This guy usually isn't too far behind in creating.pkg for the stable PostgreSQL releases. I have run it on OS X for a number of years and I have been very happy with it.
Only work done by actual employees of the federal government can not be copyrighted. Work done using federal money (grants and contracts) can be copyrighted by the grantee/contractor. See the Bayh-Dole Act. Note that grant or contract solicitations that ask the grantee/contractor to give up their Bayh-Dole rights (like requiring any software written to be GPL'd) require special permission and that permission is very rarely granted.
ALL software developed using federal funds is public domain with the exception of those classified for national security reasons.
That is only true for work done by federal employees. For work done on federal contracts or grants, the grantee or the contractor can copyright the work and do with it what they want as long as the federal government gets a free license. See the Bayh-Dole Act
Thanks for the link. If this is the best response he can come up with (I'm thinking specifically of the statistical questions) I'm glad I didn't waste my money on his book.
Gould doesn't ever deny that intelligence exists or that varys from person to person, he takes a close look at the notion that most of what we call intelligence can be expressed as a single number. The first IQ test was developed by Binet to diagnose mental problems in children and he explictly rejected the idea that the test could serve as a measuring tool to determine intelligence in an entire population. Early 20th century American researchers developed some statistical tools (factor analysis) and claimed that the principle component that comes out of this is intelligence and is valid for everyone. But mathematically there isn't any reason to favor covering the most variance in the principle component as opposed to spreading it out over more components. In fact, there were big fights when some researchers did just that and claimed that intelligence was multidimensional (ie, there was verbal, spatial, etc). The mathematics can't decide it because the different views are mathematically the same. At least in Gould's telling, despite a number of attempts, there never was any biology or other independent evidence to decide it. He makes the undebatable point that just because you can calculate a principle component doesn't mean that component can be identified with something real. So to meet Gould's objections, Murray would have to either show solid evidence that there is a reason to choose single dimensional intelligence over multidimensional or show that that distinction makes no difference to the analysis. Note that Bell Curve appeared about 10 years after Mismeasure of Man, so there is certainly no reason for Murray to not know of that argument.
For a good discussion of the games Murray and others have played with the numbers see "The Mismeasure of Man" by Stephen Jay Gould. The "facts" in Murray's books are about as solid as the "facts" in polictical shout shows.
Some government agencies require that all work done with their funding pass into the public domain
Not the US federal government (including all NIH grants). The Bayh-Dole Act gives the copyright and patents rights to the grantee or contractor and retains only limited rights to the government and taxpayers. You certainly wouldn't be alone if you don't like this, but it is the law. The law was passed (in 1980) because of a perception that federally funded research was just sitting in labs and not actually being turned into anything useful to the public. It is pretty much uncontested that since its passage, much more federally funded research is being commercialized. Much of the criticism I've read (and I don't claim to know all that's out there) argues that the trend toward more use was already in place and the law might have accelerated it a bit, but didn't really affect where things were headed anyway. So to get rid of Bayh-Dole you have to go to lawmakers and say that there was a problem, Congress passed a law, things got better, but it didn't have anything to do with the law, so you should repeal it. I think it is obvious that that is going to be a tough sell.
The problems associated with universities requiring that staff turn over all right to them is a very important, but separate issue.
I don't which is more laughable; having a completely uninformed clown posting drivel or a completely uninformed moderator saying to himself
"yeah, this makes sense".
Amen! There isn't a thing in the world stopping "open source" drug development. There are even starting points. The NCI has
data for
~50,000 compounds tested in AIDS and cancer screens. Take the data and go to town. But if you are not willing to do the work and invest the time and resources, don't bitch about the people who do.
As others have pointed out, you are completely clueless. I do computer aided drug discovery and design for a living although nothing would be better for my career to have more dependence on computer modeling, nothing would be worse for the prospect of discovering useful drugs. We feel lucky to adequately model the interaction of one compound and one protein and you think it's straightforward to model the interaction of thousands of different modelcules with thousands of different proteins?
