Of course that doesn't mean we're handling them as well as others may have in the past. I don't subscribe to the notion that we're especially better or worse though. It is just an eminence illusion, today's problems always seem worse, more urgent, more significant, than those we read about in history books (sorry, on Wikipedia). lol.
I think the assumption that there's anything impossible about doing it is what is wrong. I also think we are definitely reaching the levels of processing power necessary to meaningfully attack these sorts of problems. Nothing will every do it exactly like a human, but the path to doing it better than the best human is clearly open ahead of us. I can see a world where you can have EVERYTHING parsed for you into an ideal form for you particularly to understand it best. Is the govt going to get a super-human system out of this? No, probably not, it will be a few more generations. The work is interesting though and will lead to even better stuff. It may well be usable as-is too.
Are you maybe confusing metaphor with idiom? Idiomatic speech could truthfully be impossible to interpret without knowledge. A metaphor OTOH is a type of analogy. It operates on the basis of similarity or likeness with something else. It is thus amenable to logical analysis. Take your example, would someone who has not heard this example before understand it? I think it is self evident the answer is yes. It is therefor proven that this metaphor is open to computational analysis.
As language changes some metaphors could become incomprehensible or nonsensical, sure. This is just a problem for all linguistic analysis at some level. Again, the question is are you confusing metaphor with idiom. Most idiom arises somehow out of logical association, but it can be so idiosyncratic that it isn't really amenable to much analysis. Still, I think analysis of these kinds of speech constructs is possible.
So, got a schedule for the big world takeover? Really, I haven't seen even the faintest sign of fact-based ANYTHING in this country since 1980. Things are starting to go seriously wrong...
Eh, I have yet to meet a firearm that gave me any more than the most trivial problems. Certainly left-handed firearms are nice, but in a practical sense both sorts work fine. Working the safety on a pistol quickly is probably the least convenient feature of wrong-handed hand guns.
It isn't a big deal. Most everyday items don't favor one hand or the other. Spears, knives, and even swords pretty much fall into this category, as do quite a lot of other tools. Scissors and similar cutters are sort of an exception. So chances are this wasn't a really large problem. In a pinch you can just use things right-handed (and most lefties are less hand-polarized than righties, maybe do to the convenience of being able to use either hand, for instance I'm perfectly happy doing many tasks with either hand like cutting with scissors).
As for greater cooperation if you have two warriors and one is left handed he can guard the more exposed right flank of his partner and vice-versa. Alexander the Great had a whole cadre of left handed warriors who would take up the far right hand flank of his army (and I'd suspect this was probably fairly standard for ancient armies). The question then is really is there a strong enough difference in cooperation vs competition advantage from different-handedness for it to matter. This study doesn't seem to have addressed that question. At best it can be used like a prior to say "well, maybe it is that way since we DO have 10% lefties." I'm always a little suspicious of that sort of "Bayesian" kind of interpretation since it also leads to such absurdities as the notion that this generation will be the last one in history (do some reading on Bayesian Analysis, lol).
"Cooperation favors same-handedness—for sharing the same tools, for example. Physical competition, on the other hand, favors the unusual. In a fight, a left-hander would have the advantage in a right-handed world."
This is simply taken axiomatically as a starting point for the study. I see no indication that it was determined by any sort of analysis. I'm not even sure such an analysis is feasible. You'd have to know what activities people carried out in prehistoric times, how, and what the value of cooperation was for each one. There could be various advantages and disadvantages of same or opposite handedness depending on the activity, etc. The entire concept of their study rests ENTIRELY on the validity of this same-handedness is better for cooperation proposition. I'm not saying it is untrue, but without demonstrating it to be true and to what degree I cannot see how any meaningful conclusions can be drawn.
are like the wolves amongst you right-handed sheep, right!!!!???
I'm not ENTIRELY convinced, what about situations where it is advantageous for people to have opposite handedness for optimal cooperation? There seems to be a built-in assumption here that different-handed assortments of people will always have more problems working together. I'm not sure there's a practical way to test this as a general thing though.
