I'm relatively certain that most fifteen year old "emo kids" don't hold computer science masters degrees, running from conference to conference to give talks on changing the nature of computing using the web, in-between friending classmates on facebook.
A patent is supposed to contain, among other things, a list of claims, and a disclosure section. The claims contain what is protected by the patent, in disjunctive normal form - by which I mean, in order to violate the patent, you have to violate the entirety of at least one clause. After reading what's claimed, you're supposed to think, "Gee, so that's protected, but how did they achieve those results?" At that point you look at the disclosure and find out. The whole point of the patent is that the inventor is disclosing to the public what might otherwise remain a trade secret, and in exchange they are protected from imitators and competition for a certain time.
Now here's the thing: if you can read the claims, and know exactly what's going on without even looking at the disclosure, then that's a bad patent. It seems all the patents we hear about these days have that problem. In this case, it was so bad that you can figure out the process just by reading the abstract! So when I saw that, I really wanted to see what they could possibly put in the disclosure to justify this document. Unfortunately, the provided link has no disclosure! It includes one "by reference", but I cannot find "U.S. patent application Ser. No. 10/792,405, filed Mar. 3, 2004" online. I'm sure it exists in the physical world, or else it would not even be technically grantable, but they don't seem to have made it available.
I'm also quite miffed that so many people here are concentrating on prior art when the obvious factor seems to outweigh it (although prior art certainly is easier to prove).
Now excuse me as I close all my firefox tabs to get back 50% of my cpu.
At this point, I'm confused enough about the particular definitions being used here, and the model of Turing machine / cellular automaton that they're using, that I don't know. My original impression was that the definition of the machine had to be slightly modified so that the simple pattern of non-zero bits was built into the tape at initialization, but it looks like it could be that the pattern is supplied as input to the system. I'm not familiar with their model, if it's not just a normal semi- or bi-infinite tape Turing machine.
Whoops, I spoke before I saw the pdf, and was under the impression that the article was referring to a cellular system. Still, such emergent properties can take a long, long time to manifest themselves.
The initial parameters don't include the pattern, as the pattern is part of the machine. If we have the machine, we have the pattern, and we have Turing completeness.
(I hate it when I can't find the exact original quote online to verify it, before I look like a fool in front of hundreds of thousands of Simpsons quoters)
If I had mod points (and hadn't posted all over this conversation) I'd mark you insightful for being one of the few people in a discussion of cardinality on slashdot to understand what they're talking about. And I agree, powerset tapes fail too quickly.
Strange thing about public domain: there's no explicit copyright concept of putting a work in the public domain by choice. It can get in there if it's a government work, or if it expires (Ha!), but there's nothing in the law that specifically gives a holder the right to manually place it in there (source: Luis Ibáñez; neither of us are lawyers). This is a technicality, of course, and probably resulted from the fact that when copyright law was changed in the 70s to take affect without registration, nobody thought of an instance where copyright would be undesirable, given the choice. The prevailing model then was still creating a real work worthy of protection, and filing paperwork so you could sue the pants off people and get statutory damages; that copyright could extend its reach so far into the daily activities of individuals was probably unforeseen. And really, it wouldn't be half the problem it is today if it weren't for the fact that the internet blurs the line between written reproduction and free dialogue.
Funny, I always considered using cryptography itself to be security through obscurity.;)
Did you know that if you built a Dyson sphere around Bruce Schneier... Well, I forgot the rest, but it implies that he's made of something besides matter and occupies something besides space.
Recently, seismologists have concluded that a series of tremors are in actuality the result of Bruce Schneier and Chuck Norris communicating through a series of cryptographic roundhouse kicks. (That one's from someone on slashdot)
I'd like to see a solution a bit more involved than simple whitelisting. This may fall into that famous "Why your idea to stop spam can't work" rubric, but I'll suggest it anyway.
