Yeah. The upshot of a (much) shorter term of coverage would mean that "false positives" from USPTO wouldn't have as much effect. Also there would be more true positives anyway, since we'd have something interpolating between "design patent" and "real patent". All this should be conducive to drastically shortening judgment times, and also I get the sense that most anti-software patent people are bringing up cases which rightfully should have ended years ago.
You seriously wouldn't have a problem, in principle, with going down to 2 or 3 years? Are you an outlier? Why haven't we done it already? It seems obvious.
What do you think of moving to first to file?
I don't think that comparing us to China, and speculating about causes, is particularly meaningful.
Since you seem to be savvy and serious, what's your take on the 17-ish year timeframe? Isn't it an anachronism of great proportions, left over from when it took that long for an invention to make it by stagecoach across a country?
We've implicitly extended patent time, when measured against the background progress of the scope in which the patents are granted. Shouldn't we bring software patents down to cover maybe at most 1/2 of a "generation" of technology, i.e. 2-3 years? This could be tuned for different fields of technology, as well as perhaps different inventions, in line with the spirit of "nonobviousness." Neat ideas should get enough cover to exercise a first-mover advantage, and nothing more, maybe a few months? Ideas which solve a problem that's been under serious attack for years, should get a few years. &c.
Human brains work associatively. If I am accused of helping a terrorist, and I start thinking about all the terrorists I've heard about or studied, then what happens? What if I'm (undiagnosed) paranoid? Or what if I know that the man you're showing me a picture of has committed tax fraud, but not that he's a terrorist?
Look, the "study participants" in these experiments are often naïve psych undergrads being paid $8 an hour. They are just not representative of the weird, convoluted and often flawed minds that make up the adult population. We need real controls drawn from real-world populations.
I just want more robust testing than these tinkertoy "pretend you're a terrorist" scenarios. Hell they didn't even have a real control.
She supposedly had strings pulled by her father, a government nuclear scientist.
Some people say it's (mostly) a hoax and that she just brought along a photo-op helmet on the standard tour for state officials: http://www.uer.ca/forum_showthread_archive.asp?threadid=8951. The denials of this claim are so weak and evasive that I tend to believe it is a hoax.
The "reason" here may amount simply to the fact that the `suspect' had the city involved in the attack on his mind for whatever reason.
There's no sense in thinking that the p300 signal won't trigger for spurious reasons such as expert knowledge in bioweapons domain. I'm thinking for example of the late Dr. Ivins here...
Once a witch-hunt is under way, we can't expect evidence to be evaluated rationally.
With an 83% accuracy rate the odds of all suspects correctly identifying no targets is 0.83^50 = 8.99 x 10^-5 = 0.00899%.
There is no information whatsoever about specificity (rate of recognizing negatives as negatives), so your extrapolation is not valid. The false positive rate may be anywhere from 0% through 100%.
This is why "accuracy" is often a worse-than-useless statistic.
The poseurs and morons (like the hotdog stand owners and other angry rubes who are deliberately set up by American intelligence for the sake of budget justification and media fluff)
Wait, what? I pass by a lot of hotdog stands; what proportion are on the FBI/DHS dole?
Information theory is a mathematical science which has been very well tested in its basic principles. It was only by applying principles derived from information theory that our modern communication devices could be developed.
Yeah, but saying that the p300 "measures" entropy in the brain is pseudoscience of the highest order. It may be true (in some sense, the formulation of which would be highly nontrivial) and it's probably false.
The 10/12 is a new study "without advanced knowledge of the `terrorist' plans," whereas the 100% was with this knowledge. The presence of this distinction should further set the stage for skepticism about their experimental design.
Further, they have achieved only 100% (resp. 83%) sensitivity (=true positives) with an unknown (or unreported) specificity (=true negatives) since they had no controls. What if I'm having an affair or high-stakes slightly-shady business deal in New York?
The rules are clear, it's just that there are no reasonable expectations since software happened to hit the big time during an economically- and technologically-undiscerning period of human history.
You can reverse the charges if the product doesn't conform to reasonable expectations and is not sold "as is". I did this when I bought a used thinkpad that didn't even POST, and the seller refused to communicate with me. To clarify: it was not sold "as is", and the seller did not even try to disclaim the implied warranty of merchantability. Then again, probably most people expect anti-virus to not work anyway.:-/
The strippers may be more contentious, but if they actually had misleading photographs on display... Most people would probably not try though.;-)
I'll be happy to sell you my mod points and a subscription to a series of pamphlets detailing many "life hacks" including my patent-pending technique for obtaining 15 mod points a week; and how to get free product out of those 25-cent bubblegum dispensers at shopping malls. Please post your credit card number; verification number; and billing address in a reply.
It's like the old saw about asking a Christian, "Is it okay to smoke while praying?" versus "Is it okay to pray while smoking?"