You don't give your support to something that is broken, you oppose it and negotiate the bad parts out. How exactly by supporting this thing do they think they're going to get it changed? This is Bullshit. If I was Mr Zuckerberg I'd be careful what I wish for...
Hmmm, that's similar to setting yourself on fire to get revenge on someone, hehehe.;)
Not that I'm unsympathetic to the AC, I have no children either, and no such plans. I may get some pleasure out of that I guess, as I starve with the rest of you locusts!
Heh, well, we were using ADA and/or PACE PLI which did the same thing. We still had unit testing to the Nth degree. Heck, we had complete design verification by simulation at the module, assembly, and system levels on top of that, lol.
Of course no 747/57/67/A320/30/40 etc with any of my code flying in it has to my knowledge yet fallen out of the sky due to bad code, lol. I really have to wonder about all these rocket guys and their bad code. Can't even imagine how you could use tools that could allow a type mismatch in that kind of system, crazy. Heck, we coded (and built the hardware for) the range safety destruct box on Titan 34D. Now talk about "this is not allowed to ever have a failure mode, period" yeah. Ain't no stinking C++ code on that baby, nope. You push the big red button, there's only one answer, kaboom!;)
First and Last Men, and Starmaker. Olaf Stapleton is interesting.
And then there is the weird stuff, I suggest William Hope Hodgeson. Boats of the Glenn Carrig is pretty strange. Nightland is a work of genius, but also pretty much unreadable. It is probably worth reading the early Nightland part of it just for the atmosphere though. The House of Silence is CREEPY.
See, this is how we'll eventually achieve general purpose AI. People will just keep making more and more elaborate bot checks and AI will just get better and better at fooling them until its able to do anything a human can do, lol.
Oh, I have a firewall as well, which isn't a bad idea of course. Honestly though 99.99% of the threats home systems will face are coming right in over port 80 of course, so it isn't a huge benefit, just more a matter of thoroughness and keeping a low profile.
I think actually a good 50% of what keeps me nice and clean is just running Linux. Nobody really bothers to build targeted attacks for it unless they're trying to spear you specifically, and if they know what distro I'm running then well I've already slipped up some.
I agree with you, there are many ways to be reasonably secure. In fact there will never be a 'one size fits all' set of practices. I think something like NoScript is worth it though. You can always dial it down some and it will still nail those annoying clickjack attempts and whatnot that show up now and then. I think blacklisting like 5 big ad networks will get you probably 80% of the benefit there too. XSS, CSRF, etc are all too common and can be nasty. You won't catch them all, but it is basically a numbers game at this point.
Heh, you'd be surprised at how 'dirty' the general web is... Believe me, I've taught infosec courses and built a number of line-of-business web facing applications. You're less safe than you would assume. Only a fraction of the bad stuff out there is routinely detected.
I can't say what other people consider an inconvenience, but I spend a LOT of time online and I think not having to deal with the consequences of bad stuff is well worth the minor inconvenience. In any case I also get a nice performance boost since I'm not stuck waiting for a lot of ad servers and BS. 99% of sites function fine if you just enable the one or two scripts that they routinely use. Most other stuff is either extraneous or is 3rd party ad scripts served off some ad network somewhere. Reduces the amount of tracking you're subjected to as well.
Each to his own though. I just like sleeping well at night...
lol, you guys still allow ANYTHING 3rd party to load on a page, at all? Forget it. NoScript, block everything, selectively enable stuff that I want to see. Once you get used to the idea that many sites will need a temporary permission or two its great and only a relatively small subset of ads get through.
Well, I can't speak to costs in other places. Where I am you have 2 choices of network, AT&T or Verizon. Verizon will require a 2 year contract on a data plan, which is basically $100 a month. Subtract out the $35 a month I'd pay anyway for my existing plan, and then add $300 for the hardware. Toss in an app or two now and then and you're getting right up there in the $80+ a month range. Of course I could buy a phone unsubsidized and just drop it onto my existing plan, but there's no data at all on that except at the 'pay as you go' highway robbery rate, and a decent phone will set you back north of $600. I'm pretty sure AT&T is at least equally expensive and their network has crappier coverage (as it is neither of them will sustain a reliable connection at my office or home). Honestly though even if the reception was rock solid it is still a rather expensive luxury.