I'd like a system where all incoming email was accompanied by some chain of validation. For instance you may have a root set of friends' addresses from which you accept, and each of them would have an identifying/authenticating code. Then you may decide to allow your friends' friends, in which case those emails would have two codes, and so on, up to a certain number. Subscribing to any online entity that requires an email address would also require another unique code in order for their mail to not bounce. The idea is that in order for any mail to arrive at your box, it has to have some connection to an entity you already trust to not spam you. If it turns out that you do receive spam, you simply look at the offending mail's chain of codes and determine which one was too broad in its decision to grant further people right-of-way to your inbox.
This would make establishing new relations out-of-the-blue more difficult, as does normal whitelisting, but is more flexible. It would also provide an easy way to categorize messages. So, I could post my email to a project's wiki, along with a code, and revoke that code if spam gets through using it. It would also make it impossible for companies to sell/share your email with their associates without you knowing who was responsible.
If this type of thing became widespread, then as you said, the only people without such filters and who are thus susceptible to spam, would be people who require unrestricted communication via email and hopefully know enough not to support the business model.
No, it's *obvious* that there's a tradeoff between additional confirmation and accuracy. Furthermore, it's obvious that in situations where immediate subsequent cancellation is possible, accuracy can be sacrificed for the sake of having fewer confirmations.
Even if it wasn't obvious that it would work as well as it did, its existence in itself was not extraordinary or unique or so worthwhile that it deserved a patent.
In answer to your sig, perhaps it's because monospace fonts are more common on the internet.
Er.. I'm not sure where you're disagreeing with me. I never implied that your performance in smash is not a function of skill. I was arguing that the smash experience is about skill mingling with chance, and especially one's skill in effectively using items. Anyone who plays smash exclusively without items does not understand the point of the game.
> "They've played the HECK out of Smash using its usual setups, and have simply discovered ways to play the game competitively that your casual players wouldn't have even thought of." True, but they also play the game competitively in ways that are obvious, yet I simply wouldn't do. Mostly this pertains to choice of characters.
> "They are the most dedicated, die-hard Smash Bros fans you'll ever meet." For certain definitions of what Smash Bros is, yes. For the way I play, and what Smash Bros really means to me, they are the worst group of players.
> "I can promise you that most of the folks in the Smash tournament scene would be very offended at what you wrote " Of that I have no delusions.
> "because it's so unbelievably inaccurate. The game *can* be based on raw skill under certain setups; " That's not the point I was making. When I say it's *not* based on that, I mean that any situation under which chance (in the form of items) and asymmetry is not a factor, is not a proper instance of a Smash Bros game. I know you can take the Smash Bros title and play on final destination with sword characters and fox, and get a fairly even game out of that; but that is not Smash Bros.
> "that's how strong Smash's customizeability is!" Agreed. I just wish there were more tournaments that required you play pichu or ness on the Ice Climber's level, so we could actually see some professional usage of that customizability, rather than finding the absolute minimum variance in options and playing the hell out of it.
> "And yes, competition in a level playing field can be load of fun (Starcraft/Halo anyone?), I refuse to acknowledge Halo's existence. As for Starcraft, suppose that the races weren't as perfectly balanced as they are: would you condone the removal of all but one race, just to ensure balance via symmetry? Would you get rid of all maps that didn't have exactly the same resources, positioned in a manner that worked with the game engine to make sure that worker units had exactly the same opportunity to harvest them at a particular rate? At what point do you stop and say, "This isn't the game I bought, this isn't why I play it"?
> "especially after you've had the game since 2001." Check.
Oh absolutely. When machines get more powerful and can perform the same function for a tenth of the cost, they won't sell you the same machine at the reduced price running the same software. They'll sell you a more powerful machine at the same price, and upgrade your software's bloat to make you require the horsepower.
Funny how I can pretty much do everything I do with my new lenovo T61 windows (formally vista, now XP) laptop, on my six (?) year old 1.2 GHz Sempron running gentoo.
So who hear joined me in screaming out a nice, James Earl Jones cry upon reading the subject title?
Still, as has been echoed above, I'd rather an awesome game than an early game.... Ok, somehow I hit the lameness filter for too much repetition. Somebody's gotta tell Taco or whoever else is in charge of that, that English is 75% redundant. Maybe I should just post an ascii representation of the gzipped contents of my post. Hm.. is my post long enough yet? Apparently so.
