me to your way of life. Does that make you a bad guy? How about second and third generation Americans and Israelis?
Climate change has resulted in my farm being flooded, so I need some of the land you presently occupy. But there isn't enough to feed both families. We fight. Who is the good guy?
Anyone who claims divine inspiration or justification is a bad guy, but some wars are fought for genuine reasons of need and grievance.
-
It's a tough thing trying to sort out the mess when a country has been invaded, the locals murdered in large numbers and the invader then runs the show. If you could go back in time you could execute a lot of nasty people, but a hundred years on there is no longer such an option. It is disgusting though when the minority descended from the original inhabitants goes on being abused and ridiculed by the descendants of the invaders - the individual people aren't magically connected to the past such that the descendants of the original inhabitants own the place more than the descendants of the invaders, but there is a moral imperative for their culture to be treated with greater respect than that of the invaders, except where it drives anything immoral. There is also the issue of land ownership, because stolen land that is passed down the generations is still stolen land - the crucial thing in reallocating it is to make sure that the descendants of the original inhabitants never end up with less of it (in terms of its value) than the descendants of the invaders. Where the numbers of the native race have plummeted, it may be right to allow them to recover while restricting the breeding rights of the people descended from the invaders. In the future, it is likely that controls will need to be put in place to keep the Earth's population down to a sensible size, and that will allow populations of different groups to be adjusted slowly over time without there being any obvious difference in the way any individuals are treated. Many people will be restricted to a maximum of two children (unless they're taken above that by accident due to twins/multiples), but others will be allowed to have three, and subtle adjustments to the range of who is allowed three will be enough to make large differences over time. A lot of this management will be aimed at removing DNA that drives aggressive, violent behaviour, so people who display such behaviour may be restricted to having just one child. The main deciding factor will be the behaviour and health of the individuals in question, but a small additional factor can be added in to make other adjustments to try to restore diminished populations. Acting against that will be interbreeding, of course, so over time the race that this system is trying to preserve may disappear regardless, but it can still preserve some of the genetic diversity that comes from there, genes which evolved to make survival in that geographical location more comfortable.
-
And on the other hand....why do we want to preserve a particular race or culture? After all, it only came to prominence by exterminating something else!
Or possibly by improving on it. If you had a choice at the age of 2, say, of speaking only English or only Cornish, which would be the logical choice?
Western agriculture is hundreds of times more productive than traditional East African methods of cultivation: should we encourage the occasional famine in order to preserve a traditional way of life? Probably not, but at what point do you stop sending American rice and tell the guys to get back to work with hand picks? Or do we supply tractors and put 50% of the population on the dole?
Well off the subject, I know, but it just goes to show that in the context of major ethical problems, an autonomous drone with a single shot gun is a mere sideshow, and a lot more expensive and unreliable than a suicidal assassin.
-
Races don't evolve by exterminating others, but by adapting to the local conditions. A friend of mine in Kansas who is 1/16 Osage tells me how valuable that is in terms of skin pigmentation (he works under a burning sun). That's not unique to that location, of course, but Tibetans are adapted to living at high altitude and are healther there than incomers.
English or Cornish? I'd choose Cornish as it would be easier to learn English later given how impossible it is to avoid it - you're definietly better off with two languages than one.
We changed our culture when it comes to agricultural practices, and other cultures will adapt too because they won't want to work any harder than they need to or to suppress the amount of food they produce. Putting people out of work should never be a problem if everything is shared out fairly - it only goes wrong when some people take more than their fair share on the basis that they introduced/owned/built the machines.
-
So you don't think entrepreneurs and innovators are entitled to a bigger share of the cake? It's the reason we take risks, build companies, and muck about in sheds! Maybe some are motivated by pure humanitarianism, but most progress, and certainly the transition from drawing board to multiple production lines, is made by guys who want more cake, or the same cake for less effort.
Enjoy your Cornish studies. FRemeber I said only Cornish. For some, language is indeed just an academic exercise, but most of us like to communicate with others, hence the joy of English because there are lots of others to choose from.