That certainly hasn't been my experience. Working in drug discovery for over 20 years (mostly for government) I would say that the majority of actual drug discovery efforts (as opposed to more basic research in biology of diseases) is done by the drug companies themselves. Also look where the market is for novel drug discovery techniques. There are many such techniques that have strong roots in acdemic labs, but the hard work of turning these interesting ideas into useful tools is funded either directly by the drug companies or indirectly because the drug companies are the only places that will pay big bucks for the finished product.
Re:Your cells are free, HeLa Cells cost $$
on
The Immortal Cell
·
· Score: 1
I'd like to know where you get your estimates. ATCC will sell you a vial of HeLa cells for $167.
The NCI will ship cells to researchers for what is estimated to be just under cost and that fee is $150. A lab usually needs only one sample and they can grow out as many cells as they want for as long as they want. I'd be surprised if there are more than a few dozen orders for HeLa cells per year and there certainly aren't thousands of orders. Add to that the fact that HeLa is just one of ~700 human tumor cell lines ATCC will ship you and I just can't see how you get to HeLa cells being a big money maker. If you know someone making millions, let alone billions on HeLa cells, I'd sure like to see some hard data.
I don't know anything about the distributed computinfg effort, but I do know about the microarray data and the NCI Lab. of Molecular Pharmacology. You can find out more about the microarray effort at the Brown-Botstein group at Stanford or at the Weinstein group at NCI. This work is part of a larger effort to characterize the cell lines in the NCI cell line screen. For more information about the cell screen, see the DTP web site. A real quick summary is that >35,000 compounds and >15,000 natural product extracts have been tested for their ability to inhbit the growth of 60 human tumor cell lines in culture. There has been a lot of work that shows the usefulness of analysing correlations in this data. The latest effort is to use the microarray data to look at the correlation between gene expression and ability of compounds to inhbit cell growth. I contrast to the above poster, I think this is going to be a very valuable tool and I could expand on that if anyone is interested. Also, all this data is available for download at the DTP site, so any of you that think they have a nifty idea on how to mine the data are quite welcome to download and have at it.
Slashdot Mirror
This guy usually isn't too far behind in creating .pkg for the stable PostgreSQL releases. I have run it on OS X for a number of years and I have been very happy with it.
Only work done by actual employees of the federal government can not be copyrighted. Work done using federal money (grants and contracts) can be copyrighted by the grantee/contractor. See the Bayh-Dole Act. Note that grant or contract solicitations that ask the grantee/contractor to give up their Bayh-Dole rights (like requiring any software written to be GPL'd) require special permission and that permission is very rarely granted.
ALL software developed using federal funds is public domain with the exception of those classified for national security reasons.
That is only true for work done by federal employees. For work done on federal contracts or grants, the grantee or the contractor can copyright the work and do with it what they want as long as the federal government gets a free license. See the Bayh-Dole Act
Thanks for the link. If this is the best response he can come up with (I'm thinking specifically of the statistical questions) I'm glad I didn't waste my money on his book.
Gould doesn't ever deny that intelligence exists or that varys from person to person, he takes a close look at the notion that most of what we call intelligence can be expressed as a single number. The first IQ test was developed by Binet to diagnose mental problems in children and he explictly rejected the idea that the test could serve as a measuring tool to determine intelligence in an entire population. Early 20th century American researchers developed some statistical tools (factor analysis) and claimed that the principle component that comes out of this is intelligence and is valid for everyone. But mathematically there isn't any reason to favor covering the most variance in the principle component as opposed to spreading it out over more components. In fact, there were big fights when some researchers did just that and claimed that intelligence was multidimensional (ie, there was verbal, spatial, etc). The mathematics can't decide it because the different views are mathematically the same. At least in Gould's telling, despite a number of attempts, there never was any biology or other independent evidence to decide it. He makes the undebatable point that just because you can calculate a principle component doesn't mean that component can be identified with something real. So to meet Gould's objections, Murray would have to either show solid evidence that there is a reason to choose single dimensional intelligence over multidimensional or show that that distinction makes no difference to the analysis. Note that Bell Curve appeared about 10 years after Mismeasure of Man, so there is certainly no reason for Murray to not know of that argument.
