You are probably thinking about Latanya Sweeney's study out of CMU. There are standards for de-identification that use her work (and others) to prevent the ability to re-identify probabilistically.
Larry Page is just complaining that Google doesn't have the data. These data already exist and are being extensively studied by researchers in academics, government, health insurers, employers, and pharmaceutical companies. The de-identified data can be licensed and analyzed by anyone. The fully identifiable information is routinely analyzed by the owners of the data.
The problem is not access to the data, the problem is that it is difficult to make valid inferences about causation from observational data.
Even at private institutions with very high tuition rates, your tuition payment represents a fraction of the costs of your education. All students are on a scholarship.
You've left out another important factor. Publishers often charge the researchers that are providing the content "page charges" to defray publication costs. Researchers are willing to pay this fee (often hundreds of dollars for a single article) because success in their field is judged in part by the reputation of the journals in which they have published.
Then you misunderstand what it is that you and I, as taxpayers, are paying for through government grants. We are paying for the immediate costs of the personnel, equipment, and overhead to carry out the work. In exchange for this support we get the benefit of the knowledge generated through the research and disseminated through peer-reviewed publications. In some specific cases, the support may include a stipulation that the government gets a piece of the intellectual property pie. However, in the vast majority of cases, the ideas for the research pre-date even the application for grant support by a large margin and the control of the ideas (and thus the intellectual property) stays with the researchers and their institutions. Licenses are paid to *use* the IP, they are not a purchase of the IP per se.
Re:How much is that in square furlongs?
on
Giant Ice Shelf Snaps
·
· Score: 2, Informative
The last several plans I have had include a "max out of pocket" clause that limits the total expenditure to something like $2-3000 per person per year. This certainly adds up, but even at the relatively high co-pay rate of 20% that you mention it is not like one faces imminent bankruptcy with a major illess. I don't know how common they are, but if there are plans out there that have no max out-of-pocket limit I would stay away.
Re:Paris Hilton or Madame Curie... hmmm
on
Top Ten Geek Girls
·
· Score: 1
Among the many insults perpetrated against Rosalind Franklin by Jim Watson and to a lesser degree Francis Crick, robbing her of the Nobel prize was not one. Watson and Crick received the award in 1962 but Rosalind Franklin died in 1958, and the Nobel is not awarded posthumously.
Kentsfield is a processor, not a chipset as the parent states. According to other articles, the Kentsfield processor will work with Intel's 965 and 975 chipset-based boards that are running Core2Duo's, providing a super-sweet upgrade path.
Folate and folate-related polymorphisms certainly play a role in the pathogensis and etiology of cancer through abnormal methylation and thymidine synthesis. That is not in dispute. However, your proposal that folic acid supplementation (adding more methyl groups) is somehow going to fix a problem related to having too much methylation just doesn't make any sense.
Folate acts as a methyl group donor. The article describes a gene that is inactivated by hypermethylation (over methylation). More folate, and therefore more available methylgroups, is not likely to solve problems caused by methylation of a tumor suppressor gene.
On the other hand, folate may be beneficial in preventing tumors that would arise from double stranded breaks in DNA caused by insufficient methyl group availability interfering with the creation of thymidine or through hypomethylation of oncogenes.
Not to complicate things too much, but most vitamin D supplements (particularly pills and milk) also contain substantial vitamin A supplementation. There is considerable information that retinol (a form of A) can antagonize many of the benefits of vitamin D. I would not be at all surprised to see vitamin manufacturers start to reformulate their products with more D and less A.
Just buying it might not be enough. You might want to take the next step and drink it as well.:-)
The real debate underlying this article surrounds the appropriate "dose" of vitamin D. The current recommendations in the US (400 IU per day) are entirely based on requirements for maintaining normal bone mineral composition. This has absolutely no relation to other biological effects of vitamin D (cellular differentiation, immune cell activity).
Whereas you can get 400 IU per day by drinking vitamin D fortified milk, full-body exposure to solar ultraviolet radiation can produce as much as 10,000 - 40,000 IU of vitamin D. In the winter and at high latitudes vitamin D production from solar UV can drop to zero. Between diet, supplements, and sun exposure, the ideal combination and target dose for cancer prevention has not been established. It is almost certainly considerably above the 400 IU that you need to maintain healthy bones.
Holick's primary appointment is in Endocrinology. Gilchrist "sptripped him" of a largely symbollic secondary appointment in Dermatology. This gave Holick publicity and made her look like a git.
Slashdot Mirror
You are probably thinking about Latanya Sweeney's study out of CMU. There are standards for de-identification that use her work (and others) to prevent the ability to re-identify probabilistically.
Larry Page is just complaining that Google doesn't have the data. These data already exist and are being extensively studied by researchers in academics, government, health insurers, employers, and pharmaceutical companies. The de-identified data can be licensed and analyzed by anyone. The fully identifiable information is routinely analyzed by the owners of the data.
The problem is not access to the data, the problem is that it is difficult to make valid inferences about causation from observational data.
