Actually Viagra was invented to treat angina, at which it was a spectacular failure. The better-known use of Viagra was actually a side-effect that appeared in (if I remember) 80% of test subjects. So even Viagra was a sort of accident.
There exists a moderately sized computing world outside of games. 80 cores, as you have pointed out, are clearly not directed towards gamers, or even personal computing at the moment. I would personally love one of these for my simulations, but I can use up absolutely any number of cores without too much trouble. If you want to extend it to games, it isn't very hard to imagine. As someone else mentioned, with a handful of cores you could probably do real-time ray tracing, which is naively parallel and can eat up any number of cores. You'd get photorealistic graphics, probably indistinguishable from real life. Throw in 40 cores there to run that big-screen plasma TV. Throw in a couple of cores for AI, which is much more than path-finding. You can have any number of learning algorithms that monitor every aspect of your gameplay and build better agents. In fact, throw in..say..a dozen cores and you'd have enemies that might actually seem intelligent. A few more cores to track millions of particles and objects in the game.
Remember that a lot of algorithms can be parallelized to use any number of cores (it gets inefficient after a point, but there definitely is an initial speedup).
There's a (somewhat questionably) related application in the real world that was on this new "firehose" thing yesterday: Feng-Gui. It creates a heat-map overlay for any website supposedly highlighting areas that stick out first to human perception.
So PatRobertsonFreeVideoUpload.com has no right to take down a Muslim video posted on their site? ChristianTube.com has no right to take down a Church of Satan video posted on their site? While I agree with what you're generally saying, as someone else pointed out earlier, posting absolutely any video you like on YouTube is not a "right", and discrimination applies when you are denied your rights. In Google's opinion -- and this is one that will hold up in court -- the video was potentially inflammatory. Sure, so was the Christian one by the same definition, but good luck to you if you're going to try to sue out hypocrisy from corporations.
Completely agree, but I think the key word in your post is "religious grounds should"... i.e. "should" as in a guideline, but not a law. I don't see how you could legally bar Google -- which has done nothing unquestionably illegal -- from taking down a video posted on their private site.
Do you have any idea what anti-discrimination laws are about, or are you just throwing words out there? No one discriminated against anyone. The videos in question were uploaded in accordance with Google's TOS. By uploading a video, you agree that they can do whatever they damn well want with your video with respect to deleting it for no damn reason at all. Here's the line taken directly from their TOS:
YouTube reserves the right to remove Content and User Submissions without prior notice.
Bottom line: they don't NEED a reason. It's completely within their rights to delete your videos even if they don't like your username.
This reminds me of the Enron e-mail data that was released, with similarly "shocking" emails. Actually, in the Enron case, they really were illuminating because a lot of e-mails addressed to Ken Lay towards the end of the company's life included the words "you bastard". Also, you didn't have to look very hard to find rampant corporate nepotism (Ken Lay's daughter Elizabeth pimping her friends). The original dataset is at CMU, and a web-browsable version is at enronemail.com, although you have to register for the latter one. The first link lets you download the zipped contents of a bunch of executive's email boxes (sent items, deleted items, inbox, etc.)...it's really nuts.
Most academics think that if an expert operator has been able to do something once on a single dataset they have "solved the problem"
Where are you getting this from? I speak as an academic, and I have found very few academics who think like this, and they definitely aren't respected either in industry or academia. Very few academics will conclude that one technique, much less one paper or operator, has solved any significant problem.
OK, I just checked the actualy iBridge network. Here are some tasty papers and their "human readable" summaries:
Paper:Apparatus and Device for Analyzing Nerve Conduction
Human-readable summary: an improved device for evaluating the performance of nerve or muscle
Paper:Methods and Systems for the Identification of Components of Mammalian Cells as Targets for Therapeutic Agents
Human-readable: Methods and Systems for the Identification of Components of Mammalian Cells as Targets for Therapeutic Agents
Paper:INVASIVE SPECIES PREDICTIVE SOFTWARE
Human-readable:"The procedures represent a new process in predicitng, analyzing, and strategizing for combat of species invasions. It uses a novel combination of tools drawn from the emerging field of bi...
And that's just on the first page of "Information Technology"
Agree with the poster. Here's the summary for the lazy:
The iBridge Network aggregates research materials, technologies and discoveries into an online, easy-to-search forum. Through the iBridge Web site, researchers and commercial end users can find what they need by using community tagging and open interfaces -- and then obtain the materials via e-commerce.
