You write that "The science shows that free markets cause peace", and that lack of economic freedom causes both poverty and war. But correlation is not causation.
I can also argue that poverty and war cause lack of economic freedom: poverty can be seen as a lack of property (or means to acquire it), and there can't be much of a free economy, without circulation of property and means. War destroys civilian lives, property and freedoms, and results in poverty for those involved. Not exactly the recipe for a free economy.
Or you can conclude that free markets, individual freedoms and peace are all positively interrelated. I have the hope that China will need to embrace political freedoms as its markets grow, if only because lack of these freedoms results in a suboptimal economy: free press and political accountability are the only way to stem corruption and reward good risk-taking.
Sure, Java has a great security model and will not cause buffer overflows. But you have to write it in (duh) Java.
The fun part about NaCL is that it can eat existing (C, C++, pick-your-own compiled language) code with only minor modifications to the compile chain, as long as that code does not make weird system calls. Just make sure that the compiler does not echo any of the 'forbidden' instructions, aligns jumps to 32-bit boundaries, and uses the prescribed instruction sequences for jumps and system calls. And they provide modified versions of GCC that do just that. The paper also says that, in their experience, modification of most programs they tried was, at most, a problem of "a few hours".
If I had to port an existing app to run as a sort of browser plugin, guess what sounds better: a full rewrite in Java, or a few changes to the Makefile. Because *that* is the selling point of NaCL.
Not an expert in biology, but unless these contaminated areas have been contaminated for a very long time (read tens of thousands of years), and are quite large, the chances for life to have sprung up seem very, very slim. Current life needed millions of years to gain a firm foothold and start building up complexity. Lucky meteorites aside, starting from zero is bound to be hard.
If the experiment succeeds (here or elsewhere), and something "life-ish" is found, the results will still be tricky to classify. Can a given chemistry lead to increasing complexity, or is it just a dead end? Without hindsight, this seems like a very difficult question.
You can turn that around, and say that without freedoms, safety is worth very little. Identity theft and computer hacks could be stopped on their tracks by restricting the 'net to a whitelist of "licensed" sites, outlawing anonymous access, and mandating logs of all traffic for careful inspection by "authorities". Would you advocate such a design?
it amazes me how many stories along the lines of 'we can make scientific publishing work better' that get on Slashdot.
So, in your humble opinion, science publishing should continue essentially as it is, perhaps with a shift towards open publishing. And the current 'science accounting' methods (adding up citation sum-totals, relying on journal weights, or combining both) are perfectly adequate. And there the pressure from these flawless accounting systems does not drive scientists to publish un-matured, partial results in a frenzy to score more papers than their peers.
No, I do not have a silver bullet, and yes, the system sort-of, kind-of works. But I am sure that it can be improved. Funding depends very heavily on accounting.
A social network, if divorced from editors and anonymous reviewers, may not be so bad as you paint it. I, for one, would not mind to see rated and commented papers...
When building its databases, Google does not treat all links as equal. You have the 'rel=nofollow' link attribute to indicate that you don't want to attribute trust to the destination of the link, and Google can discount link weight based on the outgoing anchor text.
I would be very interested in seeing that done in scientific publications. As you said, notability can come in many flavors, and they are not equally yummy.
[the number of references] already exists and is widely used as a metric
Citing a paper does not necessarily mean endorsement for its contents. Only that the paper you are referencing was relevant to a part of your discussion. References can be used to provide counter-examples or denounce bad research; but they are counted as a citation anyway. In scientific citation number-crunching, any publicity is good publicity.
A crazy idea would be to add metadata to references, describing the type and relative importance of the source. That would make 'paper A, main inspiration, very important' actually count more than 'paper B, cited in an off-hand comment to exemplify bad research in this field'. The crazy part is changing the established format of scientific papers to accomodate this metadata, standarizing the metadata, and convincing authors to adhere to standards and editors to enforce them.
Placing landmarks on a one-dimensional scale is a poor way of defining political thought. Additionally, if you ask a random sample for their definitions of 'left' and 'right' there would be little consensus. Trivia: 'left' and 'right' come from seating arrangements in the french parliament. The scale is not only one-dimensional: there is no consensus on the exact dimension.
Defining individualism and colletivism, beyond the extremes (cave or hive), is also quite hard. Would social solidarity count as collectivism? Is an active voice in political matters a signal of individualism (by adding liberty), or collectivism (by collaborating with the government)?