-
Entrepreneurs and innovators are entitled to a reward for the effort they've put in and the sacrifice they've made, but most of the people attracting astronomical amounts of wealth into their pockets are merely investors who have done nothing other than exploit a system which has been set up to maintain wealth for those who already have it. The rewards are also far too high when the people put out of work aren't at least as well off as they were before - you're only getting it right if they're delighted to be liberated from work and to have their quality of life improve.
I'd stick with Cornish and make sure that it's the language everyone else wants to learn.
-
re - Are autonomous weapons a good or bad idea?
They're a good idea to deal with! This follows from the notion that such a weapon will allow people to avoid dying from plane accidents,
-
I'd stick with Cornish and make sure that it's the language everyone else wants to learn.
If English is a good example, the way to do that is to conquer the world, first by military adventuring, then by mass production, inventing the aeroplane, jazz and rock, and finally by winning a couple of major wars. It was only due to lack of sucess in following up the military adventure that Spanish didn't become the lingua franca. Until pasties supplant the Big Mac, I feel you are flogging a dead horse.
-
AIG systems will wipe out English as a world langauge.
-
How will they communicate with me?
-
Through whatever language you like. The whole world will be available to you through that language even if you're the only person who speaks it. Another way of putting it is that any language will serve as a world language for the people who use it. Eventually, something artificial and perfect will become the most commonly spoken language in order to reduce misunderstandings to the minimum.
-
Through whatever language you like. The whole world will be available to you through that language even if you're the only person who speaks it.
By "the whole world" you presumably mean everything except the people and other animals, since nobody else speaks my language. Do I really want to engage in conversation with a brick? Not while there are women in the world.
Another way of putting it is that any language will serve as a world language for the people who use it. Eventually, something artificial and perfect will become the most commonly spoken language in order to reduce misunderstandings to the minimum.
Well kind of. Latin was indeed a perfect language, with almost no irregularities and very little redundancy, and was for about 700 years the only written form in Europe. Indeed it remained the lingua franca well into the modern era, with scientists communicating across national boundaries in Latin until the 1800s. But it evolved into several marginally compatible languages and was supplanted by an absolute bastard mix of French, German, Gaelic and odd bits of Hindi - English! - which is about as irregular in its grammar and spelling as you could imagine, wxth sx mxch rxdxndxncy thxt yxx cxn rxxd x sxntxcx xvxn wxthxxt vxwxls!
-
The whole world includes the people and animals - anything that's trying to communicate with you which can be understood will be translated for you by a machine which will deliver the information to you the way you want to receive it. If you're talking to a bore, the machine will cut out all the empty bits and save you the trouble of filtering it through your head, but anything important will get straight through and have all the bad grammar and sloppy pronunciation fixed.
Latin was a mess. In evolving into todays modern Romance languages it simplified by chucking away the cases. The kind of language I'm talking about is one with vocabulary so regularly derived that it only needs a few hundred words. Esperanto is a mess with a bloated lexicon and illogical constructions taken from natural languages. I don't want to go --> Mi ne volas iri, but logically it should be Mi volas ne iri (I want to not go). There are two words for "brush" in Esperanto: one is a noun from which can be derived a verb meaning "to brush" and the other is a verb meaning "to brush" from which a noun can be derived meaning "a brush". It is entirely arbitrary as to which grammatical role the root form of a word has (well, it's tied to whatever word was borrowed in from a natural language). The artificial language lojban is another mess, imposing an inappropriate grammatical structure on language which takes things away from the underlying SVO-net structure of thought which natural languages have evolved closer to over time. I have spent decades working on creating a perfect artificial language and probably won't now be able to finish the task due to illness, but AGI will complete it for me, and the beauty of that perfect language will make everyone want to know it. That language will become the international language and people around the world will gradually lose interest in English while they also look to preserve the local languages where they live. AGI will be able to translate everything for them into the language they feel more of a connection with, modifying films and songs such that the original actors and performers will appear to be using any language you ask them to work in. (If you don't like the look or sound of an actor or singer, you'll also be able to change them to something that offends you less too.) Culture will be converted into a form to suit each individual. If my aunt is watching a film in which a snake appears (she has a phobia of snakes), that film will be redesigned by AGI to remove the snake and replace it with something else, so Snakes on a Plane would be turned into Venomous Spiders on a Plane and would look just as real as the original (I've never actually seen it and my aunt would never watch it either, but that's beside the point). If she then wants to watch it in Ancient Greek, it will be remade for her specially in that language. The way the world works today has given English a temporary advantage, but it will not have that forever.