Yeah, the cost is just rough. With the feature phone I've got I can do simple events/alarms. It has a calculator, though honestly I haven't used it more than a couple times in 5+ years. In a pinch I could access my email. I can play MP3s if I want, take pictures, etc. If I need to console into some machine it isn't going to be from the road and if it is I'd probably go insane using a smartphone. Now, I'd be mighty pleased if my phone had all the features that a smartphone has, but I'd use them for anything vital once in a blue moon, and at $1000 a year I can think of many more interesting ways to spend my money.
I don't really doubt that there are people for whom it is worth the extra cost. Sales guys that are on the road all day, etc. Maybe a few tech people. If I could get all those features for an extra $5 a month? Yeah, I'd do it. At close to $100 a month premium all told it is just too pricey. I'd guess in 5 years or so what is now a smartphone will be a $20 device and service will hopefully be equally much cheaper. Of course by then the cutting edge people will be using something far more advanced I suppose.
LOL, I just think that about 95% of the people who have them don't need really need them. It is a luxury/toy, which is OK, but its darned expensive one when you figure you're paying an extra $100 a month probably by the time you add it all up. There's a LOT you can do with that much cash. I consult for traders and constantly have equipment that is up 24/7 and people scream if it isn't, yet I still don't need a smart phone, just something that will get an SMS. Heck, a pager would do fine if such things still existed.
I suggest you read the system reliability analysis. Your assumptions are quaint but wrong. Yes you can argue that the our avoidance of a terrible accident for the last 60 years or so is some kind of proof, but your result is like 0.5 sigma.
Beyond all of that there are so MANY issues. Here we are, the great man-ape twisting ALL the dials on EVERY natural geochemical cycle, carbon, nitrogen, phosphorus, mercury, etc. I could name 20 other critical issues without even needing to go look them up. Many will be trivial or non-issues, but which ones. Where's the killer lurking? What thing is modern technological civilization doing right now that is going to rear up its ugly face and whack us down? We probably won't even realize it matters until after its too late. Look at H5N1, there are people debating whether or not its ethical to release the recipe that WE ALREADY HAVE for a strain that could kill half the human race (probably exaggerated, but 20 years from now's biotech will do it in a garage). You think we can survive that? What if we DID invent effectively limitless energy as say cheap fusion power? You think the possession of that kind of energy isn't a means to completely rework the Earth? You think we have the wisdom to know what not to fuck with? I don't.
No, asteroids and some fantastical coming techno-day-of-revelation are seriously the least of our worries. This race and this planet are on the edge of a whole slew of cliffs. If you want to survive you look where your feet are being planted, not at some horizon you will never reach if you stumble today.
I mean really, we have a human race which is eating itself out of house and home, destroying the environment and every other species and the entire biosphere at a rate never before encountered in the history of life on Earth, AND rapidly acquiring ever greater capabilities to destroy itself on a daily basis while retaining the basic ethical outlook of fire-wielding cavemen. Meanwhile these people are wasting their time wool-gathering about infinitely more remote possibilities like asteroid impacts and total hypotheticals like 'technological singularity' which may well simply not even exist. Doc, don't you realize there are 50,000 nuclear weapons on hair-trigger alert pointed at us every day and that a reasonably systems analysis of US and Russian nuclear 'defense' systems indicates there's roughly a 50/50 chance we will set them off within the next 30 years? Seriously?
This kind of speculation is perhaps intellectually interesting, but the probability that it is in any way a meaningfully useful line of inquiry is remote in the extreme. Can we please have SOME degree of effort put into what is clearly threatening us today? What fool worries about getting his mortgage payment out on time when there's a HUGE FIRE BURNING THE LIVING ROOM! Duh!!!???
Slashdot Mirror
Of course that doesn't mean we're handling them as well as others may have in the past. I don't subscribe to the notion that we're especially better or worse though. It is just an eminence illusion, today's problems always seem worse, more urgent, more significant, than those we read about in history books (sorry, on Wikipedia). lol.