Sounds like the put up a bunch of research paper links, allow tagging, search and a forum and then once you click on the link -- and this is the best part -- you can obtain the material via "e-commerce", or by paying for it. I presume the "translated academic speak" feature is tagging and forums, which is hardly earth-shattering.
Also, there is a lot of research out there that simply cannot be reliably translated into lay speak. You can't take some research papers and condense it to "so yeah, send 12k of data through the pipe for best performance". There's a reason that academic papers are complicated, and believe me, it's not to confuse the public. And the papers on which you can translate reliably into layspeak are probably shit in the first place.
There's another site where you can access research papers (largely for free), citation lists, bibtex entries: Google scholar. Also, CiteSeer. Sure, there's no forums, but then there's Usenet and the age-old technique of email-the-author (actually works sometimes).
Absolutely. Also, even though you have 60 cores in parallel, each core is running your good ol' vanilla serial code. So at the end of day, you take your Turing machine, slap in a sync/communication library like MPI or PVM and before you know it, your 60 cores are chugging along happily.
Also, non-deterministic Turing machines are not really good models for parallel computation as we know it. Simulating non-determinism exactly would require an exponential number of cores. Nevertheless, there are other computational models that deal with parallelism in the way we know it. My favourite complexity class in this respect is the bizzarely-named NC (Nick's class).
Is your bank Sovereign Bank in the states, by any chance? Their online authentication system was Social Security / 4-digit pin (same as your ATM pin). I got so sick of it that I changed to Chase.
I never said the prize motivated Perelman. Thus the poster's comment was an implied rebuttal to my own. While his statement was ironic (as it claimed to be) -- and I have no problem with irony and even agree with his statement -- it was as a whole based on the premise that I had implied Perelman "did it for the prize".
No, because you're asking for silly evidence, along the lines of "Can you prove that any politician has ever claimed to be in politics for the kickbacks (before being indicted)?" You're right about one thing though -- they'd be stupid to admit it to the Nobel committee. I'm also not saying that winning the prize motivates individuals to go for the prize 100%. What your original comment claimed is that the promise of the Prize (with capital P), or any prize or recognition for that matter, has absolutely no part to play in the endeavors of "true" scientists. In reality, I suspect there's a middle ground: where good scientists are indeed motivated by the reward of personal accomplishment and advancement, but also --- in some --- by the need for recognition. And prizes offer just that.
While I can't provide the evidence you ask for, there are a LARGE number of examples of scientists pulling underhanded stunts to appropriate enough credit to look good for a prize committee, or hell even a grant committee. You'd be naive to think otherwise. These aren't "bad" scientists -- their work is often excellent and advantageous to the field at large. It's just that their motivation, or social methods, might be less than pure. Have you read Watson's "The Double Helix"? He comes across as quite a bastard, and he wrote the book himself. The bottom line is that scientists -- good or bad -- are humans, and suffer from the same shortcomings that everyone else does, and that includes the need to be recognized sometimes.
This isn't offered as hard evidence, but read Carl Djerassi's book "Cantor's Dilemma" sometime for a plausible look at science (he's the biochemist who invented the birth control pill, also writes fiction on the side).
The point about Perelman was to illustrate that breakthroughs can indeed come from a single person in math/TCS. I understanding not reading the article, but not reading even the comments?
Plus, a contest provides a standardized (usually very difficult) dataset, and standardized metrics for evaluating the result. Provides a fair comparison of everyone's bright idea on a given real-world dataset.
Prizes are interesting for 'gold-hunting' pseudo-scientists, not for the actual hard working *REAL* scientists.
Just wanted to point out a slight flaw in your idealistic view of science and academia. We'd all LIKE it to be that way, but perhaps you've heard of one other prize that motivates some of the most brilliant scientists in the world in many fields? People spend their whole careers trying to get this prize, not just for the money but for the validation. Say what you will, but very few scientists have shrugged off the Nobel Prize as the goal of "gold-hunting pseudo-scientists".
Finally, in theoretical computer science and mathematics, it IS still possible for one person or a very small group to come up with a breakthrough. The Poincare conjecture was recently solved largely due to the efforts of a single mathematician. There are other examples, but TCS/math are not as vastly invested in massive research groups as say, particle physics.
Not to mention the fact that a million buckaroos would make even the most Bohemian/isolationist academic think "hmm...how could I put that kind of money to good use?"
The leading team changes every few days in the Netflix prize. For the longest time, it was a guy from U Toronto called NIPS Reject, then it was the whole ML team at the same uni, then it was wxyzconsulting.com, and now it's Team Gravity. It's come to the point where successive improvements are incremental and hardly significant over the previous leader. What should be interesting now is if anyone has the big breakthrough that actually wins the prize.
