inspired by the fact that many print outs have a life-span of a few hours (think of the emails you may print out just to read, or the content you proof read on the train journey back home)
About the only thing I can think of that I ever print for "transient" use is driving directions, and usually by the time I'm done with them the paper wouldn't be usable even if the ink did disappear. I can't imagine, more generally, printing material for read-once purposes; if I print something, its not transient.
I expect Xerox has done some market research here, but it sounds to me a solution desperately searching for a problem.
They should've fought, legal fees would very likely be less than $1M.
Fighting the FTC would have resulted in publicity for the fight, which would likely have led to pressure on schools and other places to block access to Xanga, which would have potentially cost them much more than the legal fees and expenses. Though I doubt it would cost them less than $1 million in legal expenses to fight it if the FTC was really serious.
Well, as this settlement reveals, its provided the FTC the ability to force social sites to create rating systems for user generated content. Some people might argue that that has little to do with children's online privacy.
Xanga settled with the FTC for $1 million dollars plus certain policy changes (such as setting up a community moderation system which allows users to flag other users as "underage", and setting up a rating system for user created content), they weren't fined $1 million.
The difference is important; a settlement doesn't mean anything was proven, it means Xanga felt the cost of complying with the settlement terms was worth paying considering the cost of fighting the issue and the risk that they might be fined.
And, really, looking the settlement terms, it looks like the FTC may well have been looking not for real violations, but an opportunity to impose defacto standards by finding someone who'd be willing to settle; particularly the requirement for a rating system for user-generated content looks like an effort to start imposing new rules on the web through bullying that would have little chance of being imposed by legislation or regular public rulemaking (but, once established in several settlements like this with large sites, might have a better chance of being imposed by law or rule on every site allowing public access and community-generated content.)
An opinion piece is an "article" ("piece" and "article" in the relevant senses are synonyms.) It is not a "news article". But the existence of the opinion piece is itself news, as are the underlying facts it relates too, so a Slashdot article pointing to it is not inconsistent with the slogan "News for nerds."
Of course, the full slogan is "News for nerds. Stuff that matters." Whether the second part is a limitation on, or addition to, the first is debatable.
I think MMORPGs have weaker AI simply because its viewed as less of a priority in a game where interaction with lots of other players is a key feature.
Or: Three, Window was a good person to lead a security team, but her work was compromised by policies from Microsoft higher ups. I'm sure there are other possibilities as well. The two options you present aren't the only too possibilities, so there is no reason people would be forced to admit either one.
Long story short, aren't emergent results of a large number of individual AI's essentially the same that you would get from statistical random generation?
Sure, if you got lots of little "agents" acting on the same inputs, applying a function with some random variation, and producing results, and then aggregate them, you could do the same thing with one big function and a different, suitably chosen, random distribution to produce the same results.
OTOH, if the agents are reacting to their immediate environment and interactions, have some sort of individual "memory" and "preference" model driving their behavior, etc., they may not produce results that can be easily modelled by a simple function over the aggregate measures of the conditions.
So, I think the extent to which AI agents produce results that look like statistical random generation based on overall conditions depends on the kind of AI agents used; the shallower the AI, the less possibility for interesting emergent properties.
I dunno. With the assumption that longer viewing distances are becoming desirable in the console market, with giant screens and wireless controllers, anything that makes you physically walk back to the console mid-game is likely to be, if not a deal breaker, a major detraction from the experieince for lots of people.
I supposed if consoles came with multidisc changers or just multiple disc drives and the capacity to use them intelligently automatically, this wouldn't be an issue. But that'd drive up cost, too.
OTOH, lots of modern computers aren't exactly made to just seemlessly drop in another CPU, whereas most are made to accept expansion cards of some sort. You don't have to beat the CPU at its own game with the co-processor, you just need the CPU and co-processor combination to beat the CPU.
But, yeah, this seems like a case where getting a computer with another CPU would be more likely more cost effective to start with than getting one with an "AI coprocessor".
Is it conceivable to create a method where the AI is presented with a view of the game that would closely match the level that a player would experience?
Its certainly conceptually possible to, say, pack away the engine in a "black box" module that exposes the same API to the UI client and the AI client, and only exposes to the UI/AI things a "player" should have access to. How well that works may vary for different types of games.
Any language where (for instance) the subject can be dropped and indicated by verb conjugation in certain cases requires proper conjugation for interpretting meaning, as does any language in which time or other factors can be indicated through verb conjugation loan.
English most certainly is one of those languages, in the latter case: "I [to go] to the store." means different things depending on the conjugation of "to go":
I went to the store. I go to the store. I am going to the store. I will go to the store. I had gone to the store. I will have gone to the store.
In the former case, English is less of a problem, though (e.g.) Spanish becomes quite ambiguous in many cases if the verb is conjugated incorrectly and the verb is the only thing indicating, say, first or second-person subject. (Third person would normally have a specified referrent.)
