Yes, the size of the problem is of course paramount, even for polynomial time algorithms. My point (which I didn't express perfectly) was that the GP took "unsolvable" to mean "not Turing computable", while I just took it to mean "the hardware will probably break or the programmer will die long before the thing finishes", i.e. the problem is practically (though not abstractly) unsolvable.
I read it as "unsolvable" in the sense that computers in the real world cannot currently solve them with current algorithms, instead of unsolvable in the abstract sense of, for instance, a Turing machine with unlimited steps and tape solving them in a finite number of steps. But, I agree that the wording is imprecise.
Many processes have near-linear behavior for a time before switching to more complex behavior later. Using percentages (instead of raw usage numbers) can help create non-linear behavior, but on the right time scale any sufficiently smooth function is approximately linear. In any case, ignoring linear extrapolations out of hand is as bad as relying on them without noting their limitations.
Without doing any halfways meaningful statistical analysis, IMO it looks like the trends didn't significantly change when the ballot was introduced. Perhaps that means some effect was canceled, or the ballot was nearly ineffective--who knows, though I lean toward the latter without more info. Averaging deltas over time looks to be a reasonably effective way to predict future trends over a few months within perhaps 1-2%, with the error margin and time scale (ironically) ~proportional.
What we do see from the chart is that everyone won because it appeared that many made a choice, and when they did over half of them went with a browser that was not MS.
What's the source of your "over half" remark? The only thing I see from the chart is that the trends don't appear to have wildly changed post-dotted-line. Perhaps you meant many people made a choice to install a non-MS browser independent of the choice offered them by the browser ballot? I wouldn't say "everybody won" so much as "everybody is winning" in that case.
I really need to stop reading anything on that site. I can't edit it (it'd be horrible to get started even attempting to correct anything... and they don't have an automated signup form), but my inner mathematician just cringes on
Minkowski space is predicated on the idea of four-dimensional vectors of which one component is time. However, one of the properties of a vector space is that every vector have an inverse. Time cannot be a vector because it has no inverse.
I've studied quite a bit of abstract and linear algebra and a chunk of relativity but this is news to me. "Time" is both a vector and a component of a vector! And of course it can have no inverse--no more reasoning or justification could possibly be necessary or reasonably expected to explain what the hell this statement means! All hail the fount of knowledge and wisdom that is Conservapedia! Seriously, do they think every scientist in the world just makes shit up all day long? Actually, that would make sense--they figure everyone just does what they themselves do: say whatever nonsense comes to mind, write it down, and defend it to the death when challenged, not because you're really right, but because that's just what you do.
Wow. That explains basically all of their behavior.... They were never good at figuring things out themselves, but they can argue decently in an indecisive way, so they figure everyone else does the same. And then Mr. Shlaffy (sp? no way am I looking it up) started a Wiki where like-minded folk can be shockingly stupid together.
If only it had been blocked for me.... I spent the last 45 minutes or so reading bits and pieces and now I'm slightly ill--as in the process that leads to vomiting has started. How can people be so stupid. I'm stopping before it gets worse.
I'm sorry, I didn't check authors and assumed you were the author of both the great-grandparent and great-great-great-grandparent posts. My "Fundamentally,..." comment was directed at the great-grandparent's author and not you.
I still think you're being too extreme with your clarified point. "Christianity" doesn't have a well-defined set of beliefs to subscribe to. Many self-described Christians disagree on what defines a Christian. This differs from your soccer example, where it's very clear to everyone that you can't use your hands and the rules are written down and agreed upon. I would agree with your point ("if you don't subscribe to all the beliefs of a particular religion, then you cannot say you belong to that religion") if each religion had a well-defined set of beliefs to subscribe to. In some particular cases I agree: a Catholic woman who tries to be a priest does seem to contradict a well-enough-defined belief of Catholicism that she shouldn't call herself Catholic anymore. A "Christian" who rejects the book of Hebrews is much less clear; some people would presumably require belief in the usual non-apocryphal books for someone to qualify as "Christian", while some wouldn't care.