I think the assumption that there's anything impossible about doing it is what is wrong. I also think we are definitely reaching the levels of processing power necessary to meaningfully attack these sorts of problems. Nothing will every do it exactly like a human, but the path to doing it better than the best human is clearly open ahead of us. I can see a world where you can have EVERYTHING parsed for you into an ideal form for you particularly to understand it best. Is the govt going to get a super-human system out of this? No, probably not, it will be a few more generations. The work is interesting though and will lead to even better stuff. It may well be usable as-is too.
Are you maybe confusing metaphor with idiom? Idiomatic speech could truthfully be impossible to interpret without knowledge. A metaphor OTOH is a type of analogy. It operates on the basis of similarity or likeness with something else. It is thus amenable to logical analysis. Take your example, would someone who has not heard this example before understand it? I think it is self evident the answer is yes. It is therefor proven that this metaphor is open to computational analysis.
As language changes some metaphors could become incomprehensible or nonsensical, sure. This is just a problem for all linguistic analysis at some level. Again, the question is are you confusing metaphor with idiom. Most idiom arises somehow out of logical association, but it can be so idiosyncratic that it isn't really amenable to much analysis. Still, I think analysis of these kinds of speech constructs is possible.
So, got a schedule for the big world takeover? Really, I haven't seen even the faintest sign of fact-based ANYTHING in this country since 1980. Things are starting to go seriously wrong...
Eh, I have yet to meet a firearm that gave me any more than the most trivial problems. Certainly left-handed firearms are nice, but in a practical sense both sorts work fine. Working the safety on a pistol quickly is probably the least convenient feature of wrong-handed hand guns.
It isn't a big deal. Most everyday items don't favor one hand or the other. Spears, knives, and even swords pretty much fall into this category, as do quite a lot of other tools. Scissors and similar cutters are sort of an exception. So chances are this wasn't a really large problem. In a pinch you can just use things right-handed (and most lefties are less hand-polarized than righties, maybe do to the convenience of being able to use either hand, for instance I'm perfectly happy doing many tasks with either hand like cutting with scissors).
As for greater cooperation if you have two warriors and one is left handed he can guard the more exposed right flank of his partner and vice-versa. Alexander the Great had a whole cadre of left handed warriors who would take up the far right hand flank of his army (and I'd suspect this was probably fairly standard for ancient armies). The question then is really is there a strong enough difference in cooperation vs competition advantage from different-handedness for it to matter. This study doesn't seem to have addressed that question. At best it can be used like a prior to say "well, maybe it is that way since we DO have 10% lefties." I'm always a little suspicious of that sort of "Bayesian" kind of interpretation since it also leads to such absurdities as the notion that this generation will be the last one in history (do some reading on Bayesian Analysis, lol).
No, it wasn't actually...
"Cooperation favors same-handedness—for sharing the same tools, for example. Physical competition, on the other hand, favors the unusual. In a fight, a left-hander would have the advantage in a right-handed world."
This is simply taken axiomatically as a starting point for the study. I see no indication that it was determined by any sort of analysis. I'm not even sure such an analysis is feasible. You'd have to know what activities people carried out in prehistoric times, how, and what the value of cooperation was for each one. There could be various advantages and disadvantages of same or opposite handedness depending on the activity, etc. The entire concept of their study rests ENTIRELY on the validity of this same-handedness is better for cooperation proposition. I'm not saying it is untrue, but without demonstrating it to be true and to what degree I cannot see how any meaningful conclusions can be drawn.
are like the wolves amongst you right-handed sheep, right!!!!???
I'm not ENTIRELY convinced, what about situations where it is advantageous for people to have opposite handedness for optimal cooperation? There seems to be a built-in assumption here that different-handed assortments of people will always have more problems working together. I'm not sure there's a practical way to test this as a general thing though.
That's new! ROFL. Lets see, which member's districts will this money go to...
You don't give your support to something that is broken, you oppose it and negotiate the bad parts out. How exactly by supporting this thing do they think they're going to get it changed? This is Bullshit. If I was Mr Zuckerberg I'd be careful what I wish for...