Truly nice way of getting high-level information from graphs. I'll probably use it sometime in the future, as I actually work with graph viz, and was looking for ways to visually compare related graphs.
And for those too lazy to actually open the PDF and look at the pictures, the idea is to draw histograms of 1st order vertex degree up to nth order vertex degree (where order-l vertex degree gets defined as how many other vertices can be reached in exactly 'l' forward hops, as found by a breadth-first search).
First of all, congratulations. No sarcasm here. You rose to the challenge of offering a solution where you picked a side based on objective considerations (i.e. - Israel's greater ability to impose a consistent policy), and that plan only relates to one side.
Thanks; you have also managed to keep your text free of claims of historical righteousness. No mean feat with the conflict so near to your doorstep.
A TV satire from the time I talked about earlier showed Shimon Peres, then prime minister, with VR goggles on talking about "a new middle east, peace in our times", while the background showed burning buses. No country should be asked to withstand that, and no politician in any democracy can afford to.
I have always wondered how the problem is reported in Israeli media. I did expect something similar, but can't really judge how widespread that "stop it now, no matter how" sentiment was and is. On the other hand, there are bound to be people on the Israeli side that will be opposed (either by principle, political profit, or plain greed) to any peace gesture. I never said the "being generous" plan was easy, but I can't think of any other.
I think the key to ending the conflict is not by agreeing to tolerate terror, but by raising the standard of living for the Palestinians. Give them hope.
Completely with you on the hope side. No hope means nothing to lose. On the other hand, terrorists will keep terrorizing regardless of what you do (either to keep their stranglehold or to avenge Israeli raids). If you fight them heads on and any damage spills (and it will, and it does), you are feeding their ranks with new desperadoes. You *do* need a lot of courage to hold back fire, but I don't see any other long-term answer; except keeping things like they are now: a slow trickle of deaths, for a long, long time.
Another part of the problem is that the Palestinians have lost sight of what their objectives are. Analyzing their actions show that "having their own state", "living in safety" and "economical prosperity" cannot conceivably be their goals, as their actions do not further and of them.
So what would their goals be? My bet is that Gazans thought that anything would be better than their previous corrupt politicians. They probably saw some kind of hope in voting for Hamas: after all, Hamas seemed like half-way competent (as in not stealing so blatantly), and besides, how much worse could things get? Don't underestimate the effect of nationalistic propaganda and lack of options on desperate people. Anything that makes them less desperate will increase chances for rational thought.
Maybe the solution is to continue doing what we are doing with Gaza, while doing everything we can to make life in the west bank better.
Good for the West Bank, but you won't convince too many Gazans that way. Stiff blockades breed a lot of resentment, and it is always easier to blame the guys imposing the blockade than to blame your own people. And nobody likes to be called a traitor.
Currently, Palestinians face a quite hopeless situation. Even if a great majority [1] wants peace and quiet for a change (and I am sure most would take it offered a credible chance), a stifling blockade (however justified) and a complete lack of institutions makes this hard to achieve. And the extremists will make sure that Israel's "no terrorism" conditions are never met [2] and the situation stays put. The future, from a Palestinian point of view, is not bright at all.
From the Israeli side of the equation, a great deal of political courage is required to stop the current knee-jerk response [3] to terrorism (namely: blockade and retaliation). However, Israel is the only side that can actually change the dynamics of the situation. Unless Palestinians see a "true chance" of stable peace, in spite of the inevitable terrorists [4], the current state of affairs will go on and on. Israel has well-established institutions and can actually enforce its own decisions.
I really believe that only great generosity on the Israeli side can put a stop to things. I don't care who started what - but the only way out of this hole is to either eliminate one side entirely (proposed by some, but unlikely and abhorring) or peaceful coexistence. Coexistence requires a compelling future for both. The Israelis have it (in spite of a constant trickle of deaths); the Palestinians don't.
[1] I don't care what statistics say about Palestinian bloodthirstyness - most humans like raising their kids in peace and prosperity. Palestinians are no different; however, when this is impossible, all humans will wish ill towards whoever they see as "preventing" this.
[2] Yes, some like things the way they are and profit from terrorism. Religious fanaticism, power, prestige, or plain greed. All the worse when combined. However, point [1] still stands.
[3] When rockets rain on a civilian neighborhood, it is perfectly understandable to want to beat the crap out of whoever is launching them, and to make sure that rocket-building materials are hard to obtain. But, in the longer run, this can backfire.
[4] Palestinian terrorism will continue, even if Israel lifts sanctions and gives Palestinians a future worth caring about. It will, given time and sound policy, peter out. Peace is a long-term strategy.
