To reinforce this: Building your own engine is a great way to fall into a trap that you won't escape for years if not decades. If you're interested in building a game, then you should build a game, and use the toolchain that lets you get to your game as quickly as possible. Right now that's a race between Unity and Unreal Engine 4, and while neither one has perfect support, I'd say the userbase for Unity is sufficiently deep that it's a better starting point. Don't worry, you'll still get plenty of opportunity to code (and to learn C# - another reason to go with Unity over UE4, if that's the language you really want), but there'll be enough that you don't have to code that you can focus on the fundamentals of building your game.
My reading of Norvig's blog post is that he suggests his specific approach (stapling together short regexps with ORs) requires solving the NP-Complete Set Cover problem, but he doesn't actually say anything about whether the core problem (match everything in list A and nothing in list B) does; it's conceivable that e.g. some sort of dynamical programming approach could do the job more efficiently than Norvig's algorithm does. Does anyone know whether the root problem (to avoid having to do the optimization, just phrase it as 'is there a separating regexp for the sets A and B of length less than k?') is specifically known to be NP-complete?
With the exception of entertainment and the rare 'culturally relevant' application, the vast majority of software is primarily a tool to get its job done, rather than an item of artistic merit in its own right. The New York Times reviews are — for the most part — cultural reviews; they're not the appropriate venue for most software reviews.
With that said, there are those exceptions where one can speak about the artistic or cultural merits of a piece of software, and my strong impression is that the Times has never really stopped speaking about those. The difference between the '80s and today is that at that point, there was so much less understood and so much more that was new in the world of software that a lot of what came out was of cultural relevance and worth talking about on those merits.
I'm not going to argue with 'severely over budget' (thank you, Prop 13) but the last information available for California suggests that federal spending in the state was substantially less (by roughly 25%) than federal revenues from the state; California is, on a per-capita basis and certainly on an overall basis, one of the largest net givers to the federal budget, not a taker. Do you have any specific reason to believe that that's changed in the last few years?
Not just mammalian neurons, but invertebrate neurons too. I think that until we surpass what MomNature has already bioengeineered and abandoning the VonNeumann/Turing model of how a computer is "supposed to be" that we will not construct anything AI that is more performant than what already exists in biological systems.
Almost every day I move around in a vehicle that's faster on land than anything 'Mom Nature' has produced. Several times a year I fly through the air in one that's almost an order of magnitude quicker. We took one of Nature's apex predators, carefully crafted through millions of years of evolution, and in maybe _one one-thousandth_ the time we turned it into the Pomeranian. I'm not sure why you believe we're so far behind nature, or why 'artificial' approaches are so doomed to failure compared to a natural simulationist approach, when we have overwhelming evidence to suggest the opposite.
While not often directly mathematical, several of Jorge Luis Borges's short stories are interesting efforts on his part to grapple philosophically with many of the concepts of infinity: The Library of Babel most famously, but also great stories like The Book of Sand, The Aleph, and even Death and the Compass. They won't necessarily tickle you in the same way that Stephenson's work did, but they're still a fine jumping-off point into fascinating and deeply philosophical mathematics.
Everyone else on the net seems to point to the article in the NY Post (not exactly known for its careful fact-checking) and the Post article talks about Hulu 'taking its first steps' without a single mention of what those steps are. No other news stories I can find in the last several days talk about any changes occurring to Hulu's model (other than more original programming) or the Hulu user experience. So what the hell is the Post talking about, exactly? What evidence is there — beyond some editorial negative-wishcasting — that anything like this is going on?
Compared to Windows as of 15 years ago, maybe. The Windows APIs the last few years have been mature enough that while diverse hardware testing can still improve the user experience, it's gotten substantially less necessary for game developers. That just isn't the case for Android games.
I already have half a dozen different devices in my house that I can compile on, most of which have better storage and faster CPUs. Why do I care if I have another one? My iPad isn't trying to replace one of my computers and it seems silly to judge it by the same standards.
TFA seems to be long on speculation and short on actual data. Obviously streaming video isn't bandwidth-cheap, but does anyone have real figures on how much data streaming, say, a standard 24-minute TV show would take, and how many episodes it would take to hit the 2GB monthly cap? If they can, for instance, stream a low-quality episode in 10-20 MB then this seems like much sound and fury over very little...