I don't care on their behalf. Dunn, HP, and their agents appear to have deliberately broken the criminal laws of my state for private gain and I want them punished, not to get away with it because they happen to be rich and powerful.
I care on my behalf, because I like living in an ordered society where the rich are not given license because of their position to violate, flagrantly, the laws that apply to everyone else.
so, my take away is that since there is not a corresponding statue that relates to telephone records, even if it was illegal for the hired investigators to pre-text, it probably was not illegal for HP to ask the investigators to obtain the information.
Anyone who "aids, abets, counsels, commands, induces or procures" the commission of an offense against the United States is punishable exactly as if they had committed the offense (18 USC 2(a)), so no statute regarding procuring someone else to commit the particular illegal act is necessary for federal offenses (some have such provisions to modify the punishment in those cases.)
Further, federal law is not the whole of the law, and most states have false personation laws which could be applicable: most relevantly, probably, given where HP is located, California, in particular, clearly does.
I'm thinking (IANAL, obviously) that if HP paid for all the phone lines, then this would be permissible?
If HP had paid for all the phone lines, they wouldn't (1) have had to lie ("pretext" as a verb, gah!) to get the records, (2) have had to have gone through outside investigators to get them, (3) pretend that they instructed that the outside investigators follow the law and nope, no-sirree, we didn't know anything questionable going on in the investigation until after they gave us the results, and we won't comment now about what we know because its a subject of possible litigation.
As a former employee of Pattie Dunn when she worked at Wells Fargo Nikko Investment Advisors which became Barclays Global Investors, I always found Pattie to be a person who really cared about her employees and their personal lives.
Apparently, she still cares very much about their personal lives.
There's no law that makes it a crime to get someone to give you information (unless it's banking information).
In California, where HP is headquartered, it is a crime to obtain labor through "fraudulent representation or pretense" is guilty just as if they had stolen services with similar value (California Penal Code 532). By representing themselves as the customers of the phone company whose records were requested, they obtained the labor of customer service staff under false pretense.
It is likewise criminal, in California, to willfully obtain "personal identifying information" (including, among many other thingsother things, name, address, telephone number, place of employment, or social security number) of another and then use that information for any unlawful purpose, including "to obtain, or attempt to obtain, credit, goods, services, or medical information" (Penal Code 530.5, emphasis added), without the consent of the person whose information was used. Here, they used several pieces of personal information concerning the directors targetted to obtain services from people with whom those directors did business, and did so without the directors consent.
So to say there is no law which makes it illegal to use someone else's personal information to enable yourself to impersonate that person to get someone to give you information is, well, not exactly true, even outside of banking information.
If you actually read the linked patent, it isn't a patent on conjugating words. It's a patent on automatically providing all of the different possible conjugation forms of any verb on the fly, which is something I, for one, haven't seen before and think could be pretty useful...
Verb conjugation software is hardly even remotely novel.
Right. If the USPTO isn't doing significant filtering, then at the very least the patent holder in a lawsuit should bear an affirmative burden in demonstrating the validity of the patent. Presumptive validity of the patent makes sense only if the USPTO's review process is meaningful.
few people were going to buy one because of the price, and now that supply will be limited, those who did want one, will not be able to, unless they want to pay even more as some enterprising individuals will buy one solely to sell it on ebay.
The people that buy them to sell them on e-Bay (1) are, in fact, buying them, and (2) won't sell them at higher prices unless there are people out there willing to pay higher prices for them.
Slashdot Mirror
About the only thing I can think of that I ever print for "transient" use is driving directions, and usually by the time I'm done with them the paper wouldn't be usable even if the ink did disappear. I can't imagine, more generally, printing material for read-once purposes; if I print something, its not transient.
I expect Xerox has done some market research here, but it sounds to me a solution desperately searching for a problem.
Well, as this settlement reveals, its provided the FTC the ability to force social sites to create rating systems for user generated content. Some people might argue that that has little to do with children's online privacy.
Because:
Xanga settled with the FTC for $1 million dollars plus certain policy changes (such as setting up a community moderation system which allows users to flag other users as "underage", and setting up a rating system for user created content), they weren't fined $1 million.
The difference is important; a settlement doesn't mean anything was proven, it means Xanga felt the cost of complying with the settlement terms was worth paying considering the cost of fighting the issue and the risk that they might be fined.
And, really, looking the settlement terms, it looks like the FTC may well have been looking not for real violations, but an opportunity to impose defacto standards by finding someone who'd be willing to settle; particularly the requirement for a rating system for user-generated content looks like an effort to start imposing new rules on the web through bullying that would have little chance of being imposed by legislation or regular public rulemaking (but, once established in several settlements like this with large sites, might have a better chance of being imposed by law or rule on every site allowing public access and community-generated content.)