Of course competent programmers should handle C and all that. The point is, however, that the new ground to be discovered will be done by having non-programmers be able to quickly, easily, and accurately practice their craft with the aid of numerical analysis and data processing. It's the difference between doing something in an afternoon and needing to incorporate a company; pitch ideas; apply for grants/labs; &c.
The cynical programmer will say that the easy languages will inevitably allow fuckups. Well, that's true I guess (although isn't it an interesting project to reduce these?), but in any real project there are already several categories of fuckups (often methodological/statistical) which the programmer is, usually, blissfully unaware of, and will screw the data/results on a much grander and more subtle scale.
Here is a great example of an "empowering" language: http://processing.org/ Yeah, a `true programmer' may call it inefficient, but if it allows someone to do what they previously could not even conceive of, isn't this an infinite gain in efficiency?
I'm exclusively a ubuntu user and I, both unfortunately and certainly, would not recommend ubuntu to someone new with computers.
Wireless is flaky. Sound is (sometimes) touchy. Updates sometimes kill X.org or other crucial components. The official way to simulate a mouse wheel in gnome doesn't install by default (?!) and even after installing it, it won't persist after a suspend. This is all with a thinkpad x61s, which is one of the better laptops for linux.
OpenOffice sucks, and telling someone that they would need to pirate and/or buy Crossover Office and MSOffice to "easily" run a decent office suite is embarrassing and awkward, especially since it comes after you tell them that repositories are a good idea in general (which yeah, they are).
Finally, I just want to say that the first thing I add to the panel on a fresh install is the Force Quit applet. There's a reason for this, but even this isn't enough. I also need to writeup and add scripts to run stuff like "killall -r flash" and "killall -r chromium".
Could you please give the cite for that effect? I've tried them to no apparent effect, but I'd be very interested in trying again if there's actual evidence. Thanks.
The patent-holder could then sue the users. Even though you're free to communicate the code, you still aren't free to implement it (except on a very limited basis to "try out" the patent in consideration of licensing it).
The fact that source code has a communication aspect means you can distribute it, but you still can't make use of the patent for free, whether compiled or interpreted.
Slashdot Mirror
Yeah. The upshot of a (much) shorter term of coverage would mean that "false positives" from USPTO wouldn't have as much effect. Also there would be more true positives anyway, since we'd have something interpolating between "design patent" and "real patent". All this should be conducive to drastically shortening judgment times, and also I get the sense that most anti-software patent people are bringing up cases which rightfully should have ended years ago.
You seriously wouldn't have a problem, in principle, with going down to 2 or 3 years? Are you an outlier? Why haven't we done it already? It seems obvious.
What do you think of moving to first to file?
I don't think that comparing us to China, and speculating about causes, is particularly meaningful.
Since you seem to be savvy and serious, what's your take on the 17-ish year timeframe? Isn't it an anachronism of great proportions, left over from when it took that long for an invention to make it by stagecoach across a country?
We've implicitly extended patent time, when measured against the background progress of the scope in which the patents are granted. Shouldn't we bring software patents down to cover maybe at most 1/2 of a "generation" of technology, i.e. 2-3 years? This could be tuned for different fields of technology, as well as perhaps different inventions, in line with the spirit of "nonobviousness." Neat ideas should get enough cover to exercise a first-mover advantage, and nothing more, maybe a few months? Ideas which solve a problem that's been under serious attack for years, should get a few years. &c.
Human brains work associatively. If I am accused of helping a terrorist, and I start thinking about all the terrorists I've heard about or studied, then what happens? What if I'm (undiagnosed) paranoid? Or what if I know that the man you're showing me a picture of has committed tax fraud, but not that he's a terrorist?
Look, the "study participants" in these experiments are often naïve psych undergrads being paid $8 an hour. They are just not representative of the weird, convoluted and often flawed minds that make up the adult population. We need real controls drawn from real-world populations.
I just want more robust testing than these tinkertoy "pretend you're a terrorist" scenarios. Hell they didn't even have a real control.
Ah, a taste for the classics.
But it's aged fairly well for being at least six years old.
Courtesy of the inimitable Emo Philips.
She supposedly had strings pulled by her father, a government nuclear scientist.
Some people say it's (mostly) a hoax and that she just brought along a photo-op helmet on the standard tour for state officials: http://www.uer.ca/forum_showthread_archive.asp?threadid=8951. The denials of this claim are so weak and evasive that I tend to believe it is a hoax.
they'll at least have reason
The "reason" here may amount simply to the fact that the `suspect' had the city involved in the attack on his mind for whatever reason.
There's no sense in thinking that the p300 signal won't trigger for spurious reasons such as expert knowledge in bioweapons domain. I'm thinking for example of the late Dr. Ivins here...
Once a witch-hunt is under way, we can't expect evidence to be evaluated rationally.
With an 83% accuracy rate the odds of all suspects correctly identifying no targets is 0.83^50 = 8.99 x 10^-5 = 0.00899%.
