A quibble: When used indoors in a house that is heated electrically (a bad idea in terms of efficiency, I've heard), are incadescent bulbs at all inefficient? I would guess that all of the inefficiency translates to waste heat. But the waste heat heats the house. Granted, they're inefficient for the part of the year where there is no need for heating, and even worse when air conditioning is needed... And nobody changes bulbs by the season.:-)
I don't have a "fancy media PC", but with practice it takes no time at all to take our home laptop (a low-end Dell), plug an S-Video cable into the end of it (the other end of the S-Video cable is always plugged into the TV), unplug the speakers from the DVD player, plug the speakers into the side of the laptop, and press fn-F2. A remote control would be nice, but it works fine without it, too. We've done a lot of Netflix "Watch Instantly" watching this way. Hulu works, too.
It's worth noting that if you post the code, it will get into Open Source code search engines, and then someone searching for, say, a particular subroutine, or a newbie trying to figure out how to do something (perhaps something very simple), will be able to benefit and either use or learn from the code.
I've posted a lot of very messy code that does various Palm-related things. I get a lot more downloads of binaries than source typically, but nobody's complained that the code is messy, ugly or whatever. It's a gift horse, after all, and in my experience everybody understands that. Put a disclaimer about the code being messy if you like.
My advice is: Swallow your pride and post it. Nobody loses (unless you can make some money from its staying closed), and someone might gain.
The statement "Science finds every soil barren in which miracles are taken literally and seriously and revelation is considered to provide authentic knowledge of the physical world" is really, really odd given that modern science started in Christian countries where miracles were taken literally and revelation was considered to provide authentic knowledge of the physical world.
It has been argued that it was seeing the universe as created by a rational being enamoured of order (in particular, mathematical order) who created us to know both him and the universe that produced the confidence in reason and observation that was needed for the scientific enterprise. (Why else should we suppose in the first that there should be discoverable deep, non-accidental patterns beneath surface disorder?)
I am not an Islamic studies scholar, but my understanding is that Islam, historically, has been less willing to agree that God made us to know the truth (there are Qu'ranic passages about God deceiving people) or that he endowed the universe with a rational structure knowable by us. Moreover, while occasionalism--the view that there is no real physical causation, but instead God does everything directly (so that it is not the fire under the kettle that causes the water to heat up, but rather God, on the occasion of the fire, makes the water heat up)--was very much a minority view in the West, it has been a much stronger force in Islam. And occasionalism can make one very skeptical about the prospects of understanding how things work in the world, unless one has a view on which God's actions have clearly discernible patterns, and I don't think Islam is that committed to this.
I wonder if they have the proper mpeg-2 visual patent licenses for mpeg2enc. They may be caught in a bind. If they obtain the patent licenses for mpeg2 encoding, then they may be violating the GPL since they are not allowing their users to pass the patent licenses on (they can't allow that, as the mpeg2 encoding license won't allow them to allow them that). And if they don't obtain the patent licenses, they're likely to get sued. Since I suspect they're more likely to get sued by someone with money for good lawyers for patent violation than for GPL violation, they may be making a shrewd--though immoral and illegal--decision to pay for the patent licenses but to violate the GPL.
Well, I can't distribute to other people, because the codecs are GPL and it would probably be a GPL violation for me to distribute them given the patent restrictions that I am under (one of these was that I had to include a patent message with the software). Fortunately GPL2 doesn't have any restrictions on one's own use, so I don't have to worry about this.
As for a HOWTO, well I just went to www.mpegla.com, and sent a very polite email asking how to get an MPEG-4 VISUAL license. Since they have (or had--this was in '05) non-discriminatory licensing, they are required to send one a license agreement even if one is only "manufacturing" half a dozen copies. They were very polite, sent me a thick booklet that they sent me by FedEx (even though they knew that I would only be making a few copies of the encoder/decoder), I signed it and mailed it back, and every year they send me a spreadsheet where I fill out how many copies I sold. ("Sold" is defined in the license in such a way that money doesn't have to change hands.) I just keep track of the number of times I install a build on some device I own (new versions require new licenses), and fill that out. I think I've probably used up 20 of my free licenses. (I think the 100,000 may have been a mistake: I am entitled to 50,000 encoders and 50,000 decoders.)