I think they think they're sacrificing time and effort to ensure their child's safety, which they actually are. They're also sacrificing some of their child's privacy--which they probably see as a good thing in small doses, so sacrificing it is a small but necessary evil in their eyes. It's short-sighted and probably selfish (the parent focuses on their own feelings of worry instead of their child's honest well being), but it makes sense to some people. In some situations it might even be a good thing to monitor your kids with a keylogger--if they're really stupid, for instance, so long as you told them about it.
Having objections to Christianity is different from saying people who disagree with some point (possibly a large point) of their religion should throw it away wholescale. To be clear, that was the point I "wasn't so sure" about:
The solution when your chosen religion conflicts with your lifestyle and biology is not to try and reinterpret and redefine that religion's beliefs to align with yours, it's to stop believing in that religion and choose another (or none at all).
Your argument seemed to be that if your beliefs are at all inconsistent with your religion, you should throw the entire religion's beliefs away and start anew (or not at all). I gave an example from math (which most people agree is a logical discipline) where that behavior is too extreme. Fundamentally, you seem to just want to encourage people to get away from Christianity by coming to some realization that Christianity is wildly inconsistent / made up, and allowed that desire to fudge your thinking in general.
To be honest, I don't think the message is serious. For reference:
'We, the collective super-consciousness known as ANONYMOUS – the Voice of Free Speech & the Advocate of the People – have long heard you issue your venomous statements of hatred, and we have witnessed your flagrant and absurd displays of inimitable bigotry and intolerant fanaticism,'
"super-consciousness" is more silly than descriptive; the caps remind me of terrible New Age books trying to make everything seem important; the tone is high and mighty in a way that it's almost a charicature of God talking to Moses; and the diction is like a teenager trying to sound important by using a thesaurus. It's just over the top. I think this is a cry for attention more than a serious statement of (sucky) principles.
I'm not so sure. If I disagree with constructivist logicians (very roughly, mathematicians who don't let you say "not not X is X"), I don't throw out all logical reasoning--I just allow proof by contradiction and go on my merry way. The opposite also holds. If I'm a constructivist, I just have to make sure your proofs didn't use the forbidden rules before I accept them.
In a sense splits in religion are the same. They didn't want to throw everything out, so they just excised pieces and went on their merry way. To assume your religion has to be 100% consistent for you to believe in it is tempting, but ultimately silly. It's unclear if number theory itself is consistent, and that's probably the best shot we humans have at a nontrivial consistent set of beliefs.
My response was mostly to your sentence, "Take that [Google's first or second hit on some search], add a tiny bit of code to clean it up, and you're done," which greatly oversimplifies the process. My example showed how oversimplified your idea is, which makes it relevant. You made another point, that Watson's behavior could be reproduced via Google; my example is irrelevant to this point, as you say, though that's immaterial.
Well... yes and no. A decently competent human can answer the vast majority of Jeopardy questions very quickly if given access to Google. Automatically sifting through the results is difficult (as you can see if you try pasting a few questions verbatim), and Google's database is much larger than Watson's.
As an example of the non-triviality of "sifting" Google results to find the correct answer, try to look answer this: "As of 2010, Croatia & Macedonia are candidates but this is the only former Yugoslav republic in the EU". I'd have to search to find out what the former Yugoslav republics were, get a list of EU members, and compare the two.
I was wondering if Watson didn't include (or didn't effectively include) the category in its search. From the clue text alone, it was actually hard to tell it was asking for a US city.
In effect, Watson is not playing the same game as what we normally call "Jeopardy."
It's interesting you say that, since Watson's strategy was remarkably different from a human's. It fished for Daily Doubles by calling high-value questions early. Most humans go sequentially through categories, starting with the low value questions, which is almost traditional at this point. Watson also often jumped around in the categories, while humans tend to finish off a category before proceeding to another.