Check out the actual Netflix leaderboard
Hmm...now that you put it in the context of plants, I'm inclined to agree with you. Of course, given the nature of hackers, it's more likely that possession would become a tax break.
Let's keep our fingers crossed... maybe some day in the near future we can express constitutions or other legal code in some sort of unambiguous formal language. Then we can finally replace lawyers with algorithms and trained data entry technicians. Sure, we could still have judges to validate the reasoning, but I'd rather trust a formal proof of why I'm right or wrong rather than the whims of someone whose wife could have left him the very morning that I'm up for trial.
They think they are "cool" and "hip," they don't care about
Seriously, who uses the word "hip" anymore? This reminds me of a scene from Fear and Loathing in Las Vegas (book & movie) during the D.A. conference and the talk about drugs.
Nobel winners - that's a MASSIVE sample size, eh? Especially when comparing against the general population. This sounds NOT like cheesy made-for-CNN sensationalism.
Slashdot Mirror
Actually Viagra was invented to treat angina, at which it was a spectacular failure. The better-known use of Viagra was actually a side-effect that appeared in (if I remember) 80% of test subjects. So even Viagra was a sort of accident.
Remember that a lot of algorithms can be parallelized to use any number of cores (it gets inefficient after a point, but there definitely is an initial speedup).
Feng-Gui
When I first visited the site, they had a porn site in their "Sample heatmaps" section, and I must say it was pretty spot-on.
So PatRobertsonFreeVideoUpload.com has no right to take down a Muslim video posted on their site? ChristianTube.com has no right to take down a Church of Satan video posted on their site? While I agree with what you're generally saying, as someone else pointed out earlier, posting absolutely any video you like on YouTube is not a "right", and discrimination applies when you are denied your rights. In Google's opinion -- and this is one that will hold up in court -- the video was potentially inflammatory. Sure, so was the Christian one by the same definition, but good luck to you if you're going to try to sue out hypocrisy from corporations.
Completely agree, but I think the key word in your post is "religious grounds should"... i.e. "should" as in a guideline, but not a law. I don't see how you could legally bar Google -- which has done nothing unquestionably illegal -- from taking down a video posted on their private site.
YouTube reserves the right to remove Content and User Submissions without prior notice.
Bottom line: they don't NEED a reason. It's completely within their rights to delete your videos even if they don't like your username.
This reminds me of the Enron e-mail data that was released, with similarly "shocking" emails. Actually, in the Enron case, they really were illuminating because a lot of e-mails addressed to Ken Lay towards the end of the company's life included the words "you bastard". Also, you didn't have to look very hard to find rampant corporate nepotism (Ken Lay's daughter Elizabeth pimping her friends). The original dataset is at CMU, and a web-browsable version is at enronemail.com, although you have to register for the latter one. The first link lets you download the zipped contents of a bunch of executive's email boxes (sent items, deleted items, inbox, etc.)...it's really nuts.
Where are you getting this from? I speak as an academic, and I have found very few academics who think like this, and they definitely aren't respected either in industry or academia. Very few academics will conclude that one technique, much less one paper or operator, has solved any significant problem.
Paper:Apparatus and Device for Analyzing Nerve Conduction
Human-readable summary: an improved device for evaluating the performance of nerve or muscle
Paper:Methods and Systems for the Identification of Components of Mammalian Cells as Targets for Therapeutic Agents
Human-readable: Methods and Systems for the Identification of Components of Mammalian Cells as Targets for Therapeutic Agents
Paper:INVASIVE SPECIES PREDICTIVE SOFTWARE
Human-readable:"The procedures represent a new process in predicitng, analyzing, and strategizing for combat of species invasions. It uses a novel combination of tools drawn from the emerging field of bi...
And that's just on the first page of "Information Technology"
The iBridge Network aggregates research materials, technologies and discoveries into an online, easy-to-search forum. Through the iBridge Web site, researchers and commercial end users can find what they need by using community tagging and open interfaces -- and then obtain the materials via e-commerce.
Sounds like the put up a bunch of research paper links, allow tagging, search and a forum and then once you click on the link -- and this is the best part -- you can obtain the material via "e-commerce", or by paying for it. I presume the "translated academic speak" feature is tagging and forums, which is hardly earth-shattering.