I wonder what they would do if everybody did this? Would they at least stop FedEx'ing the forms? Would they make an easy web form? Or would they maybe start up some minimum fee like $1000 to keep out folks like me, but that wouldn't matter to for-profit developers? It's quite ridiculous, really, the costs for them given that I will never reach the maximum of free copies.
My mailbox still has my emails:
From me:
"Dear Sir/Madam:
I was wondering whether I automatically qualify for your "no royalties" for under 50,000 units deal if I simply generate a couple of copies (no more than three or four) of a software MPEG4 visual encoder and of a software MPEG4 visual decoder (by adapting and compiling freely available source code) for the personal use of myself and my immediate family (rather than for distribution), or whether there is something I need to sign with you.
Thanks for any help! [my name]"
From them:
"Dear Mr. [my name],
"Thank you for your inquiry below. We appreciate your interest in assuring that your MPEG-4 Visual intellectual property responsibilities are properly addressed. Therefore, please allow me to briefly explain how our MPEG-4 Visual Patent Portfolio License works which I believe will be helpful in your further understanding.
"Under the MPEG-4 Visual License, coverage is generally provided and royalties are paid on end products that include MPEG-4 Video Encoders/Decoders. In the normal course, the party that manufactures and/or offers such end products for Sale is responsible for paying the applicable royalty. Therefore, as I understand from your explanation that you are creating a product that will have MPEG-4 video encoding and/or MPEG-4 video decoding functionality, such a product would benefit from coverage under our MPEG-4 Visual License. For that purpose, we would welcome you to become a Licensee.
"Meanwhile, as you point out, no royalty is payable for the manufacture and sale of the first 50,000 MPEG-4 Video Encoders and the first 50,000 MPEG-4 Video Decoders in each Calendar year. Therefore, it sounds from your description that you would produce volumes that would fall well below the annual minimum thresholds, and as a result, no royalty would be paid for the manufacture and sale of such units.
"If you care to provide us with your physical mailing address, we will be happy to send you a hard copy of our MPEG-4 Visual License for your review and signature. The process for becoming a Licensee is very easy. Once you have the materials, all you will need to do is fill in the necessary information where indicated, have the documents signed, and return the signed documents for execution by MPEG
I actually got an MPEG LA license to manufacture divx players/encoders. The license charge is $0 for the first 100,000 units. I then compile the codecs and players (with some local modifications), and give them away to myself . Every year I need to send in a report of how many I've sold.
This doesn't take care of mp3, but my understanding is that Thomson doesn't mind people using free mp3 decoders (they said something about that years ago) as long as they don't sell them (that's GPL incompatible, I know).
I am not a lawyer, but this stuff seems to me to be fraud, if the company knows that they are making false statements. Suppose I set up a booth by the side of a public road and put up a sign that says that by federal law you must either take a different road or negotiate a passage fee with me. That's fraud, and surely I'd be rightly arrested for it. Well, in the case of false copyright statements, someone is telling us that by federal law they must either refrain from copying or negotiate a license fee with them. If the company knows the statements to be false, that seems to me like fraud with an intention either to restrain competition (by preventing copying) or to profit directly (by selling license fees).
Criminal prosecution doesn't seem likely, though, because it would be hard to prove beyond a shadow of reasonable doubt that the company knew the statements were fraudulent. Maybe, though, a class action lawsuit alleging that the company made statements that they should have known are false and made them in order to restrain competition or profit would be possible, the afflicted class being all the people who bought the item and who refrained from exercising their legal rights to copy as a result of the false statements. Maybe this can't be done--I am not a lawyer.