Unfortunately, there is no touchscreen equivalent to just moving the mouse pointer without clicking.
Depends on technology. In many cases you can, indeed, spot fingers a short distance away from the display. On the other hand, if you have multi-touch, you can define an alternate gesture to indicate the same type of interaction. You don't need to actually hover your finger to mean "do as if hovering a mouse pointer".
Hovering is vital for traditional mouse-based interfaces, since you can't know where the pointer is unless you see it on screen. Its use is less clear if you are pointing directly at the point you want to interact with. It sure can't help you to fine-tune the exact point you are selecting, since touch-based displays suffer from occlusion -- although several methods have been proposed to deal with this problem.
True (to my surprise). However, most architectures use stacks to implement argument-passing (I am not aware of any that do not), and understanding how something is expected to work is key when finding out what went wrong.
Java's means of keeping you from writing to random locations in memory is something of a hindrance when trying to use anything memory-mapped.
True, java would be the wrong tool for that job. Does that disqualify java for teaching algorithms or high-level program architecture?
It's a tradeoff. Can you use the big-boy software, or do you need to have the kind that doesn't have small parts that come off and you can choke on?
Not really. It is about using the best tool for the job, the job being teaching people how to program. I still think it is better to learn a few things a time (algorithms are complex enough without memory management issues) than to try to do everything at once, badly.
Yes, typing on a touchscreen is hard. Especially touch-typing. Fortunately, you can still use a keyboard, and there is no reason to stop using a mouse if you feel comfortable with it.
On the other hand, imagine a large enough screen; say, 5k pixels wide. Positioning a mouse pointer accurately in such a large area requires progressively more time, and it is hard to locate the cursor if you forget where you left it last time. Touching at a coordinate and knowing that it is fully equivalent to a mouse-click sounds like a nice proposition. Being able to trigger actions at distant parts of the screen without long "travel times" also sounds fine. You have probably have two hands with lots of fingers on each - it would be easy to use one to select a tool from a toolbox and another one to use the tool in a distant area of the desktop. Yes, keyboard shortcuts are faster - but you have to learn them first.
Of course, if you do have such a large touchscreen, you will probably want to mount it more like a drawing table than like a traditional upright screen, easing strain on your arms if you have to reach for things... but there is still no reason to stop using keyboards or alternative pointing devices. The worst-case scenario is that you will have to move windows around a bit to avoid occluding the active window with the keyboard or mouse.
Have you tried a pen on a suitable touch-enabled surface? Touch-enabled displays need not be limited to blunt fingers.
In the linked project, researchers demonstrated that interaction with pen & hand beat interaction with two pens or two hands on a single large touch-enabled display. Are you sure that CAD/CAM would be harder with this setup than on your own?
Machines "can" do all sorts of wonderful things, but so far nobody has been able to get them to do them. Please cite examples or research that demonstrates accurate real-time 3d modeling from a synthetic 2d video, or stop making things up. Yes, I am sure it can be done. No, I very much doubt anybody can do it right now.
You say that synthetic video is finite. So is 2^256, or the number of grains of sand in the beach - what is your point? Are you suggesting that finite is always manageable?. Ok, maybe you can pick up a head from the background, and even shoot at it real quick. Good luck with navigating the map and defusing the bomb with visual input alone; fortunately, CS is not all about sniping, and a good human team can easily counter a dumb-but-accurate ai-sniper team. Probably with lots of grenades and shields, and good ol' cooperation to rush weak spots.
My analogy to the RC car hints at a larger problem. FPS games are not entirely self-contained, they try to reflect a lot of things that, for an AI, should also be considered rules: a 3D world, geometry, light, physics, ballistics, sound propagation. That is a huge amount of implicit rules we are talking about. And some of those (such as robust visual processing) are tough to teach. Playing chess is easy - just throw in more processing power. Vision is hard: nobody knows exactly what to do with extra power.
ANNs are still in their infancy. It is not that our current models break down in certain cases -- they only work well in certain very specific tasks, and we are quite clueless regarding how to design, say, a roach.
games have rules. Once you've learned the rules, you're unstoppable
Ah, but I want both to play the same game, while your are suggesting giving the machine a special representation with a high-level vocabulary. The interface I am proposing is the same one you are using: images and sounds come out and commands are executed. Not "high-level game data" - only images and sounds.