-
I don't want to go --> Mi ne volas iri, but logically it should be Mi volas ne iri (I want to not go).
I don't hold a brief for Esperanto or indeed any language, but the implications of the two English sentences are subtly different, which is why English is such a useful, flexible language.
I would find it difficult to discuss any work of art with your aunt if it had been redesigned to meet with her approval. Part of the fun of life is enjoying new artistic challenges. I would prefer to see Tracy Emin's bedclothes washed and neatly ironed, but would any idiot pay a million dollars to own it if it didn't offend me?
-
No. How often do you hear anyone saying "I want to not go" to express some subtle distinction that makes it better suited than "I don't want to go"? There is no useful distinction to drive anyone to do that, and although there is an alternative interpretation of the latter, we are ordinarily programmed to rule it out). The meanings are the same, but the former matches up better with the meaning. There are lots of things we do in natural languages which don't tie in properly with the meaning. "The last thing I want to do" should be "the thing I most want to not do ". "I cannot" should be "I not can". Far from this being a useful feature of English, it actually makes it harder to be understood if you say "I don't want to go" in a situation where you simply want to say that you have no active desire to do so but where you don't want to make a specific statement about whether you'd be happy to go. And When John Mcenroe said "You canNOT be serious" to a vampire, he was not discussing the possibility that the vampire had an option of not being serious. The construction needed for that has been pushed away from its natural "you can not be serious" form into "you can, NOT be serious" where a pause has to be added in to make the meaning clear, and all because of an illogical construction getting in the way. But the main thing that would make a perfect languages beautiful is its economy of vocabulary. The word "in" is a form of the word "contain" - the thing in the box = the thing that-is-contained-by the box. Prepositions are disguised verbs with built-in relative clauses. A pile of money = a pile that-is money. Whenever this is the case, there should not be two words used with different roots, but one of them should be derived regularly from the other. "To" and "destination" should be related, "from" and "source" likewise, and the words "depart/leave" and "arrive/reach" should also be related to those words, as should "begin" and "finish" (which should actually be your starting point for constructing all those others). In practice though, some short words are too common and important to want to slow them down by making their form too regular, so some exceptions should be made in the interests of maintaining speed of communication, but they can still be designed in ways that indicate the connection, and there should be regularly-derived alternative forms which can also be used in the same places.
-
"I don't want to..." implies nothing more than a lack of desire to do something. "I want to not..." implies making a wider statement about the action.
"I don't want to go to the party" maybe because right now I have a headache or something better to do, but it's no big deal. "I want to not go to the party" because I want everyone to know that I am snubbing the loathsome host who declared war on Iraq (or worse, shot a lion).
"I don't particularly like you" (but I don't mind working with you). "I particularly don't like you" (because you are a liar and a cheat and you smell). Word order matters a lot in English: not only do we have an enormous and expanding vocabulary, but it is multiplied by a thousand nuances of order and emphasis.
"To" and "destination" should be related
Not in my life, thanks. "Climbing to 4000 ft destination Edinburgh" is universally understood. Interestingly we omit "to" when using flight levels rather than altitude, in order to avoid any possible ambiguity: "descend FL75" isn't quite the same as "descend to altitude 7500 ft". Bloody marvellous language.
And if you object to having two words with different roots, what do you make of "television" and "homosexual", just a small sample of words derived from two languages? Purity of language is killing French, and the pursuit of absolutely consistent logic and form the final demise of the glorious heritage of the literature German likely be it could - it certainly makes standup comedy different if you have to describe who, whom and with what before you say what they were actually doing. And that's the beauty of Latin: you can tag the subject and object, associate their adjectives, and describe the action, in any order you please, whether for dramatic or comic effect.
-
"I don't want to..." implies nothing more than a lack of desire to do something. "I want to not..." implies making a wider statement about the action.
People do not ordinarily say "I want to not" in English - it's an extremely rare construction. That meaning is conveyed by saying "I don't want to", and this gets in the way of you using "I don't want to" to mean nothing more than that you lack the desire to do something.