Re:Sure it is possible to search 10^60
on
Cracking Go
·
· Score: 1
Of course Moore's law is exponential: that's why we're able to do things like brute-force DES, which requires 2^47 operations. That's why I included "(assuming we remain limited by Moore's law *and level off in the foreseeable future*)" (emphasis added) in my post. A better way to phrase the semantics of that statement would have been "(assuming Moore's law no longer holds and levels off in the foreseeable future)"
A fun game. It's not based on raw skill and straightforward proper-sequence button mashing. There's some strategy, prediction, knowing your opponent, and moments when all that goes to hell because the other guy got a star, heart, and hammer at the same time. Tournament players don't know what smash is because they're too busy studying how to break it rather than have fun.
Re:Sure it is possible to search 10^60
on
Cracking Go
·
· Score: 2, Insightful
The greatest shortcoming of the human race is our apparent inability to understand the exponential function. You, the parent, the grandparent, and the summary seem to all not understand the nature of an intractable problem. "Trillions" of operations a second means nothing when the search space is that big, and no advance in computing (assuming we remain limited by Moore's law and level off in the foreseeable future) will help.
So while you, the parent, and the grandparent were trying to be funny, you were all grossly understating the difficulty. Although I like that Self->Grandchildren->Species is possibly an exponential progression - I would have been annoyed if it had been Self->Grandchildren->GreatGreatGrandchildren.
Of course it sounds like they made their decision; they *did* hear all the arguments. If the requirement of having to establish actual infringement had been placed on the RIAA, the jury feels that it would not change anything, *because* that point was already made in the trial they did hear.
Slashdot Mirror
I'm relatively certain that most fifteen year old "emo kids" don't hold computer science masters degrees, running from conference to conference to give talks on changing the nature of computing using the web, in-between friending classmates on facebook.
There's also a Web Ontology Language whose acronym is OWL. I propose coining the word "Anagrym" for such instances.
A patent is supposed to contain, among other things, a list of claims, and a disclosure section. The claims contain what is protected by the patent, in disjunctive normal form - by which I mean, in order to violate the patent, you have to violate the entirety of at least one clause. After reading what's claimed, you're supposed to think, "Gee, so that's protected, but how did they achieve those results?" At that point you look at the disclosure and find out. The whole point of the patent is that the inventor is disclosing to the public what might otherwise remain a trade secret, and in exchange they are protected from imitators and competition for a certain time.
Now here's the thing: if you can read the claims, and know exactly what's going on without even looking at the disclosure, then that's a bad patent. It seems all the patents we hear about these days have that problem. In this case, it was so bad that you can figure out the process just by reading the abstract! So when I saw that, I really wanted to see what they could possibly put in the disclosure to justify this document. Unfortunately, the provided link has no disclosure! It includes one "by reference", but I cannot find "U.S. patent application Ser. No. 10/792,405, filed Mar. 3, 2004" online. I'm sure it exists in the physical world, or else it would not even be technically grantable, but they don't seem to have made it available.
I'm also quite miffed that so many people here are concentrating on prior art when the obvious factor seems to outweigh it (although prior art certainly is easier to prove).
Now excuse me as I close all my firefox tabs to get back 50% of my cpu.
> "After intensive study of the proposed method in the flowchart, it appears to be an if statement."
Heh. I like how that's put.
"Upon further inspection, these are loafers."
At this point, I'm confused enough about the particular definitions being used here, and the model of Turing machine / cellular automaton that they're using, that I don't know. My original impression was that the definition of the machine had to be slightly modified so that the simple pattern of non-zero bits was built into the tape at initialization, but it looks like it could be that the pattern is supplied as input to the system. I'm not familiar with their model, if it's not just a normal semi- or bi-infinite tape Turing machine.
Whoops, I spoke before I saw the pdf, and was under the impression that the article was referring to a cellular system. Still, such emergent properties can take a long, long time to manifest themselves.