Hmmm, that's similar to setting yourself on fire to get revenge on someone, hehehe. ;)
Not that I'm unsympathetic to the AC, I have no children either, and no such plans. I may get some pleasure out of that I guess, as I starve with the rest of you locusts!
Yeah, and anyone who is childless of course will be fine... Remember, there ain't no justice in this world, we're all in it together.
Heh, well, we were using ADA and/or PACE PLI which did the same thing. We still had unit testing to the Nth degree. Heck, we had complete design verification by simulation at the module, assembly, and system levels on top of that, lol.
Of course no 747/57/67/A320/30/40 etc with any of my code flying in it has to my knowledge yet fallen out of the sky due to bad code, lol. I really have to wonder about all these rocket guys and their bad code. Can't even imagine how you could use tools that could allow a type mismatch in that kind of system, crazy. Heck, we coded (and built the hardware for) the range safety destruct box on Titan 34D. Now talk about "this is not allowed to ever have a failure mode, period" yeah. Ain't no stinking C++ code on that baby, nope. You push the big red button, there's only one answer, kaboom! ;)
Yeeehaaa! ;)
The tough problems aren't about running the code and seeing what happens, they're about setting up very specific situations and testing them easily.
World Enough and Time, and its sequel (The name of which I forget).
First and Last Men, and Starmaker. Olaf Stapleton is interesting.
And then there is the weird stuff, I suggest William Hope Hodgeson. Boats of the Glenn Carrig is pretty strange. Nightland is a work of genius, but also pretty much unreadable. It is probably worth reading the early Nightland part of it just for the atmosphere though. The House of Silence is CREEPY.
See, this is how we'll eventually achieve general purpose AI. People will just keep making more and more elaborate bot checks and AI will just get better and better at fooling them until its able to do anything a human can do, lol.
Oh, I have a firewall as well, which isn't a bad idea of course. Honestly though 99.99% of the threats home systems will face are coming right in over port 80 of course, so it isn't a huge benefit, just more a matter of thoroughness and keeping a low profile.
I think actually a good 50% of what keeps me nice and clean is just running Linux. Nobody really bothers to build targeted attacks for it unless they're trying to spear you specifically, and if they know what distro I'm running then well I've already slipped up some.
I agree with you, there are many ways to be reasonably secure. In fact there will never be a 'one size fits all' set of practices. I think something like NoScript is worth it though. You can always dial it down some and it will still nail those annoying clickjack attempts and whatnot that show up now and then. I think blacklisting like 5 big ad networks will get you probably 80% of the benefit there too. XSS, CSRF, etc are all too common and can be nasty. You won't catch them all, but it is basically a numbers game at this point.
Heh, you'd be surprised at how 'dirty' the general web is... Believe me, I've taught infosec courses and built a number of line-of-business web facing applications. You're less safe than you would assume. Only a fraction of the bad stuff out there is routinely detected.
I can't say what other people consider an inconvenience, but I spend a LOT of time online and I think not having to deal with the consequences of bad stuff is well worth the minor inconvenience. In any case I also get a nice performance boost since I'm not stuck waiting for a lot of ad servers and BS. 99% of sites function fine if you just enable the one or two scripts that they routinely use. Most other stuff is either extraneous or is 3rd party ad scripts served off some ad network somewhere. Reduces the amount of tracking you're subjected to as well.
Each to his own though. I just like sleeping well at night...
lol, you guys still allow ANYTHING 3rd party to load on a page, at all? Forget it. NoScript, block everything, selectively enable stuff that I want to see. Once you get used to the idea that many sites will need a temporary permission or two its great and only a relatively small subset of ads get through.
