Many readers seems worried about the ethical side of the Tonga gene project. As an example of how a company can try to meet those concerns, you may be interested in an ethics document published by a Swedish competitor to Autogen, Uman Genomics.
Personally, I am not so worried about the privacy aspect of these projects. It seems to me that both deCode and Uman Genomics work hard on solving that problem. More pressing to me is the kind of deals they are making. Is it really right for a society to favour one company before many others when the resources are actually public? This is an old problem in that countries has sold mining rights, fishing rights, and other natural resources for a long time, but we are not talking about a scarce resource here. This is information and it is not expensive or impossible to share that resource with others as long as they can pay.
And Celera has shown that it is certainly possible to have a business model an a truly open resource: The human genome. No one has exclusive rights to this data, and yet you can make a good business selling it.
Deals like this can also be made in democracies. As mentioned, Iceland's deCODE struck a deal with the Icelandic government, giving them exlusive rights (time limited, I think) to commercialize the data. The public health system in a part of Sweden has done the same with a comany called Uman Genomics. Here, the company gains access to anonymized public health records. I think there is a similar initiative in Estonia.
Technically, I think the samples belongs the one collecting them. What these companies buy is the exlusive cooperation with public healt authorities. Presumably, Jonah donates blood to a company which also collects genealogy info, it would all be in order. However, without clinical records, this information will no be so valuable.
The deal seems good to me. A crucial fact is that Autogen is setting up a facility in Tonga, instead of shipping all the samples to Australia. Lars
__
Re:A useful feature to add to such a device...
on
New All-In-One Nokia
·
· Score: 2
There are companies developing GPS-less solutions for positioning too. The idea is that GSM cells are fairly small, especially in cities, so by determining what cell (and possibly what cells has recently been used) a user is in you can also fairly well pin-point where that user is.
The applications mentioned for this kind of stuff is emergency call support, taxi finding out where a caller really is, or keeping track of employees in the field.
There are interesting privacy/security issues here though!
I would like to recommend the book "Robo Sapiens: Evolution of a new species" (released by MIT press this fall) for all robotics geeks out there. It is a very interesting book with lots of great photography surveing the state of the art in a pop science way. The Honda P-series is only one of many projects with its own section in that book. They cover everything from basement hobbyists to multi millon dollar corporation project.
This is not an accurate picture of broadband roll-out in Sweden. After a long debate about government funding of high-speed networks, a decision was taken last spring (summer?), but I don't think they have actually started working. Meanwhile, corporations are speeding like mad to connect people in (at least) densly populated areas. In my neighbourhood we'll get broadband access in the next few weeks, and it will cost me $50 initially (which also buys me a network card) + $20 per month. The neighbourhood will own the LAN hardware, and we are locked in for two or three years to the same ISP. I don't see how they can spew ads on us though, except by distributing special browsers.
What the government should do now is to make sure rural areas are connected, but I am not sure they have understood that. Also, many small rural towns are pointing out that they are already connected, so the money could be spent on better things.
Instead of/.-ing the NBC wannabes you should aim your browser at the Nobel e-museum (organized by the Nobel foundation is seems) where the announcements are made. They are hopefully better prepared!
Lars
__
Re:Nobel Prize Research Refuted?
on
Nobel Prizes
·
· Score: 2
Although I have not followed this closely, I recall seeing titles of articles in Nature and/or Science announcing support for Prusiner's idea. Of course, he might still be wrong, but right now there seems to be more support for him than criticism...
True, there's no Linux or BSD version of Eudora available, but might the algorithms involved one day make it into other mail programs?
Are you implying that I would need stuff like that? Take that back you miserable bastard! I have been on the Net for 13 years and know bloody well how to communicate with others. I don't need a stupid "AI" program to tell ME how to write damn it! You are of course an exception, huh? Thinking "Yeah, let me give those slashdotters a hint --- there is way to much flaming there" right? Well let me tell you something: you suck! Just because there are stypid minors out there that loves to spew there guts over things at/. to make themselves look cool to their friends does not make all of us spammers. SO THERE!
And Hemos, you better find an app for reading/. submissions and writing article blurbs, because what you do today stinks and I am sure a program could do a better job!
Common practice in DNA fingerprinting today looks at more than one little piece of DNA (there are some caveats to this depending on what kind of DNA material you have available). In Europe, several regions with small repeated fragments are studied. The length of repeat regions typically vary a lot. The actual regions they use are quite studied, and combining the length pattern from several regions, you essentially get a DNA fingerprint.