You talk about teaching the AI how to headshot. I am talking about the difficulty of processing a 2D image and interpreting it as a PoV rendering of an (unknown) 3D model, and locating a set of pixels that happen to compose the 'head' of something that is an 'enemy'.
Tell you what: give me $50,000 in funding, six months to train the AI to general FPS rules (headshot, movement, general weapon effectiveness, etc.), and another six months for the GA to advance it for CS, and it will beat anything.
If I give you 50k $, will you sign a document agreeing to deliver accurate 3D model-building from a (sythetic) video in real-time or give me my money back plus interest rates?. That's what I was referring to when I wrote about the DARPA Grand Challenge. You and me can both pilot a remote-control car with a wireless lo-res camera slapped on top. Try to teach a computer to do that reliably, and you have solved many, many problems in AI.
You are overly optimistic regarding GAs. I very much doubt that, whatever the AI cycles, you can evolve an AI team capable of winning against a good (as in top-third of the table) human team at, say, counter-strike.
Gotcha: the AIs would not be allowed to cheat, and would have the exact same information at their disposal as the human team: a lot of visual input (but not the actual noise-free game geometry) and the same set of commands as human players.
If you can evolve that kind of AI, the DARPA Grand Challenge is looking for you.
Depends on the setting. It makes perfect sense within a networked-battlefield scenario: what one unit sees, it will try to convey to others, and commanders can take decisions based on everything that is seen by any of their units.
In a medieval setting, they would have to shout to each other (and be within hearing distance) to request assistance. Or wave colored flags, or send messenger pigeons.
The radio squawking in Half-Life added an element of realism to this - you could actually "hear" what the bad guys were "saying" to each other. Even though it was crude: it would have been even better to hear things like "he's hiding behind the wall" or "he's nailed Bob! the #@!"
Additionally, a multilevel AI helps to even out the use of CPU. You don't want to take high-level command decisions every quarter second, because that would not leave enough time for any of the lower tiers to do anything useful, and would require too much CPU. With a multi-tiered AI, you can dedicate small and frequent AI slots to low-level decisions, and hold high-level decisions for less frequent and more CPU-intensive thinking.
C hides stacks from you. I don't know why you have to teach them to anyone.
You don't have to teach it, until people start complaining that they get weird errors after overflowing a stack-allocated array. Knowing what the stack is and how it is used really helps then.
And Java and C++ don't eliminate stacks.
I never said that Java and C++ did not use them. But overwriting the stack with garbage is much harder under Java and other memory-managed languages. Which C++ is not, even if you can overload 'new'.
In theory, you don't need to teach pointers as arrays of memory locations. You can show them as references with different behaviors and rules for correct use.
You can teach pointer semantics out of thin air, but I think it makes more sense if you tell people where that comes from. If you teach C, you may as say why it was designed the way it was designed - and why that makes C programs prone to fail in interesting ways.
Try to teach pointers to someone who is not familiar with the concept of memory as a large, one-dimensional string of bytes - where a byte can contain a part of an sizeof(int)-byte integer, or a single character, or a part of a sizeof(void *)-byte memory address to yet other interesting bytes.
Try to explain stack corruption to someone who does not understand the concept of pushing things into the stack when you call a function and popping them back out when you return, and using that same stack space to store your local variables.
If you can't get more out of C than its more convoluted syntax, you never learned C properly.
I think he is just mad with people getting away with "oh, yes - there's a bug somewhere, must have crept in while I was not looking". Dijkstra had a quite extreme view on programmer errors.
Written in '88, it seems to me to have aged quite well. And for all the intellectual smugness, there a few tidbits worth keeping.
The 9 orders of magnitude that have to be spanned when thinking on computation are quite true. The gap may have expanded with even faster computers and more ambitious projects. And the fact that thinking discrete is very different from thinking continuous is also well made. Knowing what is difficult is a good protection against overconfidence. One of the hard lessons of computers is that what you think it does and what it really does are not always in sync.
On the educational aspect, dumbing things down, if done only for the heck of it, does result in dumber students. If all you meet are small, uniform, incremental steps -- how can you expect to learn to leap?.
Yes, hammering only theory and proof-of-correctness to freshmen is indeed cruel and, furthermore, ineffective (except for learning algorithmic thought). As other posters have said, business requirements are not amenable to formal proofs. Ivory-towerness aside, I think it makes for a provocative read. And if profs are not allowed to shout from the tower once in a while, they would go mad.
Slashdot Mirror
You write that "The science shows that free markets cause peace", and that lack of economic freedom causes both poverty and war. But correlation is not causation.