Isn't this effectively the core purpose of copyright law? A hundred years of precedent suggest that my free speech rights don't extend to, for instance, performing my own stage production of Spiderman for all the world to see, or for writing and selling (or giving away) my word-for-word version of "Arguing With Idiots", and I'm not sure why anyone would expect results in the digital world to be any different.
The fascinating thing there is how Wizards "tricked itself" by misreading how certain cards form gamebreaker combos. So then they embarked on an elaborate "currency value adjustment" program, aka Type 2. (With all the spinoffs etc. In my areas "1.5" and "Legacy" and so on were never very popular.)
By being relegated to "Type 1" All those power cards were effectively cordoned off into a backwater, and lost most of their effective value. Then as the years rolled on, once cards left Type 2, they also dropped in value like a stone.
Well, except that this didn't happen. Yes, a number of cards definitely lost value once they fell out of Type 2 - but the price of the core type-1 power cards has never actually gone down, and in fact Legacy has meant that a number of secondary cards from that era have now skyrocketed. A white-border Black Lotus will set you back more than a thousand dollars; a black-border one you'd be lucky to find under two thousand. All the dual lands are north of $50 in white-border now and more than $200 in black border; half-blue duals are at least $5-600 each in black border. Some of the cards from the earlier sets (esp. Arabian Nights) have seen corrections, but that's more a matter of the market realigning itself around playability rather than just rarity - old out-of-print cards that see any tournament play at all (or even saw tournament play at one point) have skyrocketed (Karakas at $50-60, Sylvan Library at $25, etc.)
'Pretty much he can't talk or think about the PS3 for some time.'
I wasn't aware that the only things one could say about the PS3 were related to cracking its protection schemes and pirate! (or, okay, 'traffic in copyrighted works'). He can talk about the games, he can talk about the OS, he can talk about the hardware, he just can't, y'know, talk about how to circumvent any of it. Really, this seems like a relatively reasonable restraining order all around, at least by the metrics for such things; it stops the specific (alleged) infringing behavior and doesn't strike much more broadly than that.
Perhaps the reason the court didn't overturn the encryption restrictions is because the defendant didn't challenge those restrictions? The judgement in the linked-to article seems relatively clear (even if by omission) that the only restrictions challenged were the three restrictions (to use their lettering and wording) (A) on 'Use of Computer for Non-School-Related Purposes', (B) on 'Use of Instant Messaging or Social Networks', and (C) on 'Use of Computers Contaminated with Viruses or Unwanted Software'. If the defendant didn't request to have the restrictions on encryption (which are certainly there so that the juvenile justice system can track his communications) overturned and made no request for the total overthrow of his probation conditions, then I'm not sure the court even has standing to unilaterally throw out the encryption provision, and certainly it's little surprise that they wouldn't do so without being explicitly asked.
God, the 99/4a was a horrible system. Not bad as a games machine, but the worst possible machine to get a budding young coder (one of the most locked-down systems I've ever seen; even PEEK only came in extended basic and that still didn't get you POKE, and doing fast graphics was all but impossible unless you were an expert assembly coder and sprung the $100ish for their assembler). I still haven't forgiven my parents for that.
Parsec was also flat-out excellent, and it had one of the best versions of Miner 2049er out there, but that appeared on so many systems that it could hardly be called a hidden gem.
So far as I know, Knuth has done essentially zero work related to the P/NP question; a lot of algorithmics and tons of fantastic work in combinatorics, but I can't think of a single significant result he's contributed to complexity theory. While it's not impossible that he could have some sort of 'outsider breakthrough', it seems almost infinitesimally unlikely given the mathematical context and techniques that have had to be developed for similar complexity problems. My money would be on either a formal open-sourcing of the TeX codebase or the development of a full HTML5 rendering engine for TeX along the lines of the system that mathoverflow.net uses.
The reason why PageRank 'has to be' linear is essentially mathematical; treating importance as a linear function of the importance of inbound links means that the core equations that need to be solved to determine importance are linear and the answer can be found with (essentially) one huge matrix inversion. If you make importance nonlinear then the equations being solved become computationally infeasible.
What's interesting to me is how close the connections are between PageRank and the classic light transfer/heat transfer equations that come up in computer graphics' radiosity (see James Kajiya's Rendering equation); I wonder if there's a reasonable equivalent of 'path tracing' (link tracing?) for computing page ranks that avoids the massive matrix inversions of the basic PageRank algorithm.
