Not really, you can just tokenize the code, removing the issue of white space, comments and formatting completely, and just treat identifiers as a token type vs even looking at the name - then you're just doing a pure structural comparison of the code. Plenty of other things you could do too to normalize the code and factor out any trivial rewritings (changing parenthesis, etc).
Any smart code comparison would look for function by function matches to factor out trivial reordering, and would anyways operate on a diff-like basic looking for changes rather than a yes/no match.
Net result is that a smart code compare tool would not be fooled in the slightest if you completely reformatted the code, changed all the identifiers and comments, reordered all the functions, and even rewrote a few pieces for good measure - it'd still show a 95% match with the original, which for anything more than a trivial fucntion would be almost certain proof that id had been copied.
The system described isn't quite the same... the innovation is that they are using 2-D torque sensors on the whiskers, and apparently are able to reconstruct the 3-D surface of the object whiskered using the data... that's a rather major improvement over just detecting a profile with on/off switches!;-)
AFAIK the Soyuz cost is ~$5M per mission, so thay actually make a decent profit on this, which is the reason they are doing it.
Given how useless the ISS is for research (too undermanned - they're basically janitors, not scientists), the Russians are not losing much by occasionally selling one of their spots to a tourist.
Soon you'll be able to sit at your desk in the USA, doing your job at your current salary, even though it's been "off shored"..... and the boss will still be patting himself on the back for the savings he thinks he must be making.
True, and quite profound, but of course the US isn't a real democracy (it's some type of representative republic) nor really pretends to be one. It's a carefully crafted political system that gives the appearance of being a democracy by holding elections, while making sure via the electoral college system that no real power is given to the voters. By putting all the voting power in the hand of the few swing states, it allows the members of the oligarchy to play the game by their own rules. This was in essence, if not in spirit, actually the intent of the founding fathers - the reason the US doesn't have a real democracy is because it was beleived that the public were too uninformed to vote in the best interests of the the country (nominally themselves, but in reality the power brokers), and that the electoral college members could do a better job. The disenfranchisement of the American voter would have been less if had been left at that, but in order to better manipulate the system even the electoral votes arn't allocated democratically - each state is forced to go all left or all right, thereby COMPLETELY removing the possibility of most voters (i.e. all those not living in the handful of swing states) from having ANY influence on which party gets elected in presidential elections.
So, yeah, a democracy would be nice, and a democracy with popularly decided laws even better, but it sure isn't America.
I think you have to differentiate mass distribution of copyrighted stuff via P2P networks from simply copying/sharing with friends. There was never any real issue that I'm aware of with people taping LP's or copying music cassettes for friends before the age of digital music - we never saw the RIAA trying to crack down on dual tape decks as being an illegal copying device!
Of course digital media and the internet does change the issue of copying, but in a sane world that shouldn't result in preexisting "rights" (even if only by precedent) being taken away. By all means the RIAA should prevent people from sharing via P2P, but at the same time IMO it should not be going after people who simply let their friend tape or copy a CD.
Bad article summary...
on
The Expert Mind
·
· Score: 2, Insightful
The summary of this article doesn't really convey the content - it's not really about nature vs nurture or how long it takes to train to become an expert...
The real article content is that the expert mind works differently (i.e. uses different brain functions to achieve a better result) from the novice one. Chess is used as an example because it's easy via ratings to objectively measure expertise in this area.
In a nutshell, a novice in a field has to use general (new) problem solving skills to figure out what to do, but the expert, from years of focused experience, instead uses memory recall (not problem solving) of domain-specific chunked memories to determine the best course of action.
This result is proven for chess by brain scans of novice and expert chess players in action showing which areas of the brain are active, as well as by showing that experts perfrom better at memorizing real rather than random chess positions, while novices perform muich the same (poorly) in either case; the inference of the memorization task is that experts are able to chunk real positions into pre-learnt patterns, and therefore have less to remember, but for random positions (which therefore don't occur in their learnt patterns) they have to resort to piece-by-piece memorization like the novice.
The article quotes Casablanca being questioned on how many moves he plans ahead, and answering "one - the right one!". This isn't bragging, but rather reflects the reality of seeing (via automatic memory recall) the right position rather than having to work it out via a computer-like game alogorithm.
Apparently the limitation isn't the camera - it's the Java API to the camera which limits image resolution, as does the amount of memory available on some phones. Apparently others have the same thing in a C++/Symbian environment, but this guy says his is the only one to work on a resource limited Java phones.