I can also argue that poverty and war cause lack of economic freedom: poverty can be seen as a lack of property (or means to acquire it), and there can't be much of a free economy, without circulation of property and means. War destroys civilian lives, property and freedoms, and results in poverty for those involved. Not exactly the recipe for a free economy.
Or you can conclude that free markets, individual freedoms and peace are all positively interrelated. I have the hope that China will need to embrace political freedoms as its markets grow, if only because lack of these freedoms results in a suboptimal economy: free press and political accountability are the only way to stem corruption and reward good risk-taking.
Sure, Java has a great security model and will not cause buffer overflows. But you have to write it in (duh) Java.
The fun part about NaCL is that it can eat existing (C, C++, pick-your-own compiled language) code with only minor modifications to the compile chain, as long as that code does not make weird system calls. Just make sure that the compiler does not echo any of the 'forbidden' instructions, aligns jumps to 32-bit boundaries, and uses the prescribed instruction sequences for jumps and system calls. And they provide modified versions of GCC that do just that. The paper also says that, in their experience, modification of most programs they tried was, at most, a problem of "a few hours".
If I had to port an existing app to run as a sort of browser plugin, guess what sounds better: a full rewrite in Java, or a few changes to the Makefile. Because *that* is the selling point of NaCL.
Not an expert in biology, but unless these contaminated areas have been contaminated for a very long time (read tens of thousands of years), and are quite large, the chances for life to have sprung up seem very, very slim. Current life needed millions of years to gain a firm foothold and start building up complexity. Lucky meteorites aside, starting from zero is bound to be hard.
If the experiment succeeds (here or elsewhere), and something "life-ish" is found, the results will still be tricky to classify. Can a given chemistry lead to increasing complexity, or is it just a dead end? Without hindsight, this seems like a very difficult question.
In my opinion the French military should rather develop its own national operating system.
Not another branded "national OS" flavour of Linux.
You can turn that around, and say that without freedoms, safety is worth very little. Identity theft and computer hacks could be stopped on their tracks by restricting the 'net to a whitelist of "licensed" sites, outlawing anonymous access, and mandating logs of all traffic for careful inspection by "authorities". Would you advocate such a design?
So much for extreme examples.
it amazes me how many stories along the lines of 'we can make scientific publishing work better' that get on Slashdot.
So, in your humble opinion, science publishing should continue essentially as it is, perhaps with a shift towards open publishing. And the current 'science accounting' methods (adding up citation sum-totals, relying on journal weights, or combining both) are perfectly adequate. And there the pressure from these flawless accounting systems does not drive scientists to publish un-matured, partial results in a frenzy to score more papers than their peers.
No, I do not have a silver bullet, and yes, the system sort-of, kind-of works. But I am sure that it can be improved. Funding depends very heavily on accounting.
A social network, if divorced from editors and anonymous reviewers, may not be so bad as you paint it. I, for one, would not mind to see rated and commented papers...
When building its databases, Google does not treat all links as equal. You have the 'rel=nofollow' link attribute to indicate that you don't want to attribute trust to the destination of the link, and Google can discount link weight based on the outgoing anchor text.
I would be very interested in seeing that done in scientific publications. As you said, notability can come in many flavors, and they are not equally yummy.
[the number of references] already exists and is widely used as a metric
Citing a paper does not necessarily mean endorsement for its contents. Only that the paper you are referencing was relevant to a part of your discussion. References can be used to provide counter-examples or denounce bad research; but they are counted as a citation anyway. In scientific citation number-crunching, any publicity is good publicity.
A crazy idea would be to add metadata to references, describing the type and relative importance of the source. That would make 'paper A, main inspiration, very important' actually count more than 'paper B, cited in an off-hand comment to exemplify bad research in this field'. The crazy part is changing the established format of scientific papers to accomodate this metadata, standarizing the metadata, and convincing authors to adhere to standards and editors to enforce them.
Placing landmarks on a one-dimensional scale is a poor way of defining political thought. Additionally, if you ask a random sample for their definitions of 'left' and 'right' there would be little consensus. Trivia: 'left' and 'right' come from seating arrangements in the french parliament. The scale is not only one-dimensional: there is no consensus on the exact dimension.
Defining individualism and colletivism, beyond the extremes (cave or hive), is also quite hard. Would social solidarity count as collectivism? Is an active voice in political matters a signal of individualism (by adding liberty), or collectivism (by collaborating with the government)?