While it's unsurprising given that the current Unreal Engine is still in active development and a ton of commercial games are still being developed and shipped using it, it's worth pointing out that this isn't a source code release; instead, it's something much closer to an elaborate mod engine, with generous swaths of behavioral scripting but no real ability to get 'under the hood' as it were. Still, kudos to Epic for this; it'll be interesting to see who picks up the ball and runs with this.
Especially in a case like this, where there are other tightly-correlated variables. Why is the authors' presumption that it's the cosmic rays (or lack thereof) that are regulating tree growth, rather than solar and sunspot activity itself? It seems at least as plausible to me that sunspot activity correlates to some other solar features (e.g., solar irradiance) that would have a more natural and direct effect on tree growth than cosmic rays.
Strictly speaking, there aren't any seriously new methods of multiplying numbers here; even the techniques they use for handling multiplicands larger than the computer's memory (sectional FFTs, using the Chinese Remainder Theorem to solve the problem by reducing modulo a lot of small primes) are pretty well-established from things like computations of pi, with this group offering a few improvements to the core ideas. What they did provide, and what sounds particularly promising, is a library (judging from the article, likely even open-source) for handling bignums like this that they've made available for general use. It'll be interesting to see if anyone else picks up this ball and runs with it.
...depending on your moral stance; the company (which unfortunately owns a host of major computer book publishers, most notably Academic Press, Digital Press and Morgan Kaufmann) has had a small host of scandals, mostly concerning exorbitant journal fees and 'sponsored' pharmaceutical journals (they were the publisher behind the Merck Australasian Journal of Bone and Joint Medicine scandal, if you recall that). MK and AP publish some of the finest books in the industry, which makes this that much harder a moral stand to take, but it's worth evaluating how you feel about the publisher before you consider going down that route.
Despite waning console sales, orders for PS3-related hardware have risen sharply.
'related' is an odd word to see there; what 'PS3-related' hardware would this be, and where's the evidence for this random assertion? Is this meant to imply that Sony's suppliers are getting more orders for some of the hardware used to build the PS3 (which could mean more PS3s in the pipeline but could also be wholly unrelated), or just a consolist claim that Sony is doing better than its sales figures would indicate?
...that Google licenses this to scanner manufacturers and we see this at a consumer level at some point in the future? I know I'd pay good money for a book scanner that doesn't need to have a 'book edge' (which you already have to pay through the nose for)...
Slashdot Mirror
To reinforce this: Building your own engine is a great way to fall into a trap that you won't escape for years if not decades. If you're interested in building a game, then you should build a game, and use the toolchain that lets you get to your game as quickly as possible. Right now that's a race between Unity and Unreal Engine 4, and while neither one has perfect support, I'd say the userbase for Unity is sufficiently deep that it's a better starting point. Don't worry, you'll still get plenty of opportunity to code (and to learn C# - another reason to go with Unity over UE4, if that's the language you really want), but there'll be enough that you don't have to code that you can focus on the fundamentals of building your game.
My reading of Norvig's blog post is that he suggests his specific approach (stapling together short regexps with ORs) requires solving the NP-Complete Set Cover problem, but he doesn't actually say anything about whether the core problem (match everything in list A and nothing in list B) does; it's conceivable that e.g. some sort of dynamical programming approach could do the job more efficiently than Norvig's algorithm does. Does anyone know whether the root problem (to avoid having to do the optimization, just phrase it as 'is there a separating regexp for the sets A and B of length less than k?') is specifically known to be NP-complete?
With the exception of entertainment and the rare 'culturally relevant' application, the vast majority of software is primarily a tool to get its job done, rather than an item of artistic merit in its own right. The New York Times reviews are — for the most part — cultural reviews; they're not the appropriate venue for most software reviews.
With that said, there are those exceptions where one can speak about the artistic or cultural merits of a piece of software, and my strong impression is that the Times has never really stopped speaking about those. The difference between the '80s and today is that at that point, there was so much less understood and so much more that was new in the world of software that a lot of what came out was of cultural relevance and worth talking about on those merits.
I'm not going to argue with 'severely over budget' (thank you, Prop 13) but the last information available for California suggests that federal spending in the state was substantially less (by roughly 25%) than federal revenues from the state; California is, on a per-capita basis and certainly on an overall basis, one of the largest net givers to the federal budget, not a taker. Do you have any specific reason to believe that that's changed in the last few years?