Well, from the video, this isn't a product barcode - it's arbitrary data that's encoded. One demo app is using it to encode URL's so that you scan the "barcode" and the web page comes up on your cell phone's browser. Another thing people did with the free version was to tag real world items with barcodes containing WikiPedia references. Apparently Quest used it for a competition where people had to find and scan barcodes printed on Quest advertizements around town.
Well, considing that the guy is specifically a mineralogist, and you're apparently completely baffled as to how one might test it, I'll give him the nod. Hardness (Moh amber=2-2.5, glass=5-6) or refraction index (amber=1.517, glass=1.546) would be two ways for starters. For that matter, even the average high school science student would know that you could just rub it - amber becomes statically charged and will then attract small bits of paper.
Yes, Americans love violence, but I think the American penchant for gun ownership (viz: NRA nuts) is largely unrelated - it's because it's constitutionally protected, and therefore "an American right".
Generally something like accessing an uninitialized pointer, or freed memory, will take down the whole process. If the application is designed such that threads are independent, and it's meaningful to carry on in the event of one thread dying, then I don't see why you *couldn't* install a trap (SEGV) handler that would kill the current thread rather than exit the process, but this isn't usually done - these types of error are usually regarded as catastrophic and non-recoverable, so you exit.
The problem with IE as "baseline browser" is that it defines a very weak lowest common denominator for web developers to target (notably incomplete and buggy CSS support), and thus holds back the web as a whole. It doesn't help that more standards-compliant and feature rich browsers are available since web pages themselves can't use those standards if they also want to work with IE.
Of course this is the way Microsoft wants it - they want to control the Internet rather than have it go racing ahead out of their control. Lack of a feature rich lowest common denominator hurts Microsoft's web-savvy competitors (Google, Yahoo, etc), more than it hurts Microsoft itself, since Microsoft are still playing catch-up in the web arena.
Decent idea. I guess that's what I'll do if I want to use it and they havn't changed it. I guess it remains to be seen if there's any downside to doing that in terms of ease of use (e.g. e-mails related to purchases)... it doesn't seem to be what they intended.
Yeah - I don't like that my gmail password would now give someone access to my credit card if I were to sign up for Google checkout. I like to keep things more compartmentalized than that. For things like e-mail and other lower security things I use one set of passwords, but for PayPal I use a unique, much longer, and more secure one and make sure never to have my Browser store it.
Also, it's convenient to stay logged into Google for gmail, but I wouldn't want to do that at work if it gave access to my credit card! I think a seperate password, required each time you buy something, would be better than using your one password to the Googleplex.
First Microsoft misses the whole internet thing, at least twice (browsers, search), now they're getting on the robotics bandwagon 20 years late. Ho hum.
There's nothing wrong with using goto where there's no efficient alternative. What's bad is using goto for normal control flow (aka sphagetti code), but it's reasonable for example to use it for "exception handlng" in a laguage such as C that doesn't have the try.. catch mechanism of C++ (where throw() basically means "goto catch"). Another reasonable place, just as an example, to use goto is in implementation of a state machine where the states are labels in your code and you can then write "goto state_x" rather than the less efficient "state = state_x" and loop back to a "switch (state)" statement effectively that does the "goto" (but less efficiently via jump table or decision tree, depending on how sparse your case labels are). Of course you don't always care about this level of efficiency, but sometimes you do.
The "goto considered harmfull" is basically true, and makes a reasonable edict for beginners, but goto still has it's place (especially in C for exception processing).
Well, above the rendering level you have the physics of the objects being rendered, which is only just starting to see specialized hardware (physics accelerators) thrown at it, so plently of chance to increase performance by using the main CPUs for that, and even above that I'd have thought that the movement of various self-propelled objects (people, etc) are independent and could be parallelized, and other tasks such as collision detection can also be divided up between threads on basis of objects involved.
Well, then show me how to insert a parallel architecture in your typical 3D game
Base it on OpenGL or Direct3D, and let the massive parallism that's inherent to the task of 3D rendering be handled in parallel by the multiple shaders/etc of your graphics card.
Slashdot Mirror
Not really, you can just tokenize the code, removing the issue of white space, comments and formatting completely, and just treat identifiers as a token type vs even looking at the name - then you're just doing a pure structural comparison of the code. Plenty of other things you could do too to normalize the code and factor out any trivial rewritings (changing parenthesis, etc).
Any smart code comparison would look for function by function matches to factor out trivial reordering, and would anyways operate on a diff-like basic looking for changes rather than a yes/no match.