I have been told that the situation is a little different in the US, where DNA fingerprinting was adopted in the crime labs much earlier. A set of sites to cut the DNA has been chosen and the lengths of the fragments produced are used. This method has been critisized for not being understood well enough. This may have changed since I heard about it though. Anyone with details?
The exclusion capacity of DNA fingerprinting is something like 99.999%. That is, the probability of two people sharing the same pattern is 1 in a 100000. This means that if you can reduce the number of people that could have been involved in the crime significantly, then you can quite confidently convict someone. However, the strength of DNA fingerprinting is of course exclude people from an investigation.
But i don't think that the fruit fly is going to be the Rosetta Stone anyway, because from what i've heard, there are no wild type fruit flies left anymore. They've been used so much in genetics research that you couldnt find a natural fruit fly if you looked!
Of course there are natural fruit flies out there. They may not be used for research (I have no idea), but a genetically new fruit fly is interesting to the researchers anyway.
For example, this weeks issue of Nature has an article about constructing a model of Parkinsons disease in the fruitfly. This is probably a great tool to test drug candidates on. Fruitflies are somewhat cheaper and more convenient to experiment on than human beings.;-)
The comparisons you make are quite unfair. First of all, the Drosophila project was done in cooperation with the US funded Berkeley Drosophila Genome Project. That alone should keep you from barking at the government funded projects.
Second, the fact that HGP has been going on for 10 years (how long has Celera been going) means nothing when the sequencing capacity seems to double every year. This means that you in one year can recover what has accumulated over several years of work!
Third, HGP is making their data public within 24 hours. You think Celera doesn't make use of that data?
I am also uncertain about your oversampling claims. If Celera is content with 4x and do not use public sequences (which they can't if they are supposed to be ahead), then they will have serious problem of actually connecting the pieces. Granted, you can still go gene-mining and make important discoveries. Also, I think (due to the competition) that the HGP has settled for 5x oversampling to get a rough draft available later this year. Whether there is one or more individuals sampled doesn't really matter. You are fighting statistics which says that you need 10x if you are going to have any hope of connecting the pieces. Notice that they used 14x for Drosophila. Actually, this came up in an earlier/. discussion where it was claimed that the HGP uses a single individual as well.
For a more informative source, check todays issue of Science, which includes a special about the feat. If you register, you get to read abstracts of the articles. This alone is quite informative.
I actually made this post with my tongue firmly in my cheek, but I guess I forgot a smiley:-)
However, I think you are wrong. The FDL could very well be posted under itself due to the modification clause, point A:
A. Use in the Title Page (and on the covers, if any) a title distinct from that of the Document, and from those of previous versions (which should, if there were any, be listed in the History section of the Document). You may use the same title as a previous version if the original publisher of that version gives permission.
So either the derived licence can not be called the FDL, or you have received permission from the original author, in which case the content cannot have changed distinctively.
There was a special issue on digital watermarking in Communications of the ACM two years ago: The citation is CACM vol 41, no 7, July 1998.
I never read the articles on the subject very carefully, but I don't think they presented more than ideas there (CACM is rarely technical nowadays), but maybe it is possible to get some pointers to modern research. The articles discuss many types of watermarking, including what you probably want: cryptograhpically secure which is robust against operations such as compression, filtering, and reduction.
The field of digital watermarking seems quite undeveloped, so I wouldn't hold my breath for free software for it.
"Streambox has been stealing our streams for quite some time," says Opher Mizrahi, CEO of MovieFlix.com. "It's costing us bandwidth and we don't get the benefit of the registration. They can link to our movies illegally, and there's no reason to come to the MovieFlix Web site," he adds.
Streambox denies the theft of course, but what is the guy talking about? Deep linking? Or does the RA player provide them with some kind of statistics?
I have not used these players, but I can't see why an RA user would not be able to do the same thing as with a Streambox player.
The AC said: The optimal performance increase with 2 CPU SMP is about 45%. Linux is below 20%, Solaris is close to 40%.
Huh? Why 45%? Is this Quake performance or some kind of general claim? It sounds awfully low.
I have made many CPU intensive tasks on a 4 CPU Solaris machine that I have shared with others (also doing intensive computations) and have not noticed anything like this. I have not made any timings, but each CPU is faster than what I have on my desk and it certainly looks like a linear speedup until the machine runs out of processes.
I don't think the idea of a microwave oven cooperating with the frozen dinner will happen. Sure, the technology works (and I actually can't see why a university has to be involved to make it happen), but it needs cooperation from too many entities: more than one.