My personal bugbear are the false statements at the beginnings of videos which say that all copying is prohibited. No, that is false in the U.S., since only copying in excess of fair use is prohibited. Likewise, statements that only home use is permitted are false in the U.S., since copyright law explicitly allows the showing of films for instructional purpose in a non-profit educational institution. I suppose they might try to weasel out of this by saying that the company, not the law, is doing the prohibiting. Seems shaky to me, but I am not a lawyer.
Is there any law against offering legal advice that one knows or should know to be inaccurate?
That is how Offset Cursor works. If you read the article, Shift is much better--it targets the area under the finger, but shows a circular callout of that area above the finger so it is not occluded. A variant even magnifies the area for higher precision. It actually looks really nice from the paper. I'd be tempted to make a PalmOS implementation, but I suspect it's being patented.
Now if only they could solve the problem of screens getting smudged by fingers.:-)
The IMAX 3D films I have seen have used linearly polarized glasses. As a result, if you tilt the plane of your eyes (or more precisely of your glasses) away from the horizontal, you get ghosting. The new thing is circularly polarized glasses which eliminates this problem.
Amusingly, Sony claims the ARccOS copyprotection system is "Fully compatible with available DVD players and drives" (http://www.sonydadc.com/products.copy.arccos.go).
With practice, it takes me about 20-30 seconds to evaluate the typical turnitin.com report. Of course, if there is something suspicious, it will take longer but that is only a small minority of cases, and if there is something very suspicious the process will take hours (meeting with student, filing a report, etc.). Overall, it tends to take me about half an hour to evaluate about fifty papers if no credible plagiarism case is there.
Thus, with practice, I think evaluating a turnitin.com report is about 3-6% of total grading time for a typical paper.
3-4 words in a row will be highlighted. But it's up to the professor's judgment whether to count that as evidence of cheating (and that judgment typically can either be appealed or else only serves as as the basis for a disciplinary hearing which might go in either direction). If the professor knows what she's doing, she's going to think about how natural the choice of words was (a natural choice of words may occur in several texts independently), how likely it was to be random, etc. Before turnitin.com, a lot of graders would take distinctive sample phrases and search the web for them. That had a much higher false negative rate, and a correspondingly lower false positive rate, but even there there was the possibility of a false positive.
In practice, 90% of the papers I get have some turnitin.com highlighting. Most of that is embedded quotes, or bibliographic entries. Obviously, I disregard those. After that, there are short strings taken from my topic question or from my own class notes, and I disregard these. Then there are a few short phrases that will naturally occur in many papers on the subject, such as natural paraphrases of the author. Again, I disregard these. There may also be combinations of three words that do not at all look like likely candidates for copying (e.g., if in this sentence, turnitin.com highlighted "do not at all", I would have a quick look at the original source to make sure there was nothing more that turnitin.com didn't notice, and if there was nothing more I'd ignore it). Typically, there is nothing left in a paper.
But sometimes there is something more. And then a time-consuming investigation is needed. The original sources given by turnitin.com need to be compared against carefully. The length of matches has to be thought about. Alternate hypotheses (such as of the copying going in the other direction, or of an innocent mistake, or of randomness) need to be considered and discounted. This is a lot of work. And of course the student needs to be met with, given a chance to give an alternate explanation.
First of all, no professor who knows anything about what they're doing is going to make a plagiarism accusation based on turnitin.com's score. For one, by default, embedded quotations are usually flagged as copied (which of course they are, but it's an acceptable form copying if duly acknowledged). Rather, the professor would, presumably, look at the source of each of the flagged portions, check if the amount copied seems too large to be chance, see whether perhaps there is some innocent mistake, and so on. And if the student is revising a draft that was earlier submitted for the same course, after looking at the source--which is hyperlinked in the report--the professor should see that it is a draft by the same author. When I use turnitin.com and get a suspected plagiarist, I write a long report to our Honor Council, carefully detailing the evidence for and against innocence, considering alternate hypotheses (e.g., if the student's text matches text by some professional scholar, I check the dates to make sure that there is no possibility that the scholar plagiarized from the student), etc. Typically, I'll also have met with the student by this point (and in every case where I've met with the student, the student admitted the wrongdoing as soon as he or she saw the turnitin.com report).