Almost every day I move around in a vehicle that's faster on land than anything 'Mom Nature' has produced. Several times a year I fly through the air in one that's almost an order of magnitude quicker. We took one of Nature's apex predators, carefully crafted through millions of years of evolution, and in maybe _one one-thousandth_ the time we turned it into the Pomeranian. I'm not sure why you believe we're so far behind nature, or why 'artificial' approaches are so doomed to failure compared to a natural simulationist approach, when we have overwhelming evidence to suggest the opposite.
While not often directly mathematical, several of Jorge Luis Borges's short stories are interesting efforts on his part to grapple philosophically with many of the concepts of infinity: The Library of Babel most famously, but also great stories like The Book of Sand, The Aleph, and even Death and the Compass. They won't necessarily tickle you in the same way that Stephenson's work did, but they're still a fine jumping-off point into fascinating and deeply philosophical mathematics.
Everyone else on the net seems to point to the article in the NY Post (not exactly known for its careful fact-checking) and the Post article talks about Hulu 'taking its first steps' without a single mention of what those steps are. No other news stories I can find in the last several days talk about any changes occurring to Hulu's model (other than more original programming) or the Hulu user experience. So what the hell is the Post talking about, exactly? What evidence is there — beyond some editorial negative-wishcasting — that anything like this is going on?
Compared to Windows as of 15 years ago, maybe. The Windows APIs the last few years have been mature enough that while diverse hardware testing can still improve the user experience, it's gotten substantially less necessary for game developers. That just isn't the case for Android games.
I already have half a dozen different devices in my house that I can compile on, most of which have better storage and faster CPUs. Why do I care if I have another one? My iPad isn't trying to replace one of my computers and it seems silly to judge it by the same standards.
TFA seems to be long on speculation and short on actual data. Obviously streaming video isn't bandwidth-cheap, but does anyone have real figures on how much data streaming, say, a standard 24-minute TV show would take, and how many episodes it would take to hit the 2GB monthly cap? If they can, for instance, stream a low-quality episode in 10-20 MB then this seems like much sound and fury over very little...
Isn't this effectively the core purpose of copyright law? A hundred years of precedent suggest that my free speech rights don't extend to, for instance, performing my own stage production of Spiderman for all the world to see, or for writing and selling (or giving away) my word-for-word version of "Arguing With Idiots", and I'm not sure why anyone would expect results in the digital world to be any different.
Strange, then, that the 'bitcoin isn't worth the energy it's minted on!' article isn't the one that made Slashdot headlines...
The fascinating thing there is how Wizards "tricked itself" by misreading how certain cards form gamebreaker combos. So then they embarked on an elaborate "currency value adjustment" program, aka Type 2. (With all the spinoffs etc. In my areas "1.5" and "Legacy" and so on were never very popular.)
By being relegated to "Type 1" All those power cards were effectively cordoned off into a backwater, and lost most of their effective value. Then as the years rolled on, once cards left Type 2, they also dropped in value like a stone.
Well, except that this didn't happen. Yes, a number of cards definitely lost value once they fell out of Type 2 - but the price of the core type-1 power cards has never actually gone down, and in fact Legacy has meant that a number of secondary cards from that era have now skyrocketed. A white-border Black Lotus will set you back more than a thousand dollars; a black-border one you'd be lucky to find under two thousand. All the dual lands are north of $50 in white-border now and more than $200 in black border; half-blue duals are at least $5-600 each in black border. Some of the cards from the earlier sets (esp. Arabian Nights) have seen corrections, but that's more a matter of the market realigning itself around playability rather than just rarity - old out-of-print cards that see any tournament play at all (or even saw tournament play at one point) have skyrocketed (Karakas at $50-60, Sylvan Library at $25, etc.)
Apple stopped being a software company when the iPhone because the hottest ticket in the geek market.
Why do you think the iPhone was/is the hottest thing going? Hint: it's not the hardware, and it's sure not the carrier...
'Pretty much he can't talk or think about the PS3 for some time.'
I wasn't aware that the only things one could say about the PS3 were related to cracking its protection schemes and pirate! (or, okay, 'traffic in copyrighted works'). He can talk about the games, he can talk about the OS, he can talk about the hardware, he just can't, y'know, talk about how to circumvent any of it. Really, this seems like a relatively reasonable restraining order all around, at least by the metrics for such things; it stops the specific (alleged) infringing behavior and doesn't strike much more broadly than that.