The initial parameters don't include the pattern, as the pattern is part of the machine. If we have the machine, we have the pattern, and we have Turing completeness.
But yours is a trivial solution...
"Even BSD?"
"It. Runs. Everything! (Stupid)"
(I hate it when I can't find the exact original quote online to verify it, before I look like a fool in front of hundreds of thousands of Simpsons quoters)
If I had mod points (and hadn't posted all over this conversation) I'd mark you insightful for being one of the few people in a discussion of cardinality on slashdot to understand what they're talking about. And I agree, powerset tapes fail too quickly.
Have you ever seen a Universal Turing Machine implemented in Conway's Game of Life?
http://www.cs.ualberta.ca/~bulitko/F02/papers/tm_words.pdf
Constructing such a thing in cellular automata can take a while.
Cue the I-don't-get-it post. If that was a joke, it went over my head as I don't see a logical implication operator.
Strange thing about public domain: there's no explicit copyright concept of putting a work in the public domain by choice. It can get in there if it's a government work, or if it expires (Ha!), but there's nothing in the law that specifically gives a holder the right to manually place it in there (source: Luis Ibáñez; neither of us are lawyers). This is a technicality, of course, and probably resulted from the fact that when copyright law was changed in the 70s to take affect without registration, nobody thought of an instance where copyright would be undesirable, given the choice. The prevailing model then was still creating a real work worthy of protection, and filing paperwork so you could sue the pants off people and get statutory damages; that copyright could extend its reach so far into the daily activities of individuals was probably unforeseen. And really, it wouldn't be half the problem it is today if it weren't for the fact that the internet blurs the line between written reproduction and free dialogue.
Without a license, the recipient would not have the right to redistribute the work to friends. I think you mean public domain.
Funny, I always considered using cryptography itself to be security through obscurity. ;)
Did you know that if you built a Dyson sphere around Bruce Schneier... Well, I forgot the rest, but it implies that he's made of something besides matter and occupies something besides space.
Recently, seismologists have concluded that a series of tremors are in actuality the result of Bruce Schneier and Chuck Norris communicating through a series of cryptographic roundhouse kicks. (That one's from someone on slashdot)
I'd like to see a solution a bit more involved than simple whitelisting. This may fall into that famous "Why your idea to stop spam can't work" rubric, but I'll suggest it anyway.
I'd like a system where all incoming email was accompanied by some chain of validation. For instance you may have a root set of friends' addresses from which you accept, and each of them would have an identifying/authenticating code. Then you may decide to allow your friends' friends, in which case those emails would have two codes, and so on, up to a certain number. Subscribing to any online entity that requires an email address would also require another unique code in order for their mail to not bounce. The idea is that in order for any mail to arrive at your box, it has to have some connection to an entity you already trust to not spam you. If it turns out that you do receive spam, you simply look at the offending mail's chain of codes and determine which one was too broad in its decision to grant further people right-of-way to your inbox.
This would make establishing new relations out-of-the-blue more difficult, as does normal whitelisting, but is more flexible. It would also provide an easy way to categorize messages. So, I could post my email to a project's wiki, along with a code, and revoke that code if spam gets through using it. It would also make it impossible for companies to sell/share your email with their associates without you knowing who was responsible.
If this type of thing became widespread, then as you said, the only people without such filters and who are thus susceptible to spam, would be people who require unrestricted communication via email and hopefully know enough not to support the business model.
No, it's *obvious* that there's a tradeoff between additional confirmation and accuracy. Furthermore, it's obvious that in situations where immediate subsequent cancellation is possible, accuracy can be sacrificed for the sake of having fewer confirmations.
Even if it wasn't obvious that it would work as well as it did, its existence in itself was not extraordinary or unique or so worthwhile that it deserved a patent.
In answer to your sig, perhaps it's because monospace fonts are more common on the internet.
Er.. I'm not sure where you're disagreeing with me. I never implied that your performance in smash is not a function of skill. I was arguing that the smash experience is about skill mingling with chance, and especially one's skill in effectively using items. Anyone who plays smash exclusively without items does not understand the point of the game.