Of course, like every tool, turnitin.com can be used incorrectly. But likewise mistakes can be made without turnitin.com. It happens sometimes that a student gets accused of plagiarism because the paper looks too good. For instance, if the student was slacking on earlier papers--or just less good at that part of the course--and then hands in a really good paper, the professor is likely to be quite suspicious. If the professor relies more on turnitin.com, then there may be fewer accusations based simply on the paper looking too good. (The downside of that is that some plagiarized papers well get through.)
As for frequency, I catch about one plagiarist per one hundred students. I suspect there are probably one or two there that I don't catch.
They say the emulated system runs at 10% of the CPU speed of the emulating system. That's not bad at all. With a 1 ghz system, one gets a 100 mhz DOS system, which is more than enough!
So in an ESL textbook the list "1st, 2nd, 3rd, 4th, 5th, 7th, 8th, 9th, 10th, 11th, 12th, 13th, 14th, 15th, 16th, 17th, 18th, 19th, 20th, 21st, 22nd" is in danger of violation, because it combines two items from the list, namely "21st" and "10th"?
And the Science Fair and other well-established competitions should be careful about giving gold, silver and bronze medals, or at least calling them that?
IANAL, but I thought that for a legally binding contract, something valuable must be offered by both sides. If I tell you that I will give you $100 with no strings attached, there is no legal contract there. (Of course, I've made a promise and so I am morally bound. But that's different.)
That may be, but I hope they don't prosecute the guy who admitted this, since it would have a chilling effect on others who might like to make similar admissions.
I agree about the vagueness. I think a social networking site is defined as one that lets you maintain your own web page. I don't know what counts as "your own web page". If I start a new topic in an online forum, is that topic "a web page"? It seems to be. Is it "my own"? Well, in some sense, no, because I do not exercise complete control over it. But on the other hand, I assume facebook users don't have complete control over "their" pages.
On the other hand, about enforcibility, obviously it cannot be enforced in all cases--just about nothing can. But it is a fallacy to conclude that it cannot be enforced at all. One could require anybody who wants to join to send a notarized agreement, with the notary having examined the proof of age in the case of adults, and with a notarized parental agreement in the case of kids, with the notary having examined birth/adoption certificates. Of course, this would probably force the sites out of business, since it would be too much of a nuisance for people to sign up. But it certainly could be done. The question is whether it would be desirable to do it.
I do think that at least if kids are doing something where there is no presumption of confidentiality, it is quite reasonable that parents should be able to see it. There is no presumption of confidentiality in posting in a public arena (and that includes sites where one needs to sign up, as long as everybody is allowed to sign up; facebook is a borderline case, since only people at an educational institution can sign up). There is, however, a presumption of confidentiality in email, in IM, and in telephone conversations, and so that should be respected, except in extreme situations (just as it is OK for law enforcement to tap these in extreme situations--of course the privilege can be abused).
How friendly is the Office 2007 interface for us keyboard-centric folks? I know that they emulate menu key shortcuts, so you can do alt-f,x to exit, but I understand that it's less friendly now, because when you press alt-f with 2007 I understand you don't see the menu as a reminder of the keystrokes available. Plus I like to pull down menus by alt-key and then select options via arrow-keys and enter. I think it's often faster than mouse.
Slashdot Mirror
A quibble: When used indoors in a house that is heated electrically (a bad idea in terms of efficiency, I've heard), are incadescent bulbs at all inefficient? I would guess that all of the inefficiency translates to waste heat. But the waste heat heats the house. Granted, they're inefficient for the part of the year where there is no need for heating, and even worse when air conditioning is needed... And nobody changes bulbs by the season. :-)
Do you really lose the hyperlinking? The MOBI format certainly supports hyperlinking--it's basically a compressed subset of HTML, if memory serves.