Net result is that a smart code compare tool would not be fooled in the slightest if you completely reformatted the code, changed all the identifiers and comments, reordered all the functions, and even rewrote a few pieces for good measure - it'd still show a 95% match with the original, which for anything more than a trivial fucntion would be almost certain proof that id had been copied.
Ben
The system described isn't quite the same... the innovation is that they are using 2-D torque sensors on the whiskers, and apparently are able to reconstruct the 3-D surface of the object whiskered using the data... that's a rather major improvement over just detecting a profile with on/off switches! ;-)
If you want a decent and *ahem* modern programmer's editor for windows, then try GWD edit :
http://www.gwdsoft.com/
Free 30 day trial then $29.99 shareware. Well worth the money.
AFAIK the Soyuz cost is ~$5M per mission, so thay actually make a decent profit on this, which is the reason they are doing it.
Given how useless the ISS is for research (too undermanned - they're basically janitors, not scientists), the Russians are not losing much by occasionally selling one of their spots to a tourist.
Soon you'll be able to sit at your desk in the USA, doing your job at your current salary, even though it's been .. and the boss will still be patting himself on the back for the savings he thinks he must be making.
"off shored"...
It looks like you're taking a dump! Would your like some toilet paper for that?
If you're incapable of depression, and you're always happy, how do you know if you really are happy?
When you've got a boner.
True, and quite profound, but of course the US isn't a real democracy (it's some type of representative republic) nor really pretends to be one. It's a carefully crafted political system that gives the appearance of being a democracy by holding elections, while making sure via the electoral college system that no real power is given to the voters. By putting all the voting power in the hand of the few swing states, it allows the members of the oligarchy to play the game by their own rules. This was in essence, if not in spirit, actually the intent of the founding fathers - the reason the US doesn't have a real democracy is because it was beleived that the public were too uninformed to vote in the best interests of the the country (nominally themselves, but in reality the power brokers), and that the electoral college members could do a better job. The disenfranchisement of the American voter would have been less if had been left at that, but in order to better manipulate the system even the electoral votes arn't allocated democratically - each state is forced to go all left or all right, thereby COMPLETELY removing the possibility of most voters (i.e. all those not living in the handful of swing states) from having ANY influence on which party gets elected in presidential elections.
So, yeah, a democracy would be nice, and a democracy with popularly decided laws even better, but it sure isn't America.
I think you have to differentiate mass distribution of copyrighted stuff via P2P networks from simply copying/sharing with friends. There was never any real issue that I'm aware of with people taping LP's or copying music cassettes for friends before the age of digital music - we never saw the RIAA trying to crack down on dual tape decks as being an illegal copying device!
Of course digital media and the internet does change the issue of copying, but in a sane world that shouldn't result in preexisting "rights" (even if only by precedent) being taken away. By all means the RIAA should prevent people from sharing via P2P, but at the same time IMO it should not be going after people who simply let their friend tape or copy a CD.
The summary of this article doesn't really convey the content - it's not really about nature vs nurture or how long it takes to train to become an expert...
The real article content is that the expert mind works differently (i.e. uses different brain functions to achieve a better result) from the novice one. Chess is used as an example because it's easy via ratings to objectively measure expertise in this area.
In a nutshell, a novice in a field has to use general (new) problem solving skills to figure out what to do, but the expert, from years of focused experience, instead uses memory recall (not problem solving) of domain-specific chunked memories to determine the best course of action.
This result is proven for chess by brain scans of novice and expert chess players in action showing which areas of the brain are active, as well as by showing that experts perfrom better at memorizing real rather than random chess positions, while novices perform muich the same (poorly) in either case; the inference of the memorization task is that experts are able to chunk real positions into pre-learnt patterns, and therefore have less to remember, but for random positions (which therefore don't occur in their learnt patterns) they have to resort to piece-by-piece memorization like the novice.
The article quotes Casablanca being questioned on how many moves he plans ahead, and answering "one - the right one!". This isn't bragging, but rather reflects the reality of seeing (via automatic memory recall) the right position rather than having to work it out via a computer-like game alogorithm.
Apparently the limitation isn't the camera - it's the Java API to the camera which limits image resolution, as does the amount of memory available on some phones. Apparently others have the same thing in a C++/Symbian environment, but this guy says his is the only one to work on a resource limited Java phones.
Well, from the video, this isn't a product barcode - it's arbitrary data that's encoded. One demo app is using it to encode URL's so that you scan the "barcode" and the web page comes up on your cell phone's browser. Another thing people did with the free version was to tag real world items with barcodes containing WikiPedia references. Apparently Quest used it for a competition where people had to find and scan barcodes printed on Quest advertizements around town.