Also, there is a lot of research out there that simply cannot be reliably translated into lay speak. You can't take some research papers and condense it to "so yeah, send 12k of data through the pipe for best performance". There's a reason that academic papers are complicated, and believe me, it's not to confuse the public. And the papers on which you can translate reliably into layspeak are probably shit in the first place.
There's another site where you can access research papers (largely for free), citation lists, bibtex entries: Google scholar. Also, CiteSeer. Sure, there's no forums, but then there's Usenet and the age-old technique of email-the-author (actually works sometimes).
Also, non-deterministic Turing machines are not really good models for parallel computation as we know it. Simulating non-determinism exactly would require an exponential number of cores. Nevertheless, there are other computational models that deal with parallelism in the way we know it. My favourite complexity class in this respect is the bizzarely-named NC (Nick's class).
Is your bank Sovereign Bank in the states, by any chance? Their online authentication system was Social Security / 4-digit pin (same as your ATM pin). I got so sick of it that I changed to Chase.
I absolutely cannot. In fact, I've changed my signature to more accurately reflect my feelings. Thank you, AC.
I never said the prize motivated Perelman. Thus the poster's comment was an implied rebuttal to my own. While his statement was ironic (as it claimed to be) -- and I have no problem with irony and even agree with his statement -- it was as a whole based on the premise that I had implied Perelman "did it for the prize".
While I can't provide the evidence you ask for, there are a LARGE number of examples of scientists pulling underhanded stunts to appropriate enough credit to look good for a prize committee, or hell even a grant committee. You'd be naive to think otherwise. These aren't "bad" scientists -- their work is often excellent and advantageous to the field at large. It's just that their motivation, or social methods, might be less than pure. Have you read Watson's "The Double Helix"? He comes across as quite a bastard, and he wrote the book himself. The bottom line is that scientists -- good or bad -- are humans, and suffer from the same shortcomings that everyone else does, and that includes the need to be recognized sometimes.
This isn't offered as hard evidence, but read Carl Djerassi's book "Cantor's Dilemma" sometime for a plausible look at science (he's the biochemist who invented the birth control pill, also writes fiction on the side).
The point about Perelman was to illustrate that breakthroughs can indeed come from a single person in math/TCS. I understanding not reading the article, but not reading even the comments?
Plus, a contest provides a standardized (usually very difficult) dataset, and standardized metrics for evaluating the result. Provides a fair comparison of everyone's bright idea on a given real-world dataset.
Just wanted to point out a slight flaw in your idealistic view of science and academia. We'd all LIKE it to be that way, but perhaps you've heard of one other prize that motivates some of the most brilliant scientists in the world in many fields? People spend their whole careers trying to get this prize, not just for the money but for the validation. Say what you will, but very few scientists have shrugged off the Nobel Prize as the goal of "gold-hunting pseudo-scientists".
Finally, in theoretical computer science and mathematics, it IS still possible for one person or a very small group to come up with a breakthrough. The Poincare conjecture was recently solved largely due to the efforts of a single mathematician. There are other examples, but TCS/math are not as vastly invested in massive research groups as say, particle physics.
Not to mention the fact that a million buckaroos would make even the most Bohemian/isolationist academic think "hmm...how could I put that kind of money to good use?"
The leading team changes every few days in the Netflix prize. For the longest time, it was a guy from U Toronto called NIPS Reject, then it was the whole ML team at the same uni, then it was wxyzconsulting.com, and now it's Team Gravity. It's come to the point where successive improvements are incremental and hardly significant over the previous leader. What should be interesting now is if anyone has the big breakthrough that actually wins the prize. Check out the actual Netflix leaderboard
Hmm...now that you put it in the context of plants, I'm inclined to agree with you. Of course, given the nature of hackers, it's more likely that possession would become a tax break.
Let's keep our fingers crossed ... maybe some day in the near future we can express constitutions or other legal code in some sort of unambiguous formal language. Then we can finally replace lawyers with algorithms and trained data entry technicians. Sure, we could still have judges to validate the reasoning, but I'd rather trust a formal proof of why I'm right or wrong rather than the whims of someone whose wife could have left him the very morning that I'm up for trial.
Seriously, who uses the word "hip" anymore? This reminds me of a scene from Fear and Loathing in Las Vegas (book & movie) during the D.A. conference and the talk about drugs.
Nobel winners - that's a MASSIVE sample size, eh? Especially when comparing against the general population. This sounds NOT like cheesy made-for-CNN sensationalism.
I think implicit in the system is one casualty, if the sniper is effective. But that's better than two casualties or more.