Stanislaw Lem was Polish, not Soviet. I suspect he'd really not have liked being identified as Soviet (and that's an understatement).
I don't have a "fancy media PC", but with practice it takes no time at all to take our home laptop (a low-end Dell), plug an S-Video cable into the end of it (the other end of the S-Video cable is always plugged into the TV), unplug the speakers from the DVD player, plug the speakers into the side of the laptop, and press fn-F2. A remote control would be nice, but it works fine without it, too. We've done a lot of Netflix "Watch Instantly" watching this way. Hulu works, too.
Moreover, one can get a design patent on fonts, and MS has been getting these for the Vista fonts.
It's worth noting that if you post the code, it will get into Open Source code search engines, and then someone searching for, say, a particular subroutine, or a newbie trying to figure out how to do something (perhaps something very simple), will be able to benefit and either use or learn from the code.
I've posted a lot of very messy code that does various Palm-related things. I get a lot more downloads of binaries than source typically, but nobody's complained that the code is messy, ugly or whatever. It's a gift horse, after all, and in my experience everybody understands that. Put a disclaimer about the code being messy if you like.
My advice is: Swallow your pride and post it. Nobody loses (unless you can make some money from its staying closed), and someone might gain.
The statement "Science finds every soil barren in which miracles are taken literally and seriously and revelation is considered to provide authentic knowledge of the physical world" is really, really odd given that modern science started in Christian countries where miracles were taken literally and revelation was considered to provide authentic knowledge of the physical world.
It has been argued that it was seeing the universe as created by a rational being enamoured of order (in particular, mathematical order) who created us to know both him and the universe that produced the confidence in reason and observation that was needed for the scientific enterprise. (Why else should we suppose in the first that there should be discoverable deep, non-accidental patterns beneath surface disorder?)
I am not an Islamic studies scholar, but my understanding is that Islam, historically, has been less willing to agree that God made us to know the truth (there are Qu'ranic passages about God deceiving people) or that he endowed the universe with a rational structure knowable by us. Moreover, while occasionalism--the view that there is no real physical causation, but instead God does everything directly (so that it is not the fire under the kettle that causes the water to heat up, but rather God, on the occasion of the fire, makes the water heat up)--was very much a minority view in the West, it has been a much stronger force in Islam. And occasionalism can make one very skeptical about the prospects of understanding how things work in the world, unless one has a view on which God's actions have clearly discernible patterns, and I don't think Islam is that committed to this.
I wonder if they have the proper mpeg-2 visual patent licenses for mpeg2enc. They may be caught in a bind. If they obtain the patent licenses for mpeg2 encoding, then they may be violating the GPL since they are not allowing their users to pass the patent licenses on (they can't allow that, as the mpeg2 encoding license won't allow them to allow them that). And if they don't obtain the patent licenses, they're likely to get sued. Since I suspect they're more likely to get sued by someone with money for good lawyers for patent violation than for GPL violation, they may be making a shrewd--though immoral and illegal--decision to pay for the patent licenses but to violate the GPL.
Or they're just careless.
Well, I can't distribute to other people, because the codecs are GPL and it would probably be a GPL violation for me to distribute them given the patent restrictions that I am under (one of these was that I had to include a patent message with the software). Fortunately GPL2 doesn't have any restrictions on one's own use, so I don't have to worry about this.