Here is the story:
http://www.slate.com/articles/...
In the U.S. being "born alive" is defined by Federal Statute. http://www.law.cornell.edu/uscode/text/1/8
Some information about differences by nation is contained on page 7 of this Congressional Research Service report: http://www.fas.org/sgp/crs/misc/R41378.pdf
This is nearly a decade old, so some changes may have occurred.
Even at private institutions with very high tuition rates, your tuition payment represents a fraction of the costs of your education. All students are on a scholarship.
You've left out another important factor. Publishers often charge the researchers that are providing the content "page charges" to defray publication costs. Researchers are willing to pay this fee (often hundreds of dollars for a single article) because success in their field is judged in part by the reputation of the journals in which they have published.
How about an RFC instead of a web page?
, but The World Is Not Enough.
Then you misunderstand what it is that you and I, as taxpayers, are paying for through government grants. We are paying for the immediate costs of the personnel, equipment, and overhead to carry out the work. In exchange for this support we get the benefit of the knowledge generated through the research and disseminated through peer-reviewed publications. In some specific cases, the support may include a stipulation that the government gets a piece of the intellectual property pie. However, in the vast majority of cases, the ideas for the research pre-date even the application for grant support by a large margin and the control of the ideas (and thus the intellectual property) stays with the researchers and their institutions. Licenses are paid to *use* the IP, they are not a purchase of the IP per se.
Ah Google, what can't you do?
66 (square kilometers) = 630.89552 square furlongs
The last several plans I have had include a "max out of pocket" clause that limits the total expenditure to something like $2-3000 per person per year. This certainly adds up, but even at the relatively high co-pay rate of 20% that you mention it is not like one faces imminent bankruptcy with a major illess. I don't know how common they are, but if there are plans out there that have no max out-of-pocket limit I would stay away.
Among the many insults perpetrated against Rosalind Franklin by Jim Watson and to a lesser degree Francis Crick, robbing her of the Nobel prize was not one. Watson and Crick received the award in 1962 but Rosalind Franklin died in 1958, and the Nobel is not awarded posthumously.
Kentsfield is a processor, not a chipset as the parent states. According to other articles, the Kentsfield processor will work with Intel's 965 and 975 chipset-based boards that are running Core2Duo's, providing a super-sweet upgrade path.
Are you an engineer or marketing guy, or aspiring to be one? No more useless extra knowledge please!
Folate and folate-related polymorphisms certainly play a role in the pathogensis and etiology of cancer through abnormal methylation and thymidine synthesis. That is not in dispute. However, your proposal that folic acid supplementation (adding more methyl groups) is somehow going to fix a problem related to having too much methylation just doesn't make any sense.
Just to balance the balance, the AMD document indicates that of 136 listed problems they plan to fix all but about a dozen.
Folate acts as a methyl group donor. The article describes a gene that is inactivated by hypermethylation (over methylation). More folate, and therefore more available methylgroups, is not likely to solve problems caused by methylation of a tumor suppressor gene.
On the other hand, folate may be beneficial in preventing tumors that would arise from double stranded breaks in DNA caused by insufficient methyl group availability interfering with the creation of thymidine or through hypomethylation of oncogenes.
Oops, typo. That should be:
Azacitidine (5-azacitidine)
Decitabine (5-aza-2'-deoxycytidine)
You will be happy to hear that Azazitidine and Decitabine are existing chemotherapeutic agents that demethylate DNA.
The transcript of the podcast makes the point that this probably did not exist prior to Windows 2000.
The Logitech DiNovo is the right size and has an excellent feel, but it's wireless. Perhaps you can hold out for the Optimus?
Not to complicate things too much, but most vitamin D supplements (particularly pills and milk) also contain substantial vitamin A supplementation. There is considerable information that retinol (a form of A) can antagonize many of the benefits of vitamin D. I would not be at all surprised to see vitamin manufacturers start to reformulate their products with more D and less A.
That's 30 deaths prevented for each one caused by sun exposure.
Just buying it might not be enough. You might want to take the next step and drink it as well.
The real debate underlying this article surrounds the appropriate "dose" of vitamin D. The current recommendations in the US (400 IU per day) are entirely based on requirements for maintaining normal bone mineral composition. This has absolutely no relation to other biological effects of vitamin D (cellular differentiation, immune cell activity).
Whereas you can get 400 IU per day by drinking vitamin D fortified milk, full-body exposure to solar ultraviolet radiation can produce as much as 10,000 - 40,000 IU of vitamin D. In the winter and at high latitudes vitamin D production from solar UV can drop to zero. Between diet, supplements, and sun exposure, the ideal combination and target dose for cancer prevention has not been established. It is almost certainly considerably above the 400 IU that you need to maintain healthy bones.
Holick's primary appointment is in Endocrinology. Gilchrist "sptripped him" of a largely symbollic secondary appointment in Dermatology. This gave Holick publicity and made her look like a git.