> "They've played the HECK out of Smash using its usual setups, and have simply discovered ways to play the game competitively that your casual players wouldn't have even thought of."
True, but they also play the game competitively in ways that are obvious, yet I simply wouldn't do. Mostly this pertains to choice of characters.
> "They are the most dedicated, die-hard Smash Bros fans you'll ever meet."
For certain definitions of what Smash Bros is, yes. For the way I play, and what Smash Bros really means to me, they are the worst group of players.
> "I can promise you that most of the folks in the Smash tournament scene would be very offended at what you wrote "
Of that I have no delusions.
> "because it's so unbelievably inaccurate. The game *can* be based on raw skill under certain setups; "
That's not the point I was making. When I say it's *not* based on that, I mean that any situation under which chance (in the form of items) and asymmetry is not a factor, is not a proper instance of a Smash Bros game. I know you can take the Smash Bros title and play on final destination with sword characters and fox, and get a fairly even game out of that; but that is not Smash Bros.
> "that's how strong Smash's customizeability is!"
Agreed. I just wish there were more tournaments that required you play pichu or ness on the Ice Climber's level, so we could actually see some professional usage of that customizability, rather than finding the absolute minimum variance in options and playing the hell out of it.
> "And yes, competition in a level playing field can be load of fun (Starcraft/Halo anyone?),
I refuse to acknowledge Halo's existence. As for Starcraft, suppose that the races weren't as perfectly balanced as they are: would you condone the removal of all but one race, just to ensure balance via symmetry? Would you get rid of all maps that didn't have exactly the same resources, positioned in a manner that worked with the game engine to make sure that worker units had exactly the same opportunity to harvest them at a particular rate? At what point do you stop and say, "This isn't the game I bought, this isn't why I play it"?
> "especially after you've had the game since 2001."
Check.
Oh absolutely. When machines get more powerful and can perform the same function for a tenth of the cost, they won't sell you the same machine at the reduced price running the same software. They'll sell you a more powerful machine at the same price, and upgrade your software's bloat to make you require the horsepower.
Funny how I can pretty much do everything I do with my new lenovo T61 windows (formally vista, now XP) laptop, on my six (?) year old 1.2 GHz Sempron running gentoo.
So who hear joined me in screaming out a nice, James Earl Jones cry upon reading the subject title?
...
Still, as has been echoed above, I'd rather an awesome game than an early game.
Ok, somehow I hit the lameness filter for too much repetition. Somebody's gotta tell Taco or whoever else is in charge of that, that English is 75% redundant. Maybe I should just post an ascii representation of the gzipped contents of my post. Hm.. is my post long enough yet? Apparently so.
Of course Moore's law is exponential: that's why we're able to do things like brute-force DES, which requires 2^47 operations. That's why I included "(assuming we remain limited by Moore's law *and level off in the foreseeable future*)" (emphasis added) in my post. A better way to phrase the semantics of that statement would have been "(assuming Moore's law no longer holds and levels off in the foreseeable future)"
A fun game. It's not based on raw skill and straightforward proper-sequence button mashing. There's some strategy, prediction, knowing your opponent, and moments when all that goes to hell because the other guy got a star, heart, and hammer at the same time. Tournament players don't know what smash is because they're too busy studying how to break it rather than have fun.
The greatest shortcoming of the human race is our apparent inability to understand the exponential function. You, the parent, the grandparent, and the summary seem to all not understand the nature of an intractable problem. "Trillions" of operations a second means nothing when the search space is that big, and no advance in computing (assuming we remain limited by Moore's law and level off in the foreseeable future) will help.
So while you, the parent, and the grandparent were trying to be funny, you were all grossly understating the difficulty. Although I like that Self->Grandchildren->Species is possibly an exponential progression - I would have been annoyed if it had been Self->Grandchildren->GreatGreatGrandchildren.
Of course it sounds like they made their decision; they *did* hear all the arguments. If the requirement of having to establish actual infringement had been placed on the RIAA, the jury feels that it would not change anything, *because* that point was already made in the trial they did hear.