An opinion piece is an "article" ("piece" and "article" in the relevant senses are synonyms.) It is not a "news article". But the existence of the opinion piece is itself news, as are the underlying facts it relates too, so a Slashdot article pointing to it is not inconsistent with the slogan "News for nerds."
Of course, the full slogan is "News for nerds. Stuff that matters." Whether the second part is a limitation on, or addition to, the first is debatable.
I think MMORPGs have weaker AI simply because its viewed as less of a priority in a game where interaction with lots of other players is a key feature.
Or: Three, Window was a good person to lead a security team, but her work was compromised by policies from Microsoft higher ups. I'm sure there are other possibilities as well. The two options you present aren't the only too possibilities, so there is no reason people would be forced to admit either one.
Sure, if you got lots of little "agents" acting on the same inputs, applying a function with some random variation, and producing results, and then aggregate them, you could do the same thing with one big function and a different, suitably chosen, random distribution to produce the same results.
OTOH, if the agents are reacting to their immediate environment and interactions, have some sort of individual "memory" and "preference" model driving their behavior, etc., they may not produce results that can be easily modelled by a simple function over the aggregate measures of the conditions.
So, I think the extent to which AI agents produce results that look like statistical random generation based on overall conditions depends on the kind of AI agents used; the shallower the AI, the less possibility for interesting emergent properties.
I dunno. With the assumption that longer viewing distances are becoming desirable in the console market, with giant screens and wireless controllers, anything that makes you physically walk back to the console mid-game is likely to be, if not a deal breaker, a major detraction from the experieince for lots of people.
I supposed if consoles came with multidisc changers or just multiple disc drives and the capacity to use them intelligently automatically, this wouldn't be an issue. But that'd drive up cost, too.
OTOH, lots of modern computers aren't exactly made to just seemlessly drop in another CPU, whereas most are made to accept expansion cards of some sort. You don't have to beat the CPU at its own game with the co-processor, you just need the CPU and co-processor combination to beat the CPU. But, yeah, this seems like a case where getting a computer with another CPU would be more likely more cost effective to start with than getting one with an "AI coprocessor".
Apparently, their chief outside counsel was involved in the whole process, so I don't think not running it by legal was the problem.
Any language where (for instance) the subject can be dropped and indicated by verb conjugation in certain cases requires proper conjugation for interpretting meaning, as does any language in which time or other factors can be indicated through verb conjugation loan.
English most certainly is one of those languages, in the latter case: "I [to go] to the store." means different things depending on the conjugation of "to go":
I went to the store.
I go to the store.
I am going to the store.
I will go to the store.
I had gone to the store.
I will have gone to the store.
In the former case, English is less of a problem, though (e.g.) Spanish becomes quite ambiguous in many cases if the verb is conjugated incorrectly and the verb is the only thing indicating, say, first or second-person subject. (Third person would normally have a specified referrent.)
I don't care on their behalf. Dunn, HP, and their agents appear to have deliberately broken the criminal laws of my state for private gain and I want them punished, not to get away with it because they happen to be rich and powerful.
I care on my behalf, because I like living in an ordered society where the rich are not given license because of their position to violate, flagrantly, the laws that apply to everyone else.
Anyone who "aids, abets, counsels, commands, induces or procures" the commission of an offense against the United States is punishable exactly as if they had committed the offense (18 USC 2(a)), so no statute regarding procuring someone else to commit the particular illegal act is necessary for federal offenses (some have such provisions to modify the punishment in those cases.)
Further, federal law is not the whole of the law, and most states have false personation laws which could be applicable: most relevantly, probably, given where HP is located, California, in particular, clearly does.
In California, where HP is headquartered, it is a crime to obtain labor through "fraudulent representation or pretense" is guilty just as if they had stolen services with similar value (California Penal Code 532). By representing themselves as the customers of the phone company whose records were requested, they obtained the labor of customer service staff under false pretense.
It is likewise criminal, in California, to willfully obtain "personal identifying information" (including, among many other thingsother things, name, address, telephone number, place of employment, or social security number) of another and then use that information for any unlawful purpose, including "to obtain, or attempt to obtain, credit, goods, services, or medical information" (Penal Code 530.5, emphasis added), without the consent of the person whose information was used. Here, they used several pieces of personal information concerning the directors targetted to obtain services from people with whom those directors did business, and did so without the directors consent.
So to say there is no law which makes it illegal to use someone else's personal information to enable yourself to impersonate that person to get someone to give you information is, well, not exactly true, even outside of banking information.
Right. If the USPTO isn't doing significant filtering, then at the very least the patent holder in a lawsuit should bear an affirmative burden in demonstrating the validity of the patent. Presumptive validity of the patent makes sense only if the USPTO's review process is meaningful.
We've got Diebold, so that's covered.