Truly nice way of getting high-level information from graphs. I'll probably use it sometime in the future, as I actually work with graph viz, and was looking for ways to visually compare related graphs.
And for those too lazy to actually open the PDF and look at the pictures, the idea is to draw histograms of 1st order vertex degree up to nth order vertex degree (where order-l vertex degree gets defined as how many other vertices can be reached in exactly 'l' forward hops, as found by a breadth-first search).
First of all, congratulations. No sarcasm here. You rose to the challenge of offering a solution where you picked a side based on objective considerations (i.e. - Israel's greater ability to impose a consistent policy), and that plan only relates to one side.
Thanks; you have also managed to keep your text free of claims of historical righteousness. No mean feat with the conflict so near to your doorstep.
A TV satire from the time I talked about earlier showed Shimon Peres, then prime minister, with VR goggles on talking about "a new middle east, peace in our times", while the background showed burning buses. No country should be asked to withstand that, and no politician in any democracy can afford to.
I have always wondered how the problem is reported in Israeli media. I did expect something similar, but can't really judge how widespread that "stop it now, no matter how" sentiment was and is. On the other hand, there are bound to be people on the Israeli side that will be opposed (either by principle, political profit, or plain greed) to any peace gesture. I never said the "being generous" plan was easy, but I can't think of any other.
I think the key to ending the conflict is not by agreeing to tolerate terror, but by raising the standard of living for the Palestinians. Give them hope.
Completely with you on the hope side. No hope means nothing to lose. On the other hand, terrorists will keep terrorizing regardless of what you do (either to keep their stranglehold or to avenge Israeli raids). If you fight them heads on and any damage spills (and it will, and it does), you are feeding their ranks with new desperadoes. You *do* need a lot of courage to hold back fire, but I don't see any other long-term answer; except keeping things like they are now: a slow trickle of deaths, for a long, long time.
Another part of the problem is that the Palestinians have lost sight of what their objectives are. Analyzing their actions show that "having their own state", "living in safety" and "economical prosperity" cannot conceivably be their goals, as their actions do not further and of them.
So what would their goals be? My bet is that Gazans thought that anything would be better than their previous corrupt politicians. They probably saw some kind of hope in voting for Hamas: after all, Hamas seemed like half-way competent (as in not stealing so blatantly), and besides, how much worse could things get? Don't underestimate the effect of nationalistic propaganda and lack of options on desperate people. Anything that makes them less desperate will increase chances for rational thought.
Maybe the solution is to continue doing what we are doing with Gaza, while doing everything we can to make life in the west bank better.
Good for the West Bank, but you won't convince too many Gazans that way. Stiff blockades breed a lot of resentment, and it is always easier to blame the guys imposing the blockade than to blame your own people. And nobody likes to be called a traitor.
Currently, Palestinians face a quite hopeless situation. Even if a great majority [1] wants peace and quiet for a change (and I am sure most would take it offered a credible chance), a stifling blockade (however justified) and a complete lack of institutions makes this hard to achieve. And the extremists will make sure that Israel's "no terrorism" conditions are never met [2] and the situation stays put. The future, from a Palestinian point of view, is not bright at all.
From the Israeli side of the equation, a great deal of political courage is required to stop the current knee-jerk response [3] to terrorism (namely: blockade and retaliation). However, Israel is the only side that can actually change the dynamics of the situation. Unless Palestinians see a "true chance" of stable peace, in spite of the inevitable terrorists [4], the current state of affairs will go on and on. Israel has well-established institutions and can actually enforce its own decisions.
I really believe that only great generosity on the Israeli side can put a stop to things. I don't care who started what - but the only way out of this hole is to either eliminate one side entirely (proposed by some, but unlikely and abhorring) or peaceful coexistence. Coexistence requires a compelling future for both. The Israelis have it (in spite of a constant trickle of deaths); the Palestinians don't.
[1] I don't care what statistics say about Palestinian bloodthirstyness - most humans like raising their kids in peace and prosperity. Palestinians are no different; however, when this is impossible, all humans will wish ill towards whoever they see as "preventing" this.
[2] Yes, some like things the way they are and profit from terrorism. Religious fanaticism, power, prestige, or plain greed. All the worse when combined. However, point [1] still stands.
[3] When rockets rain on a civilian neighborhood, it is perfectly understandable to want to beat the crap out of whoever is launching them, and to make sure that rocket-building materials are hard to obtain. But, in the longer run, this can backfire.
[4] Palestinian terrorism will continue, even if Israel lifts sanctions and gives Palestinians a future worth caring about. It will, given time and sound policy, peter out. Peace is a long-term strategy.