Watson's wagers were very strange for Jeopardy in that he used all the significant figures, eg. $1234, instead of following the usual route of rounding to the nearest $100. Trebek remarked on this several times and the audience laughed when Watson made such strangely specific wagers. Watson used at least one very large Daily Double wager, while most humans are more conservative. Humans often buzz in because they "know they know" and figure out the answer during the next few seconds--Jennings used this a number of times, apparently in an effort to beat Watson to the buzzer. By contrast, Watson always answered immediately when he was called.
It struck me that some of Watson's programmers probably aren't fans of Jeopardy because they made it break established conventions. Either that, or they said "screw tradition, we want to win".
And comparing human vs machine play styles is also largely pointless
I disagree. The machine's play style (a souped up database search) doesn't generalize to many other tasks while the human's (actually understand the content, or make reasonable inferences) does. I see Watson as a wonderful database search program that appears to parse sentences well. I'd love to be able to ask it things like I do Wolfram Alpha, but Watson's use is in content retrieval and not content creation. That it can't do more than search is a key "failing" of its play style compared to a human's. That certainly doesn't make it useless.
I was also annoyed when they kept saying Watson "understood" natural language. I don't believe parsing equals understanding, and it clearly didn't truly understand its answers by the mistakes it made and the alternate choices it came up with.
I'm not so sure. TheSpec's article's first line says "THE NEW YORK TIMES NEWS SERVICE". Perhaps they have some sort of a deal with the NY Times allowing them to reprint articles. At the least, it's not as malicious as you suggest since they give some attribution.
Thank you. This was what I was wondering about and TFA implies CDs and DVDs are also affected.
I know quite a few people who would be baffled by running a CD manually, though they're competent in other ways. I can just imagine the increase in tech support calls if CDs and DVDs were affected.
Slashdot Mirror
Yes, the size of the problem is of course paramount, even for polynomial time algorithms. My point (which I didn't express perfectly) was that the GP took "unsolvable" to mean "not Turing computable", while I just took it to mean "the hardware will probably break or the programmer will die long before the thing finishes", i.e. the problem is practically (though not abstractly) unsolvable.
Exactly, that is what I meant.
I read it as "unsolvable" in the sense that computers in the real world cannot currently solve them with current algorithms, instead of unsolvable in the abstract sense of, for instance, a Turing machine with unlimited steps and tape solving them in a finite number of steps. But, I agree that the wording is imprecise.
Many processes have near-linear behavior for a time before switching to more complex behavior later. Using percentages (instead of raw usage numbers) can help create non-linear behavior, but on the right time scale any sufficiently smooth function is approximately linear. In any case, ignoring linear extrapolations out of hand is as bad as relying on them without noting their limitations.
Without doing any halfways meaningful statistical analysis, IMO it looks like the trends didn't significantly change when the ballot was introduced. Perhaps that means some effect was canceled, or the ballot was nearly ineffective--who knows, though I lean toward the latter without more info. Averaging deltas over time looks to be a reasonably effective way to predict future trends over a few months within perhaps 1-2%, with the error margin and time scale (ironically) ~proportional.
What we do see from the chart is that everyone won because it appeared that many made a choice, and when they did over half of them went with a browser that was not MS.
What's the source of your "over half" remark? The only thing I see from the chart is that the trends don't appear to have wildly changed post-dotted-line. Perhaps you meant many people made a choice to install a non-MS browser independent of the choice offered them by the browser ballot? I wouldn't say "everybody won" so much as "everybody is winning" in that case.
Minkowski space is predicated on the idea of four-dimensional vectors of which one component is time. However, one of the properties of a vector space is that every vector have an inverse. Time cannot be a vector because it has no inverse.