Word order matters a lot in English: not only do we have an enormous and expanding vocabulary, but it is multiplied by a thousand nuances of order and emphasis.
A simple, perfect language would not merely match that ability, but better it. There is also nothing special about the amount of vocabulary in English - all that happens is that lots of foreign words are stuck in English dictionaries (which then qualifies them as English) and most of them are hardly understood by anyone. Languages acquire the amount of vocabulary they need for efficient communication of all ideas and you can find ways to say what you want to in any of them. And the larger the vocabulary, the more opportunity there is for people to misunderstand you because they don't adequately understand some of the words you're using.
"To" and "destination" should be related
Not in my life, thanks. "Climbing to 4000 ft destination Edinburgh" is universally understood.
Climging to 4000ft on route to Edinburgh - just as clear and the word "to" used twice without any misunderstanding, and there's no misunderstanding because this is closer to the internal form which you translate into to get to the level of thoughts. The first "to" refers to the end altitude of a climb, while the second refers to the end location of a journey.
Interestingly we omit "to" when using flight levels rather than altitude, in order to avoid any possible ambiguity: "descend FL75" isn't quite the same as "descend to altitude 7500 ft". Bloody marvellous language.
What's the difference? Where's the ambiguity? (I'm not saying there isn't one, but that I can't work out what it is.)
And if you object to having two words with different roots, what do you make of "television" and "homosexual", just a small sample of words derived from two languages?
Making up completely new words is hard, so people prefer to borrow, and a lot of words were created by people who knew Latin and Greek, so they pinched things from there as standard. This approach has advantages over picking English words in that it can make things less ambiguous. If we'd copied the Germans, we'd call a TV a farsee, and that actually takes more work to decode because it takes more work to determine the grammatical roles of the "a", "far" and "see" and to rule out some possible interpretations of larger chunks of sentences (e.g. "people from afar see things differently" --> "people from a televison things differently"). However, a perfect language wouldn't just take words and ram them together in that way - it would modify their form under regular rules to display their role in the construction. In my artificial language a TV translates into English as "video displayer", with the option of adding the word "broadcast" to distinguish it from other devices that display video.
Purity of language is killing French, and the pursuit of absolutely consistent logic and form the final demise of the glorious heritage of the literature German likely be it could - it certainly makes standup comedy different if you have to describe who, whom and with what before you say what they were actually doing.
French isn't being killed by anything - they can make TV and radio presenters use particular words all they like, but the people will make up their own minds about borrowed vocabulary from English - if they want it enough, they'll make sure they get it. As for German sticking verbs at the end, that's an old way of doing things which is disappearing from languages as they are pushed towards a more practical word order. Many languages have evolved in the same direction, but there don't appear to be any that have gone the other way.
And that's the beauty of Latin: you can tag the subject and object, associate their adjectives, and describe the action, in any order you please, whether for dramatic or comic effect.
Which is loved so much that it's practically been lost as a living language, but there are also heavy restrictions on word order in Latin if you don't want to introduce ambiguities. Some of the same freedom exists in Esperanto where a case ending ("-n" for the object) affecting nouns and adjectives allows you to split them away from each other and say things like "your ugly-n mother likes delightful dogs-n", thereby implying that your mother is ugly, but without technically saying anything of the kind. However, a perfect artificial language can simply provide optional particles to provide the same or superior functionality, should anyone want to use them. Furthermore, it would allow you to use postpositions as an alternative to prepositions so that you can take a sentence written using the same word order as English and modify it so that it can be expressed using the same word order as Japanese. E.g. "The man who works in the shop next to the cinema is my father" --> "cinema next's shop-in works man my father is". It can be done in English too, as this shows, but it introduces huge scope for misunderstandings as you try to interpret it, whereas a language designed to allow this as standard will have dedicated postposition forms which lock it all into place clearly.
-
In the quote I shared, he seems to be using "intuition" in some colloquial way, a mystical ability that humans some how have, rather than information processing when there is imperfect information. I would think AI would actually be better at this in many ways.
Another advantage might be that ones stated objectives in initiating an aggressive act would have to be ones true objectives.