Actually that's a pretty damned impressive web server.
I'm downloading the video on a cable modem at a sustained 1.5MB/sec (bytes, not bits).
It seems unslashstoppable.
Yeah - I remember trying this as a schoolkid wearing a black school uniform. Didn't make the stuff thick enough...
.... and got covered in sticky cornstarch. :-(
Poured a big pool of it on the kitchen table and smacked my hand down on it hard
Well, considing that the guy is specifically a mineralogist, and you're apparently completely baffled as to how one might test it, I'll give him the nod. Hardness (Moh amber=2-2.5, glass=5-6) or refraction index (amber=1.517, glass=1.546) would be two ways for starters. For that matter, even the average high school science student would know that you could just rub it - amber becomes statically charged and will then attract small bits of paper.
Yes, Americans love violence, but I think the American penchant for gun ownership (viz: NRA nuts) is largely unrelated - it's because it's constitutionally protected, and therefore "an American right".
Generally something like accessing an uninitialized pointer, or freed memory, will take down the whole process. If the application is designed such that threads are independent, and it's meaningful to carry on in the event of one thread dying, then I don't see why you *couldn't* install a trap (SEGV) handler that would kill the current thread rather than exit the process, but this isn't usually done - these types of error are usually regarded as catastrophic and non-recoverable, so you exit.
The problem with IE as "baseline browser" is that it defines a very weak lowest common denominator for web developers to target (notably incomplete and buggy CSS support), and thus holds back the web as a whole. It doesn't help that more standards-compliant and feature rich browsers are available since web pages themselves can't use those standards if they also want to work with IE.
Of course this is the way Microsoft wants it - they want to control the Internet rather than have it go racing ahead out of their control. Lack of a feature rich lowest common denominator hurts Microsoft's web-savvy competitors (Google, Yahoo, etc), more than it hurts Microsoft itself, since Microsoft are still playing catch-up in the web arena.
Decent idea. I guess that's what I'll do if I want to use it and they havn't changed it. I guess it remains to be seen if there's any downside to doing that in terms of ease of use (e.g. e-mails related to purchases)... it doesn't seem to be what they intended.
Yeah - I don't like that my gmail password would now give someone access to my credit card if I were to sign up for Google checkout. I like to keep things more compartmentalized than that. For things like e-mail and other lower security things I use one set of passwords, but for PayPal I use a unique, much longer, and more secure one and make sure never to have my Browser store it.
Also, it's convenient to stay logged into Google for gmail, but I wouldn't want to do that at work if it gave access to my credit card! I think a seperate password, required each time you buy something, would be better than using your one password to the Googleplex.
http://world.honda.com/ASIMO/history/
Overloading is just syntactic sugar - it doesn't give you any fucntionality.
There's no functional difference between using an overloaded name f(a), f(x, y), f(p, q, r) or three separate ones f_a(a), f_xy(x, y), f_pqr(p, q, r).
If you want default arguments C has them, and if you want polymorphism then C has it too (function pointers).
There's nothing wrong with using goto where there's no efficient alternative. What's bad is using goto for normal control flow (aka sphagetti code), but it's reasonable for example to use it for "exception handlng" in a laguage such as C that doesn't have the try.. catch mechanism of C++ (where throw() basically means "goto catch"). Another reasonable place, just as an example, to use goto is in implementation of a state machine where the states are labels in your code and you can then write "goto state_x" rather than the less efficient "state = state_x" and loop back to a "switch (state)" statement effectively that does the "goto" (but less efficiently via jump table or decision tree, depending on how sparse your case labels are). Of course you don't always care about this level of efficiency, but sometimes you do.
The "goto considered harmfull" is basically true, and makes a reasonable edict for beginners, but goto still has it's place (especially in C for exception processing).
Well, above the rendering level you have the physics of the objects being rendered, which is only just starting to see specialized hardware (physics accelerators) thrown at it, so plently of chance to increase performance by using the main CPUs for that, and even above that I'd have thought that the movement of various self-propelled objects (people, etc) are independent and could be parallelized, and other tasks such as collision detection can also be divided up between threads on basis of objects involved.
Well, then show me how to insert a parallel architecture in your typical 3D game
Base it on OpenGL or Direct3D, and let the massive parallism that's inherent to the task of 3D rendering be handled in parallel by the multiple shaders/etc of your graphics card.