For a good discussion of the games Murray and others have played with the numbers see "The Mismeasure of Man" by Stephen Jay Gould. The "facts" in Murray's books are about as solid as the "facts" in polictical shout shows.
Not the US federal government (including all NIH grants). The Bayh-Dole Act gives the copyright and patents rights to the grantee or contractor and retains only limited rights to the government and taxpayers. You certainly wouldn't be alone if you don't like this, but it is the law. The law was passed (in 1980) because of a perception that federally funded research was just sitting in labs and not actually being turned into anything useful to the public. It is pretty much uncontested that since its passage, much more federally funded research is being commercialized. Much of the criticism I've read (and I don't claim to know all that's out there) argues that the trend toward more use was already in place and the law might have accelerated it a bit, but didn't really affect where things were headed anyway. So to get rid of Bayh-Dole you have to go to lawmakers and say that there was a problem, Congress passed a law, things got better, but it didn't have anything to do with the law, so you should repeal it. I think it is obvious that that is going to be a tough sell.
The problems associated with universities requiring that staff turn over all right to them is a very important, but separate issue.
I don't which is more laughable; having a completely uninformed clown posting drivel or a completely uninformed moderator saying to himself
"yeah, this makes sense".
Take a course in quantum mechanics and then you might be able to make sensible statements.
Amen! There isn't a thing in the world stopping "open source" drug development. There are even starting points. The NCI has
data for
~50,000 compounds tested in AIDS and cancer screens. Take the data and go to town. But if you are not willing to do the work and invest the time and resources, don't bitch about the people who do.
As others have pointed out, you are completely clueless. I do computer aided drug discovery and design for a living although nothing would be better for my career to have more dependence on computer modeling, nothing would be worse for the prospect of discovering useful drugs. We feel lucky to adequately model the interaction of one compound and one protein and you think it's straightforward to model the interaction of thousands of different modelcules with thousands of different proteins?
That certainly hasn't been my experience. Working in drug discovery for over 20 years (mostly for government) I would say that the majority of actual drug discovery efforts (as opposed to more basic research in biology of diseases) is done by the drug companies themselves. Also look where the market is for novel drug discovery techniques. There are many such techniques that have strong roots in acdemic labs, but the hard work of turning these interesting ideas into useful tools is funded either directly by the drug companies or indirectly because the drug companies are the only places that will pay big bucks for the finished product.
I'd like to know where you get your estimates. ATCC will sell you a vial of HeLa cells for $167. The NCI will ship cells to researchers for what is estimated to be just under cost and that fee is $150. A lab usually needs only one sample and they can grow out as many cells as they want for as long as they want. I'd be surprised if there are more than a few dozen orders for HeLa cells per year and there certainly aren't thousands of orders. Add to that the fact that HeLa is just one of ~700 human tumor cell lines ATCC will ship you and I just can't see how you get to HeLa cells being a big money maker. If you know someone making millions, let alone billions on HeLa cells, I'd sure like to see some hard data.
I don't know anything about the distributed computinfg effort, but I do know about the microarray data and the NCI Lab. of Molecular Pharmacology. You can find out more about the microarray effort at the Brown-Botstein group at Stanford or at the Weinstein group at NCI. This work is part of a larger effort to characterize the cell lines in the NCI cell line screen. For more information about the cell screen, see the DTP web site. A real quick summary is that >35,000 compounds and >15,000 natural product extracts have been tested for their ability to inhbit the growth of 60 human tumor cell lines in culture. There has been a lot of work that shows the usefulness of analysing correlations in this data. The latest effort is to use the microarray data to look at the correlation between gene expression and ability of compounds to inhbit cell growth. I contrast to the above poster, I think this is going to be a very valuable tool and I could expand on that if anyone is interested. Also, all this data is available for download at the DTP site, so any of you that think they have a nifty idea on how to mine the data are quite welcome to download and have at it.