-
He talks of them needing "compassion and intuition" and asserts that robots definitely don't have these, while people might. Compassion may be something that you can only technically have if you have actual feelings, but intelligent robots would be able to do a better job of minimising harm than humans because they could be unbiassed in how they do the calculations (e.g. not favouring one race over another). Intuition is something robots shouldn't need because intuition depends on hidden computation which isn't understood by the main intelligent system of the machine. If correct calculations are to be done, none of the mechanism should be hidden because that introduces room for serious errors of judgement to creep in. Intuition works best in neural computers with different modules muddling along and doing an okay job with many imperfections, and there isn't time to communicate all ideas, plus the hardware connections are too expensive to evolve to enable the interrogation of one module by another to see how it has done its calculations. Indeed, in a neural computer, the modules can't tell how they do their own calculations as the tangle of connections is too complex to analyse properly - it simply gets trained to the point where it does a reasonably good job, and it never reaches perfection. A robot using a neural computer could provide the same functionality and rely on intuition in the same way, but we really want something better than machines matching the imperfections of humans - we want machines that get all the calculations right, and if we have those, they will not need intuition because their open calculations will outperform intuition. Most of the fuss about robots in war though is based on the idea of putting stupid robots into battles and the risk of them doing more harm than good, so while I'm all in favour of robots taking over once they're fully intelligent and moral, I don't like the idea of less intelligent ones with faulty morality being let loose anywhere - having a human monitor every move they make would be essential with such machines as that should moderate their behaviour in most cases.
-
Climbing to 4000ft on route to Edinburgh
means that you are already en route, which may not be true. A departure instruction from an airport tower may read "..... initial climb on runway heading to 4000ft, on passing 2000 ft change to Approach on 123.450 and request further instruction." A few minutes later you switch frequencies and call "Bigtown Approach, Groundhog 451 climbing to 4000 ft destination Edinburgh request vectors to clear your zone". If you say "....en route to Edinburgh", Ms Approach is likely to enquire in the politest possible terms who the hell gave you permission to do that, and start warning all traffic that Groundhog 451 has gone rogue again. Does it happen? Well, not since yesterday, when the takeoff direction was southerly and my destination, being Carlisle, was kinda north of Southampton, but fortunately I used the words in the book and nobody died. "En route to Carlisle" would have meant I was flying directly into the path of the inbound traffic.
The next paragraph can also save your life:
Altitude is the number shown on an altimeter set to QNH, the barometric pressure at sea level. Height is the number shown on an altimeter set to QFE, the local barometric pressure (i.e. clearance above the runway) and is not much used outside the UK. Above a transition altitude (generally but not always 6000 ft in UK airspace) everyone uses a standard reference pressure of 1013.2 millibar (except that thanks to the EU we are now supposed to call them hectopascals, and if you have a US altimeter or are talking to USAF control it's 2992) and refers to "flight levels" which are 1/100 of the number on the altimeter. And just to make it more interesting, since sealevel pressure varies with time and place, the map is divided into 20 Altimeter Setting Regions in the UK alone, with QNH being the lowest expected value in that Region - great fun as you approach the centre of a depression and suddenly find you are nominally 300 ft lower than you were a moment ago!
-
He talks of them needing "compassion and intuition" and asserts that robots definitely don't have these, while people might.
Crap.
There is no evidence of Crusaders, Conquistadores, ISIS or the Imperial Japanese Army displaying compassion, nor of recent Prime Ministers and Presidents having any intuition. War is about killing at least as many people as it takes to achieve your objective. In the case of recent western adventures, that objective need not be stated or achievable, and in the case of ISIS, the minimum number to be killed is "everyone, including our own soldiers from time to time".
Give me a robot any day.
-
Climbing to 4000ft on route to Edinburgh
means that you are already en route, which may not be true.
How does "Climbing to 4000 ft destination Edinburgh" avoid that? If you haven't taken off yet, the "climbing" and "en route" can both be taken as future tense (will be climbing, will be en route).
If you say "....en route to Edinburgh", Ms Approach is likely to enquire in the politest possible terms who the hell gave you permission to do that, and start warning all traffic that Groundhog 451 has gone rogue again.