There is no information whatsoever about specificity (rate of recognizing negatives as negatives), so your extrapolation is not valid. The false positive rate may be anywhere from 0% through 100%.
This is why "accuracy" is often a worse-than-useless statistic.
The poseurs and morons (like the hotdog stand owners and other angry rubes who are deliberately set up by American intelligence for the sake of budget justification and media fluff)
Wait, what? I pass by a lot of hotdog stands; what proportion are on the FBI/DHS dole?
Information theory is a mathematical science which has been very well tested in its basic principles. It was only by applying principles derived from information theory that our modern communication devices could be developed.
Yeah, but saying that the p300 "measures" entropy in the brain is pseudoscience of the highest order. It may be true (in some sense, the formulation of which would be highly nontrivial) and it's probably false.
The 10/12 is a new study "without advanced knowledge of the `terrorist' plans," whereas the 100% was with this knowledge. The presence of this distinction should further set the stage for skepticism about their experimental design.
Further, they have achieved only 100% (resp. 83%) sensitivity (=true positives) with an unknown (or unreported) specificity (=true negatives) since they had no controls. What if I'm having an affair or high-stakes slightly-shady business deal in New York?
i hesitate to ask here, but: what's a clam?
The rules are clear, it's just that there are no reasonable expectations since software happened to hit the big time during an economically- and technologically-undiscerning period of human history.
how would adding and removing words from a work of fiction be correcting it?
You can reverse the charges if the product doesn't conform to reasonable expectations and is not sold "as is". I did this when I bought a used thinkpad that didn't even POST, and the seller refused to communicate with me. To clarify: it was not sold "as is", and the seller did not even try to disclaim the implied warranty of merchantability. Then again, probably most people expect anti-virus to not work anyway. :-/
The strippers may be more contentious, but if they actually had misleading photographs on display... Most people would probably not try though. ;-)
I'll be happy to sell you my mod points and a subscription to a series of pamphlets detailing many "life hacks" including my patent-pending technique for obtaining 15 mod points a week; and how to get free product out of those 25-cent bubblegum dispensers at shopping malls. Please post your credit card number; verification number; and billing address in a reply.
It's like the old saw about asking a Christian, "Is it okay to smoke while praying?" versus "Is it okay to pray while smoking?"
Of course competent programmers should handle C and all that. The point is, however, that the new ground to be discovered will be done by having non-programmers be able to quickly, easily, and accurately practice their craft with the aid of numerical analysis and data processing. It's the difference between doing something in an afternoon and needing to incorporate a company; pitch ideas; apply for grants/labs; &c.
The cynical programmer will say that the easy languages will inevitably allow fuckups. Well, that's true I guess (although isn't it an interesting project to reduce these?), but in any real project there are already several categories of fuckups (often methodological/statistical) which the programmer is, usually, blissfully unaware of, and will screw the data/results on a much grander and more subtle scale.
Here is a great example of an "empowering" language: http://processing.org/ Yeah, a `true programmer' may call it inefficient, but if it allows someone to do what they previously could not even conceive of, isn't this an infinite gain in efficiency?
I for one wouldn't, although I also don't expect you to believe that. Can't speak for other slashdotters since I don't know any of them.
I do use facebook, and with no illusions; only ignorance about the fine details about what exactly is shared.
Nothing here is particularly more amusing than the rest of humanity's sordid and nuanced lives
http://www.businessinsider.com/well-these-new-zuckerberg-ims-wont-help-facebooks-privacy-problems-2010-5
I'm exclusively a ubuntu user and I, both unfortunately and certainly, would not recommend ubuntu to someone new with computers.
Wireless is flaky. Sound is (sometimes) touchy. Updates sometimes kill X.org or other crucial components. The official way to simulate a mouse wheel in gnome doesn't install by default (?!) and even after installing it, it won't persist after a suspend. This is all with a thinkpad x61s, which is one of the better laptops for linux.
OpenOffice sucks, and telling someone that they would need to pirate and/or buy Crossover Office and MSOffice to "easily" run a decent office suite is embarrassing and awkward, especially since it comes after you tell them that repositories are a good idea in general (which yeah, they are).
Finally, I just want to say that the first thing I add to the panel on a fresh install is the Force Quit applet. There's a reason for this, but even this isn't enough. I also need to writeup and add scripts to run stuff like "killall -r flash" and "killall -r chromium".
Homicide is the most generic term for one person being directly involved in the death of another. Manslaughter is homicide.
Could you please give the cite for that effect? I've tried them to no apparent effect, but I'd be very interested in trying again if there's actual evidence. Thanks.
Posting your code online is inducing others to infringe.
No it isn't.
The patent-holder could then sue the users. Even though you're free to communicate the code, you still aren't free to implement it (except on a very limited basis to "try out" the patent in consideration of licensing it).
The fact that source code has a communication aspect means you can distribute it, but you still can't make use of the patent for free, whether compiled or interpreted.