As for a HOWTO, well I just went to www.mpegla.com, and sent a very polite email asking how to get an MPEG-4 VISUAL license. Since they have (or had--this was in '05) non-discriminatory licensing, they are required to send one a license agreement even if one is only "manufacturing" half a dozen copies. They were very polite, sent me a thick booklet that they sent me by FedEx (even though they knew that I would only be making a few copies of the encoder/decoder), I signed it and mailed it back, and every year they send me a spreadsheet where I fill out how many copies I sold. ("Sold" is defined in the license in such a way that money doesn't have to change hands.) I just keep track of the number of times I install a build on some device I own (new versions require new licenses), and fill that out. I think I've probably used up 20 of my free licenses. (I think the 100,000 may have been a mistake: I am entitled to 50,000 encoders and 50,000 decoders.)
I wonder what they would do if everybody did this? Would they at least stop FedEx'ing the forms? Would they make an easy web form? Or would they maybe start up some minimum fee like $1000 to keep out folks like me, but that wouldn't matter to for-profit developers? It's quite ridiculous, really, the costs for them given that I will never reach the maximum of free copies.
My mailbox still has my emails:
From me:
"Dear Sir/Madam:
I was wondering whether I automatically qualify for your "no royalties" for under 50,000 units deal if I simply generate a couple of copies (no more than three or four) of a software MPEG4 visual encoder and of a software MPEG4 visual decoder (by adapting and compiling freely available source code) for the personal use of myself and my immediate family (rather than for distribution), or whether there is something I need to sign with you.
Thanks for any help!
[my name]"
From them:
"Dear Mr. [my name],
"Thank you for your inquiry below. We appreciate your interest in assuring that your MPEG-4 Visual intellectual property responsibilities are properly addressed. Therefore, please allow me to briefly explain how our MPEG-4 Visual Patent Portfolio License works which I believe will be helpful in your further understanding.
"Under the MPEG-4 Visual License, coverage is generally provided and royalties are paid on end products that include MPEG-4 Video Encoders/Decoders. In the normal course, the party that manufactures and/or offers such end products for Sale is responsible for paying the applicable royalty. Therefore, as I understand from your explanation that you are creating a product that will have MPEG-4 video encoding and/or MPEG-4 video decoding functionality, such a product would benefit from coverage under our MPEG-4 Visual License. For that purpose, we would welcome you to become a Licensee.
"Meanwhile, as you point out, no royalty is payable for the manufacture and sale of the first 50,000 MPEG-4 Video Encoders and the first 50,000 MPEG-4 Video Decoders in each Calendar year. Therefore, it sounds from your description that you would produce volumes that would fall well below the annual minimum thresholds, and as a result, no royalty would be paid for the manufacture and sale of such units.
"If you care to provide us with your physical mailing address, we will be happy to send you a hard copy of our MPEG-4 Visual License for your review and signature. The process for becoming a Licensee is very easy. Once you have the materials, all you will need to do is fill in the necessary information where indicated, have the documents signed, and return the signed documents for execution by MPEG
I actually got an MPEG LA license to manufacture divx players/encoders. The license charge is $0 for the first 100,000 units. I then compile the codecs and players (with some local modifications), and give them away to myself . Every year I need to send in a report of how many I've sold.
This doesn't take care of mp3, but my understanding is that Thomson doesn't mind people using free mp3 decoders (they said something about that years ago) as long as they don't sell them (that's GPL incompatible, I know).
I am not a lawyer, but this stuff seems to me to be fraud, if the company knows that they are making false statements. Suppose I set up a booth by the side of a public road and put up a sign that says that by federal law you must either take a different road or negotiate a passage fee with me. That's fraud, and surely I'd be rightly arrested for it. Well, in the case of false copyright statements, someone is telling us that by federal law they must either refrain from copying or negotiate a license fee with them. If the company knows the statements to be false, that seems to me like fraud with an intention either to restrain competition (by preventing copying) or to profit directly (by selling license fees).
Criminal prosecution doesn't seem likely, though, because it would be hard to prove beyond a shadow of reasonable doubt that the company knew the statements were fraudulent. Maybe, though, a class action lawsuit alleging that the company made statements that they should have known are false and made them in order to restrain competition or profit would be possible, the afflicted class being all the people who bought the item and who refrained from exercising their legal rights to copy as a result of the false statements. Maybe this can't be done--I am not a lawyer.