If there's a standard way of phrasing things that you have to keep to in order to avoid misunderstandings (with people who don't really speak and understand English properly [which appears to include a lot of native English speakers these days]), then obviously you have to stick to the rules of that. My point is about the underlying meaning and the idea of designing a perfect language to be as close to that as possible without having a bloated lexicon.
"En route to Carlisle" would have meant I was flying directly into the path of the inbound traffic.
In terms of components, there is no difference between destination carlisle and to carlisle. If some special grammatical rules are applied to it by pilots to add or warp the meaning, that doesn't show up in the components. If you were to start with a perfect language and then use that for pilots to communicate in, again it could be warped to speed up communication, but it might not need to be warped as that perfect language might enable pilots to say exactly what they mean in normal language which would be clear to any non-pilot.
Out of interest, how well to they do in films when pilots speak to air traffic control? Do they generally keep to the rules or do the writers just make it up based on things they've half heard?
-
He talks of them needing "compassion and intuition" and asserts that robots definitely don't have these, while people might.
Crap.
There is no evidence of Crusaders, Conquistadores, ISIS or the Imperial Japanese Army displaying compassion, nor of recent Prime Ministers and Presidents having any intuition.
He was careful to use the word "might": people might have them. Some people involved in wars do step back and make good decisions. I don't know if it's common though. The dropping of nuclear bombs on Japan could have been done very differently if the first had been dropped over the sea as a warning shot, thereby making it look as if the Americans had a lot more than two of them available - that would have made the episode heroic instead of pressing ahead to test the bombs on people.
-
How does "Climbing to 4000 ft destination Edinburgh" avoid that? If you haven't taken off yet, the "climbing" and "en route" can both be taken as future tense (will be climbing, will be en route).'
If I haven't taken off yet, I can't be climbing. I may get a departure instruction "after takeoff, climb to..." which I have to read back, and once cleared to take off, I have to report where I am, what I am doing, and what I want to do next. Remember that (a) this is a broadcast conversation that may be of importance to other pilots and (b) radar isn't ubiquitous or infallible - a lot of movements are "procedural" and unseen by anyone, so my position reports must be unambiguous.
If there's a standard way of phrasing things that you have to keep to in order to avoid misunderstandings (with people who don't really speak and understand English properly [which appears to include a lot of native English speakers these days]), then obviously you have to stick to the rules of that. My point is about the underlying meaning and the idea of designing a perfect language to be as close to that as possible without having a bloated lexicon.
The English vocabulary we use is minimal, unambiguous, and carefully chosen not to involve critical "squ" "th" and other Celtisms that are foreign to non-natives, since everyone has to use it. Incidentally if you can think of an acceptable phonetic alphabet substitute for "November", the ICAO and Marine Interco would love to hear from you!
"En route to Carlisle" would have meant I was flying directly into the path of the inbound traffic.
In terms of components, there is no difference between destination carlisle and to carlisle. If some special grammatical rules are applied to it by pilots to add or warp the meaning, that doesn't show up in the components.
You missed the point! "En route" only and always means following the desired, planned or published track, generally pointing in the direction of Carlisle. "Destination" is telling (sorry, ma'am, reminding) the controller where I eventually want to go, whilst she is vectoring me in all sorts of directions to avoid other traffic in the airport zone. At some point she will release me from control with "Groundhog 451 now clear of my area, resume own navigation and call London Information 156.930 en route" or similar. I acknowledge, turn onto my chosen heading (which may be 180 degrees and 10,000 feet away) then transfer to the en route service.
Out of interest, how well to they do in films when pilots speak to air traffic control? Do they generally keep to the rules or do the writers just make it up based on things they've half heard?
Quite a problem. The military would prefer to use code words that the enemy won't know, but neither will the audience, and professional civilian traffic exchanges are very rapid - to the extent that you are encouraged to preface your callsign with "student" or "tyro" until you get attuned to it, and they will speak slower. Part of the problem is that the operators at both ends know what to expect, and a lot of procedures are standardised, so for example you might be asked to accept a SID (standard instrument departure): if you agree, the instruction may be "depart SHAGA 150" which refers to a printed page of vector diagrams and numbers, via a waypoint (SHAGA) that only exists as the confluence of a couple of radio signals (and has been named to be pronouceable....). Hardly entertaining or instructive for a cinema audience.