Unfortunately, there is no touchscreen equivalent to just moving the mouse pointer without clicking.
Depends on technology. In many cases you can, indeed, spot fingers a short distance away from the display. On the other hand, if you have multi-touch, you can define an alternate gesture to indicate the same type of interaction. You don't need to actually hover your finger to mean "do as if hovering a mouse pointer".
Hovering is vital for traditional mouse-based interfaces, since you can't know where the pointer is unless you see it on screen. Its use is less clear if you are pointing directly at the point you want to interact with. It sure can't help you to fine-tune the exact point you are selecting, since touch-based displays suffer from occlusion -- although several methods have been proposed to deal with this problem.
C does not require the use of a stack.
True (to my surprise). However, most architectures use stacks to implement argument-passing (I am not aware of any that do not), and understanding how something is expected to work is key when finding out what went wrong.
Java's means of keeping you from writing to random locations in memory is something of a hindrance when trying to use anything memory-mapped.
True, java would be the wrong tool for that job. Does that disqualify java for teaching algorithms or high-level program architecture?
It's a tradeoff. Can you use the big-boy software, or do you need to have the kind that doesn't have small parts that come off and you can choke on?
Not really. It is about using the best tool for the job, the job being teaching people how to program. I still think it is better to learn a few things a time (algorithms are complex enough without memory management issues) than to try to do everything at once, badly.
Yes, typing on a touchscreen is hard. Especially touch-typing. Fortunately, you can still use a keyboard, and there is no reason to stop using a mouse if you feel comfortable with it.
On the other hand, imagine a large enough screen; say, 5k pixels wide. Positioning a mouse pointer accurately in such a large area requires progressively more time, and it is hard to locate the cursor if you forget where you left it last time. Touching at a coordinate and knowing that it is fully equivalent to a mouse-click sounds like a nice proposition. Being able to trigger actions at distant parts of the screen without long "travel times" also sounds fine. You have probably have two hands with lots of fingers on each - it would be easy to use one to select a tool from a toolbox and another one to use the tool in a distant area of the desktop. Yes, keyboard shortcuts are faster - but you have to learn them first.
Of course, if you do have such a large touchscreen, you will probably want to mount it more like a drawing table than like a traditional upright screen, easing strain on your arms if you have to reach for things... but there is still no reason to stop using keyboards or alternative pointing devices. The worst-case scenario is that you will have to move windows around a bit to avoid occluding the active window with the keyboard or mouse.
Have you tried a pen on a suitable touch-enabled surface? Touch-enabled displays need not be limited to blunt fingers.
In the linked project, researchers demonstrated that interaction with pen & hand beat interaction with two pens or two hands on a single large touch-enabled display. Are you sure that CAD/CAM would be harder with this setup than on your own?
http://mi-lab.org/projects/bimanual-pen-touch/
There's a nice video scrolling down the page.
Machines "can" do all sorts of wonderful things, but so far nobody has been able to get them to do them. Please cite examples or research that demonstrates accurate real-time 3d modeling from a synthetic 2d video, or stop making things up. Yes, I am sure it can be done. No, I very much doubt anybody can do it right now.
You say that synthetic video is finite. So is 2^256, or the number of grains of sand in the beach - what is your point? Are you suggesting that finite is always manageable?. Ok, maybe you can pick up a head from the background, and even shoot at it real quick. Good luck with navigating the map and defusing the bomb with visual input alone; fortunately, CS is not all about sniping, and a good human team can easily counter a dumb-but-accurate ai-sniper team. Probably with lots of grenades and shields, and good ol' cooperation to rush weak spots.
My analogy to the RC car hints at a larger problem. FPS games are not entirely self-contained, they try to reflect a lot of things that, for an AI, should also be considered rules: a 3D world, geometry, light, physics, ballistics, sound propagation. That is a huge amount of implicit rules we are talking about. And some of those (such as robust visual processing) are tough to teach. Playing chess is easy - just throw in more processing power. Vision is hard: nobody knows exactly what to do with extra power.
ANNs are still in their infancy. It is not that our current models break down in certain cases -- they only work well in certain very specific tasks, and we are quite clueless regarding how to design, say, a roach.
games have rules. Once you've learned the rules, you're unstoppable
Ah, but I want both to play the same game, while your are suggesting giving the machine a special representation with a high-level vocabulary. The interface I am proposing is the same one you are using: images and sounds come out and commands are executed. Not "high-level game data" - only images and sounds.