My personal bugbear are the false statements at the beginnings of videos which say that all copying is prohibited. No, that is false in the U.S., since only copying in excess of fair use is prohibited. Likewise, statements that only home use is permitted are false in the U.S., since copyright law explicitly allows the showing of films for instructional purpose in a non-profit educational institution. I suppose they might try to weasel out of this by saying that the company, not the law, is doing the prohibiting. Seems shaky to me, but I am not a lawyer.
Is there any law against offering legal advice that one knows or should know to be inaccurate?
I emailed one of the researchers. Yes, the technology is being patented, unsurprisingly.
That is how Offset Cursor works. If you read the article, Shift is much better--it targets the area under the finger, but shows a circular callout of that area above the finger so it is not occluded. A variant even magnifies the area for higher precision. It actually looks really nice from the paper. I'd be tempted to make a PalmOS implementation, but I suspect it's being patented.
:-)
Now if only they could solve the problem of screens getting smudged by fingers.
The IMAX 3D films I have seen have used linearly polarized glasses. As a result, if you tilt the plane of your eyes (or more precisely of your glasses) away from the horizontal, you get ghosting. The new thing is circularly polarized glasses which eliminates this problem.
Amusingly, Sony claims the ARccOS copyprotection system is "Fully compatible with available DVD players and drives" (http://www.sonydadc.com/products.copy.arccos.go).
With practice, it takes me about 20-30 seconds to evaluate the typical turnitin.com report. Of course, if there is something suspicious, it will take longer but that is only a small minority of cases, and if there is something very suspicious the process will take hours (meeting with student, filing a report, etc.). Overall, it tends to take me about half an hour to evaluate about fifty papers if no credible plagiarism case is there.
Thus, with practice, I think evaluating a turnitin.com report is about 3-6% of total grading time for a typical paper.
3-4 words in a row will be highlighted. But it's up to the professor's judgment whether to count that as evidence of cheating (and that judgment typically can either be appealed or else only serves as as the basis for a disciplinary hearing which might go in either direction). If the professor knows what she's doing, she's going to think about how natural the choice of words was (a natural choice of words may occur in several texts independently), how likely it was to be random, etc. Before turnitin.com, a lot of graders would take distinctive sample phrases and search the web for them. That had a much higher false negative rate, and a correspondingly lower false positive rate, but even there there was the possibility of a false positive.
In practice, 90% of the papers I get have some turnitin.com highlighting. Most of that is embedded quotes, or bibliographic entries. Obviously, I disregard those. After that, there are short strings taken from my topic question or from my own class notes, and I disregard these. Then there are a few short phrases that will naturally occur in many papers on the subject, such as natural paraphrases of the author. Again, I disregard these. There may also be combinations of three words that do not at all look like likely candidates for copying (e.g., if in this sentence, turnitin.com highlighted "do not at all", I would have a quick look at the original source to make sure there was nothing more that turnitin.com didn't notice, and if there was nothing more I'd ignore it). Typically, there is nothing left in a paper.
But sometimes there is something more. And then a time-consuming investigation is needed. The original sources given by turnitin.com need to be compared against carefully. The length of matches has to be thought about. Alternate hypotheses (such as of the copying going in the other direction, or of an innocent mistake, or of randomness) need to be considered and discounted. This is a lot of work. And of course the student needs to be met with, given a chance to give an alternate explanation.