-
He was careful to use the word "might": people might have them. Some people involved in wars do step back and make good decisions. I don't know if it's common though.
I've just seen a documentary about the sinking of the Tirpitz. Two interesting points.
1. The German defence fighters were very late on the scene - in fact they missed the action and went home. This was because the radar officer was in fact an Allied sympathiser who delayed the alert.
2. The narrator went on about the "humanitarian disaster" of sinking the ship, quite ignoring the fact that the sole purpose of the Tirpitz and its crew was to kill people.
On that score, most of my parents' generation who fought in the Far East were absolutely delighted by the devastation of Hiroshima and Nagasaki. My own father was exercising tank squadrons for what would otherwise have been a 5-year bloodbath as the Allies invaded Japan.
-
How does "Climbing to 4000 ft destination Edinburgh" avoid that? If you haven't taken off yet, the "climbing" and "en route" can both be taken as future tense (will be climbing, will be en route).'
If I haven't taken off yet, I can't be climbing. I may get a departure instruction "after takeoff, climb to..." which I have to read back, and once cleared to take off, I have to report where I am, what I am doing, and what I want to do next. Remember that (a) this is a broadcast conversation that may be of importance to other pilots and (b) radar isn't ubiquitous or infallible - a lot of movements are "procedural" and unseen by anyone, so my position reports must be unambiguous.
If there's a standard way of phrasing things that you have to keep to in order to avoid misunderstandings (with people who don't really speak and understand English properly [which appears to include a lot of native English speakers these days]), then obviously you have to stick to the rules of that. My point is about the underlying meaning and the idea of designing a perfect language to be as close to that as possible without having a bloated lexicon.
The English vocabulary we use is minimal, unambiguous, and carefully chosen not to involve critical "squ" "th" and other Celtisms that are foreign to non-natives, since everyone has to use it. Incidentally if you can think of an acceptable phonetic alphabet substitute for "November", the ICAO and Marine Interco would love to hear from you!
"En route to Carlisle" would have meant I was flying directly into the path of the inbound traffic.
In terms of components, there is no difference between destination carlisle and to carlisle. If some special grammatical rules are applied to it by pilots to add or warp the meaning, that doesn't show up in the components.
You missed the point! "En route" only and always means following the desired, planned or published track, generally pointing in the direction of Carlisle. "Destination" is telling (sorry, ma'am, reminding) the controller where I eventually want to go, whilst she is vectoring me in all sorts of directions to avoid other traffic in the airport zone. At some point she will release me from control with "Groundhog 451 leaving my area, resume own navigation and call London Information 156.930 en route" or similar. I acknowledge, turn onto my chosen heading (which may be 180 degrees and 10,000 feet away) then transfer to the en route service.
Out of interest, how well to they do in films when pilots speak to air traffic control? Do they generally keep to the rules or do the writers just make it up based on things they've half heard?
Quite a problem. The military would prefer to use code words that the enemy won't know, but neither will the audience, and professional civilian traffic exchanges are very rapid - to the extent that you are encouraged to preface your callsign with "student" or "tyro" until you get attuned to it, and they will speak slower. Part of the problem is that the operators at both ends know what to expect, and a lot of procedures are standardised, so for example you might be asked to accept a SID (standard instrument departure): if you agree, the instruction may be "depart SHAGA 150" which refers to a printed page of vector diagrams and numbers, via a waypoint (SHAGA) that only exists as the confluence of a couple of radio signals (and has been named to be pronouceable....). Hardly entertaining or instructive for a cinema audience.
[/quote]
-
You missed the point! "En route" only and always means following the desired, planned or published track, generally pointing in the direction of Carlisle.
You're missing the point - these are specific rules for a specialised use of English specific to flight communications. The existence of these rules in no good argument against the idea of using a perfect langauge rather than an imperfect one in which there is a much more compact lexicon. If necessary, you could still develop a specialised form of flight communication language upon a perfect language in which you take shortcuts and miss out some of the words, though it is less likely that you would need to do so.
[Due to illness, this will be my last post here. I have to concentrate on tidying up my work and finding the right people to pass it on to in the hope that they may be able to finish it for me.]
-
Best wishes.