Perhaps the reason the court didn't overturn the encryption restrictions is because the defendant didn't challenge those restrictions? The judgement in the linked-to article seems relatively clear (even if by omission) that the only restrictions challenged were the three restrictions (to use their lettering and wording) (A) on 'Use of Computer for Non-School-Related Purposes', (B) on 'Use of Instant Messaging or Social Networks', and (C) on 'Use of Computers Contaminated with Viruses or Unwanted Software'. If the defendant didn't request to have the restrictions on encryption (which are certainly there so that the juvenile justice system can track his communications) overturned and made no request for the total overthrow of his probation conditions, then I'm not sure the court even has standing to unilaterally throw out the encryption provision, and certainly it's little surprise that they wouldn't do so without being explicitly asked.
God, the 99/4a was a horrible system. Not bad as a games machine, but the worst possible machine to get a budding young coder (one of the most locked-down systems I've ever seen; even PEEK only came in extended basic and that still didn't get you POKE, and doing fast graphics was all but impossible unless you were an expert assembly coder and sprung the $100ish for their assembler). I still haven't forgiven my parents for that.
Parsec was also flat-out excellent, and it had one of the best versions of Miner 2049er out there, but that appeared on so many systems that it could hardly be called a hidden gem.
So far as I know, Knuth has done essentially zero work related to the P/NP question; a lot of algorithmics and tons of fantastic work in combinatorics, but I can't think of a single significant result he's contributed to complexity theory. While it's not impossible that he could have some sort of 'outsider breakthrough', it seems almost infinitesimally unlikely given the mathematical context and techniques that have had to be developed for similar complexity problems. My money would be on either a formal open-sourcing of the TeX codebase or the development of a full HTML5 rendering engine for TeX along the lines of the system that mathoverflow.net uses.
The reason why PageRank 'has to be' linear is essentially mathematical; treating importance as a linear function of the importance of inbound links means that the core equations that need to be solved to determine importance are linear and the answer can be found with (essentially) one huge matrix inversion. If you make importance nonlinear then the equations being solved become computationally infeasible.
What's interesting to me is how close the connections are between PageRank and the classic light transfer/heat transfer equations that come up in computer graphics' radiosity (see James Kajiya's Rendering equation); I wonder if there's a reasonable equivalent of 'path tracing' (link tracing?) for computing page ranks that avoids the massive matrix inversions of the basic PageRank algorithm.
While it's unsurprising given that the current Unreal Engine is still in active development and a ton of commercial games are still being developed and shipped using it, it's worth pointing out that this isn't a source code release; instead, it's something much closer to an elaborate mod engine, with generous swaths of behavioral scripting but no real ability to get 'under the hood' as it were. Still, kudos to Epic for this; it'll be interesting to see who picks up the ball and runs with this.
Especially in a case like this, where there are other tightly-correlated variables. Why is the authors' presumption that it's the cosmic rays (or lack thereof) that are regulating tree growth, rather than solar and sunspot activity itself? It seems at least as plausible to me that sunspot activity correlates to some other solar features (e.g., solar irradiance) that would have a more natural and direct effect on tree growth than cosmic rays.
Strictly speaking, there aren't any seriously new methods of multiplying numbers here; even the techniques they use for handling multiplicands larger than the computer's memory (sectional FFTs, using the Chinese Remainder Theorem to solve the problem by reducing modulo a lot of small primes) are pretty well-established from things like computations of pi, with this group offering a few improvements to the core ideas. What they did provide, and what sounds particularly promising, is a library (judging from the article, likely even open-source) for handling bignums like this that they've made available for general use. It'll be interesting to see if anyone else picks up this ball and runs with it.
...depending on your moral stance; the company (which unfortunately owns a host of major computer book publishers, most notably Academic Press, Digital Press and Morgan Kaufmann) has had a small host of scandals, mostly concerning exorbitant journal fees and 'sponsored' pharmaceutical journals (they were the publisher behind the Merck Australasian Journal of Bone and Joint Medicine scandal, if you recall that). MK and AP publish some of the finest books in the industry, which makes this that much harder a moral stand to take, but it's worth evaluating how you feel about the publisher before you consider going down that route.
'related' is an odd word to see there; what 'PS3-related' hardware would this be, and where's the evidence for this random assertion? Is this meant to imply that Sony's suppliers are getting more orders for some of the hardware used to build the PS3 (which could mean more PS3s in the pipeline but could also be wholly unrelated), or just a consolist claim that Sony is doing better than its sales figures would indicate?
...that Google licenses this to scanner manufacturers and we see this at a consumer level at some point in the future? I know I'd pay good money for a book scanner that doesn't need to have a 'book edge' (which you already have to pay through the nose for)...