First of all, no professor who knows anything about what they're doing is going to make a plagiarism accusation based on turnitin.com's score. For one, by default, embedded quotations are usually flagged as copied (which of course they are, but it's an acceptable form copying if duly acknowledged). Rather, the professor would, presumably, look at the source of each of the flagged portions, check if the amount copied seems too large to be chance, see whether perhaps there is some innocent mistake, and so on. And if the student is revising a draft that was earlier submitted for the same course, after looking at the source--which is hyperlinked in the report--the professor should see that it is a draft by the same author. When I use turnitin.com and get a suspected plagiarist, I write a long report to our Honor Council, carefully detailing the evidence for and against innocence, considering alternate hypotheses (e.g., if the student's text matches text by some professional scholar, I check the dates to make sure that there is no possibility that the scholar plagiarized from the student), etc. Typically, I'll also have met with the student by this point (and in every case where I've met with the student, the student admitted the wrongdoing as soon as he or she saw the turnitin.com report).
Of course, like every tool, turnitin.com can be used incorrectly. But likewise mistakes can be made without turnitin.com. It happens sometimes that a student gets accused of plagiarism because the paper looks too good. For instance, if the student was slacking on earlier papers--or just less good at that part of the course--and then hands in a really good paper, the professor is likely to be quite suspicious. If the professor relies more on turnitin.com, then there may be fewer accusations based simply on the paper looking too good. (The downside of that is that some plagiarized papers well get through.)
As for frequency, I catch about one plagiarist per one hundred students. I suspect there are probably one or two there that I don't catch.
They say the emulated system runs at 10% of the CPU speed of the emulating system. That's not bad at all. With a 1 ghz system, one gets a 100 mhz DOS system, which is more than enough!
So in an ESL textbook the list "1st, 2nd, 3rd, 4th, 5th, 7th, 8th, 9th, 10th, 11th, 12th, 13th, 14th, 15th, 16th, 17th, 18th, 19th, 20th, 21st, 22nd" is in danger of violation, because it combines two items from the list, namely "21st" and "10th"?
And the Science Fair and other well-established competitions should be careful about giving gold, silver and bronze medals, or at least calling them that?
IANAL, but I thought that for a legally binding contract, something valuable must be offered by both sides. If I tell you that I will give you $100 with no strings attached, there is no legal contract there. (Of course, I've made a promise and so I am morally bound. But that's different.)
That may be, but I hope they don't prosecute the guy who admitted this, since it would have a chilling effect on others who might like to make similar admissions.
I agree about the vagueness. I think a social networking site is defined as one that lets you maintain your own web page. I don't know what counts as "your own web page". If I start a new topic in an online forum, is that topic "a web page"? It seems to be. Is it "my own"? Well, in some sense, no, because I do not exercise complete control over it. But on the other hand, I assume facebook users don't have complete control over "their" pages.
On the other hand, about enforcibility, obviously it cannot be enforced in all cases--just about nothing can. But it is a fallacy to conclude that it cannot be enforced at all. One could require anybody who wants to join to send a notarized agreement, with the notary having examined the proof of age in the case of adults, and with a notarized parental agreement in the case of kids, with the notary having examined birth/adoption certificates. Of course, this would probably force the sites out of business, since it would be too much of a nuisance for people to sign up. But it certainly could be done. The question is whether it would be desirable to do it.
I do think that at least if kids are doing something where there is no presumption of confidentiality, it is quite reasonable that parents should be able to see it. There is no presumption of confidentiality in posting in a public arena (and that includes sites where one needs to sign up, as long as everybody is allowed to sign up; facebook is a borderline case, since only people at an educational institution can sign up). There is, however, a presumption of confidentiality in email, in IM, and in telephone conversations, and so that should be respected, except in extreme situations (just as it is OK for law enforcement to tap these in extreme situations--of course the privilege can be abused).
Nevermind what I said--it looks like they do have potentially a very good keyboard-centric interface: http://blogs.msdn.com/jensenh/archive/2005/10/13/4 80568.aspx
How friendly is the Office 2007 interface for us keyboard-centric folks? I know that they emulate menu key shortcuts, so you can do alt-f,x to exit, but I understand that it's less friendly now, because when you press alt-f with 2007 I understand you don't see the menu as a reminder of the keystrokes available. Plus I like to pull down menus by alt-key and then select options via arrow-keys and enter. I think it's often faster than mouse.