I've studied quite a bit of abstract and linear algebra and a chunk of relativity but this is news to me. "Time" is both a vector and a component of a vector! And of course it can have no inverse--no more reasoning or justification could possibly be necessary or reasonably expected to explain what the hell this statement means! All hail the fount of knowledge and wisdom that is Conservapedia! Seriously, do they think every scientist in the world just makes shit up all day long? Actually, that would make sense--they figure everyone just does what they themselves do: say whatever nonsense comes to mind, write it down, and defend it to the death when challenged, not because you're really right, but because that's just what you do.
Wow. That explains basically all of their behavior.... They were never good at figuring things out themselves, but they can argue decently in an indecisive way, so they figure everyone else does the same. And then Mr. Shlaffy (sp? no way am I looking it up) started a Wiki where like-minded folk can be shockingly stupid together.
The ratio of registered users to readers is very low. The number that really got me was the total page views: over 217 million.
If only it had been blocked for me.... I spent the last 45 minutes or so reading bits and pieces and now I'm slightly ill--as in the process that leads to vomiting has started. How can people be so stupid. I'm stopping before it gets worse.
I had never heard of Conservapedia before. Now I really wish I hadn't.
I'm sorry, I didn't check authors and assumed you were the author of both the great-grandparent and great-great-great-grandparent posts. My "Fundamentally, ..." comment was directed at the great-grandparent's author and not you.
I still think you're being too extreme with your clarified point. "Christianity" doesn't have a well-defined set of beliefs to subscribe to. Many self-described Christians disagree on what defines a Christian. This differs from your soccer example, where it's very clear to everyone that you can't use your hands and the rules are written down and agreed upon. I would agree with your point ("if you don't subscribe to all the beliefs of a particular religion, then you cannot say you belong to that religion") if each religion had a well-defined set of beliefs to subscribe to. In some particular cases I agree: a Catholic woman who tries to be a priest does seem to contradict a well-enough-defined belief of Catholicism that she shouldn't call herself Catholic anymore. A "Christian" who rejects the book of Hebrews is much less clear; some people would presumably require belief in the usual non-apocryphal books for someone to qualify as "Christian", while some wouldn't care.
You didn't respond to my point. I'm sorry I made you defensive. There's no point in continuing this thread. Best of luck in life.
I think they think they're sacrificing time and effort to ensure their child's safety, which they actually are. They're also sacrificing some of their child's privacy--which they probably see as a good thing in small doses, so sacrificing it is a small but necessary evil in their eyes. It's short-sighted and probably selfish (the parent focuses on their own feelings of worry instead of their child's honest well being), but it makes sense to some people. In some situations it might even be a good thing to monitor your kids with a keylogger--if they're really stupid, for instance, so long as you told them about it.
The solution when your chosen religion conflicts with your lifestyle and biology is not to try and reinterpret and redefine that religion's beliefs to align with yours, it's to stop believing in that religion and choose another (or none at all).
Your argument seemed to be that if your beliefs are at all inconsistent with your religion, you should throw the entire religion's beliefs away and start anew (or not at all). I gave an example from math (which most people agree is a logical discipline) where that behavior is too extreme. Fundamentally, you seem to just want to encourage people to get away from Christianity by coming to some realization that Christianity is wildly inconsistent / made up, and allowed that desire to fudge your thinking in general.
'We, the collective super-consciousness known as ANONYMOUS – the Voice of Free Speech & the Advocate of the People – have long heard you issue your venomous statements of hatred, and we have witnessed your flagrant and absurd displays of inimitable bigotry and intolerant fanaticism,'
"super-consciousness" is more silly than descriptive; the caps remind me of terrible New Age books trying to make everything seem important; the tone is high and mighty in a way that it's almost a charicature of God talking to Moses; and the diction is like a teenager trying to sound important by using a thesaurus. It's just over the top. I think this is a cry for attention more than a serious statement of (sucky) principles.
I'm not so sure. If I disagree with constructivist logicians (very roughly, mathematicians who don't let you say "not not X is X"), I don't throw out all logical reasoning--I just allow proof by contradiction and go on my merry way. The opposite also holds. If I'm a constructivist, I just have to make sure your proofs didn't use the forbidden rules before I accept them.