Well, I can't speak to costs in other places. Where I am you have 2 choices of network, AT&T or Verizon. Verizon will require a 2 year contract on a data plan, which is basically $100 a month. Subtract out the $35 a month I'd pay anyway for my existing plan, and then add $300 for the hardware. Toss in an app or two now and then and you're getting right up there in the $80+ a month range. Of course I could buy a phone unsubsidized and just drop it onto my existing plan, but there's no data at all on that except at the 'pay as you go' highway robbery rate, and a decent phone will set you back north of $600. I'm pretty sure AT&T is at least equally expensive and their network has crappier coverage (as it is neither of them will sustain a reliable connection at my office or home). Honestly though even if the reception was rock solid it is still a rather expensive luxury.
Yeah, the cost is just rough. With the feature phone I've got I can do simple events/alarms. It has a calculator, though honestly I haven't used it more than a couple times in 5+ years. In a pinch I could access my email. I can play MP3s if I want, take pictures, etc. If I need to console into some machine it isn't going to be from the road and if it is I'd probably go insane using a smartphone. Now, I'd be mighty pleased if my phone had all the features that a smartphone has, but I'd use them for anything vital once in a blue moon, and at $1000 a year I can think of many more interesting ways to spend my money.
I don't really doubt that there are people for whom it is worth the extra cost. Sales guys that are on the road all day, etc. Maybe a few tech people. If I could get all those features for an extra $5 a month? Yeah, I'd do it. At close to $100 a month premium all told it is just too pricey. I'd guess in 5 years or so what is now a smartphone will be a $20 device and service will hopefully be equally much cheaper. Of course by then the cutting edge people will be using something far more advanced I suppose.
LOL, I just think that about 95% of the people who have them don't need really need them. It is a luxury/toy, which is OK, but its darned expensive one when you figure you're paying an extra $100 a month probably by the time you add it all up. There's a LOT you can do with that much cash. I consult for traders and constantly have equipment that is up 24/7 and people scream if it isn't, yet I still don't need a smart phone, just something that will get an SMS. Heck, a pager would do fine if such things still existed.
I suggest you read the system reliability analysis. Your assumptions are quaint but wrong. Yes you can argue that the our avoidance of a terrible accident for the last 60 years or so is some kind of proof, but your result is like 0.5 sigma.
Beyond all of that there are so MANY issues. Here we are, the great man-ape twisting ALL the dials on EVERY natural geochemical cycle, carbon, nitrogen, phosphorus, mercury, etc. I could name 20 other critical issues without even needing to go look them up. Many will be trivial or non-issues, but which ones. Where's the killer lurking? What thing is modern technological civilization doing right now that is going to rear up its ugly face and whack us down? We probably won't even realize it matters until after its too late. Look at H5N1, there are people debating whether or not its ethical to release the recipe that WE ALREADY HAVE for a strain that could kill half the human race (probably exaggerated, but 20 years from now's biotech will do it in a garage). You think we can survive that? What if we DID invent effectively limitless energy as say cheap fusion power? You think the possession of that kind of energy isn't a means to completely rework the Earth? You think we have the wisdom to know what not to fuck with? I don't.
No, asteroids and some fantastical coming techno-day-of-revelation are seriously the least of our worries. This race and this planet are on the edge of a whole slew of cliffs. If you want to survive you look where your feet are being planted, not at some horizon you will never reach if you stumble today.
I mean really, we have a human race which is eating itself out of house and home, destroying the environment and every other species and the entire biosphere at a rate never before encountered in the history of life on Earth, AND rapidly acquiring ever greater capabilities to destroy itself on a daily basis while retaining the basic ethical outlook of fire-wielding cavemen. Meanwhile these people are wasting their time wool-gathering about infinitely more remote possibilities like asteroid impacts and total hypotheticals like 'technological singularity' which may well simply not even exist. Doc, don't you realize there are 50,000 nuclear weapons on hair-trigger alert pointed at us every day and that a reasonably systems analysis of US and Russian nuclear 'defense' systems indicates there's roughly a 50/50 chance we will set them off within the next 30 years? Seriously?
This kind of speculation is perhaps intellectually interesting, but the probability that it is in any way a meaningfully useful line of inquiry is remote in the extreme. Can we please have SOME degree of effort put into what is clearly threatening us today? What fool worries about getting his mortgage payment out on time when there's a HUGE FIRE BURNING THE LIVING ROOM! Duh!!!???