You talk about teaching the AI how to headshot. I am talking about the difficulty of processing a 2D image and interpreting it as a PoV rendering of an (unknown) 3D model, and locating a set of pixels that happen to compose the 'head' of something that is an 'enemy'.
Tell you what: give me $50,000 in funding, six months to train the AI to general FPS rules (headshot, movement, general weapon effectiveness, etc.), and another six months for the GA to advance it for CS, and it will beat anything.
If I give you 50k $, will you sign a document agreeing to deliver accurate 3D model-building from a (sythetic) video in real-time or give me my money back plus interest rates?. That's what I was referring to when I wrote about the DARPA Grand Challenge. You and me can both pilot a remote-control car with a wireless lo-res camera slapped on top. Try to teach a computer to do that reliably, and you have solved many, many problems in AI.
You are overly optimistic regarding GAs. I very much doubt that, whatever the AI cycles, you can evolve an AI team capable of winning against a good (as in top-third of the table) human team at, say, counter-strike.
Gotcha: the AIs would not be allowed to cheat, and would have the exact same information at their disposal as the human team: a lot of visual input (but not the actual noise-free game geometry) and the same set of commands as human players.
If you can evolve that kind of AI, the DARPA Grand Challenge is looking for you.
Depends on the setting. It makes perfect sense within a networked-battlefield scenario: what one unit sees, it will try to convey to others, and commanders can take decisions based on everything that is seen by any of their units.
In a medieval setting, they would have to shout to each other (and be within hearing distance) to request assistance. Or wave colored flags, or send messenger pigeons.
The radio squawking in Half-Life added an element of realism to this - you could actually "hear" what the bad guys were "saying" to each other. Even though it was crude: it would have been even better to hear things like "he's hiding behind the wall" or "he's nailed Bob! the #@!"
Additionally, a multilevel AI helps to even out the use of CPU. You don't want to take high-level command decisions every quarter second, because that would not leave enough time for any of the lower tiers to do anything useful, and would require too much CPU. With a multi-tiered AI, you can dedicate small and frequent AI slots to low-level decisions, and hold high-level decisions for less frequent and more CPU-intensive thinking.
C hides stacks from you. I don't know why you have to teach them to anyone.
You don't have to teach it, until people start complaining that they get weird errors after overflowing a stack-allocated array. Knowing what the stack is and how it is used really helps then.
And Java and C++ don't eliminate stacks.
I never said that Java and C++ did not use them. But overwriting the stack with garbage is much harder under Java and other memory-managed languages. Which C++ is not, even if you can overload 'new'.
In theory, you don't need to teach pointers as arrays of memory locations. You can show them as references with different behaviors and rules for correct use.
You can teach pointer semantics out of thin air, but I think it makes more sense if you tell people where that comes from. If you teach C, you may as say why it was designed the way it was designed - and why that makes C programs prone to fail in interesting ways.
Try to teach pointers to someone who is not familiar with the concept of memory as a large, one-dimensional string of bytes - where a byte can contain a part of an sizeof(int)-byte integer, or a single character, or a part of a sizeof(void *)-byte memory address to yet other interesting bytes.
Try to explain stack corruption to someone who does not understand the concept of pushing things into the stack when you call a function and popping them back out when you return, and using that same stack space to store your local variables.
If you can't get more out of C than its more convoluted syntax, you never learned C properly.
I think he is just mad with people getting away with "oh, yes - there's a bug somewhere, must have crept in while I was not looking". Dijkstra had a quite extreme view on programmer errors.
Written in '88, it seems to me to have aged quite well. And for all the intellectual smugness, there a few tidbits worth keeping.
The 9 orders of magnitude that have to be spanned when thinking on computation are quite true. The gap may have expanded with even faster computers and more ambitious projects. And the fact that thinking discrete is very different from thinking continuous is also well made. Knowing what is difficult is a good protection against overconfidence. One of the hard lessons of computers is that what you think it does and what it really does are not always in sync.
On the educational aspect, dumbing things down, if done only for the heck of it, does result in dumber students. If all you meet are small, uniform, incremental steps -- how can you expect to learn to leap?.
Yes, hammering only theory and proof-of-correctness to freshmen is indeed cruel and, furthermore, ineffective (except for learning algorithmic thought). As other posters have said, business requirements are not amenable to formal proofs. Ivory-towerness aside, I think it makes for a provocative read. And if profs are not allowed to shout from the tower once in a while, they would go mad.