In a sense splits in religion are the same. They didn't want to throw everything out, so they just excised pieces and went on their merry way. To assume your religion has to be 100% consistent for you to believe in it is tempting, but ultimately silly. It's unclear if number theory itself is consistent, and that's probably the best shot we humans have at a nontrivial consistent set of beliefs.
My response was mostly to your sentence, "Take that [Google's first or second hit on some search], add a tiny bit of code to clean it up, and you're done," which greatly oversimplifies the process. My example showed how oversimplified your idea is, which makes it relevant. You made another point, that Watson's behavior could be reproduced via Google; my example is irrelevant to this point, as you say, though that's immaterial.
Watson getting the question wrong makes it an even better example that sifting results is non-trivial....
Has the media degraded so far that we are now counting the National Enquirer as a reliable news source?
Yes.
Actually, no; it's degraded farther.
Well... yes and no. A decently competent human can answer the vast majority of Jeopardy questions very quickly if given access to Google. Automatically sifting through the results is difficult (as you can see if you try pasting a few questions verbatim), and Google's database is much larger than Watson's.
As an example of the non-triviality of "sifting" Google results to find the correct answer, try to look answer this: "As of 2010, Croatia & Macedonia are candidates but this is the only former Yugoslav republic in the EU". I'd have to search to find out what the former Yugoslav republics were, get a list of EU members, and compare the two.
I was wondering if Watson didn't include (or didn't effectively include) the category in its search. From the clue text alone, it was actually hard to tell it was asking for a US city.
In effect, Watson is not playing the same game as what we normally call "Jeopardy."
It's interesting you say that, since Watson's strategy was remarkably different from a human's. It fished for Daily Doubles by calling high-value questions early. Most humans go sequentially through categories, starting with the low value questions, which is almost traditional at this point. Watson also often jumped around in the categories, while humans tend to finish off a category before proceeding to another.
Watson's wagers were very strange for Jeopardy in that he used all the significant figures, eg. $1234, instead of following the usual route of rounding to the nearest $100. Trebek remarked on this several times and the audience laughed when Watson made such strangely specific wagers. Watson used at least one very large Daily Double wager, while most humans are more conservative. Humans often buzz in because they "know they know" and figure out the answer during the next few seconds--Jennings used this a number of times, apparently in an effort to beat Watson to the buzzer. By contrast, Watson always answered immediately when he was called.
It struck me that some of Watson's programmers probably aren't fans of Jeopardy because they made it break established conventions. Either that, or they said "screw tradition, we want to win".
And comparing human vs machine play styles is also largely pointless
I disagree. The machine's play style (a souped up database search) doesn't generalize to many other tasks while the human's (actually understand the content, or make reasonable inferences) does. I see Watson as a wonderful database search program that appears to parse sentences well. I'd love to be able to ask it things like I do Wolfram Alpha, but Watson's use is in content retrieval and not content creation. That it can't do more than search is a key "failing" of its play style compared to a human's. That certainly doesn't make it useless.
I was also annoyed when they kept saying Watson "understood" natural language. I don't believe parsing equals understanding, and it clearly didn't truly understand its answers by the mistakes it made and the alternate choices it came up with.
Point taken, I didn't see that line.
I wish I could mod this up. Someone on /. admitted to a mistake!
I'm not so sure. TheSpec's article's first line says "THE NEW YORK TIMES NEWS SERVICE". Perhaps they have some sort of a deal with the NY Times allowing them to reprint articles. At the least, it's not as malicious as you suggest since they give some attribution.
Thank you. This was what I was wondering about and TFA implies CDs and DVDs are also affected.
I know quite a few people who would be baffled by running a CD manually, though they're competent in other ways. I can just imagine the increase in tech support calls if CDs and DVDs were affected.