My parents still have a VCR with a bad code reader. The Swedish version of TV guide carried full page ads with bar codes for a couple of years, making it possible to drag the bar code pen across a few lines and then beam the data over to the VCR using infra-red light. It was cool, but it required someone to publish the bar codes for the most popular TV programming every week and this is expensive, so naturally it died a premature death.
Similarly with a cheaper coding scheme. Some papers used to have ~10 digit code by the program title which encoded the essential data for the VCR. But who wanted to make sure that those codes were right? Seems that idea died too.
I believe that successful technology development evolves from existing technologies. For instance, I discovered this fall that my new VCR could set the time all by itself. It simply use the time info sent out in the text-TV data. That is an excellent use of an existing technology!
Actually, there are those who wants to see this happen soon (except for the brain-fry). They want a web-appliance on your fridge. I still have a hard time seeing why I would need this, but hey, who knows what people wants?
I note with interest that the Slazenger Institute at Cambridge UK has published the first batch of sequenced DNA (Nr. 23) on the Net so everyone can have it. Are US researchers doing the same?
I presume you mean the Sanger Institute?
It was researchers from several institutes/universties, including US researchers involved in that feat. The foremost private initiative, Celera Genomics, have promised to donate humage genome data early this year. Many academics worry that they will not stand to their promise though. However, that is independent of the issue of patents.
I think a better analogy is with prospecting. A company can buy the rights to prospect an area and get first dibs on a gold ore within that area.
Some gene patenting seems reasonable to me, because they require a large investment in time, money and development in general, but there are certainly many cases that are downright trivial. The concern among many academic researchers is that the patent offices are giving away gold that rightfully belongs to humanity.
New species have, sort of, been in patent applications. I have heard of a case where a company tried to patent a compound found in a plant which would make a good ingredient for toothpaste. Indigenous people have chewed on these plants for generations... The patent was supposedly turned down. Anyone heard more of this story?
There is also a village in Italy where the locals have barely no heart problems. It turns out that they carry a mutated form of a protein that is very good at breaking down colesterol (sp?). This protein has been patented and is now developed to become a drug. It remains to be seen if the company is successful.
No, this is most likely not about "synthesized genes" (is this really a used concept? Synthesized proteins makes sense to me, but synthesizing a gene seems like a strange concept.) but so called Expressed Sequence Tags (EST). These are short DNA sequences (in the range of 100-200 bases I believe) that are used to map the genome. They bind to positions on the genome that code for proteins (they are "expressed") and are simply a kind of marker. Notice also that an EST is a part of a gene, not a full gene.
EST are useful for detecting whether a gene is present and can therefore be used to piece together the big puzzle of DNA strings that the genome projects produce, but can also be used to detect a disease I presume (I don't fully understand why companies try to patent EST:s).
Yep. IIRC Reading Football Club (...). Still, last I checked, Reading FC weren't in any danger of threatening the Premier division... Or even anywhere near even being in that division, either...
Didn't the English national team also use Geller's "talents" during the last world cup? I remember reading that the former English manager (don't remember his name now, I am not big on football) had a thing for the paranormal...
They are doing the MDMI (multiple data many individuals) approach. What they get is basically an average genome. And certainly a good map of the genome.
The next step is to hunt for SNP:s, which stands for Single Nuceleotide Polymorphisms. And the race has already started. Both companies and universities are hunting for them. They are expected to be useful for things like identifying inheritedable deseases.
Slashdot Mirror
Personally, I am not so worried about the privacy aspect of these projects. It seems to me that both deCode and Uman Genomics work hard on solving that problem. More pressing to me is the kind of deals they are making. Is it really right for a society to favour one company before many others when the resources are actually public? This is an old problem in that countries has sold mining rights, fishing rights, and other natural resources for a long time, but we are not talking about a scarce resource here. This is information and it is not expensive or impossible to share that resource with others as long as they can pay.
And Celera has shown that it is certainly possible to have a business model an a truly open resource: The human genome. No one has exclusive rights to this data, and yet you can make a good business selling it.
Lars
__
Technically, I think the samples belongs the one collecting them. What these companies buy is the exlusive cooperation with public healt authorities. Presumably, Jonah donates blood to a company which also collects genealogy info, it would all be in order. However, without clinical records, this information will no be so valuable.
The deal seems good to me. A crucial fact is that Autogen is setting up a facility in Tonga, instead of shipping all the samples to Australia.
Lars
__
There are companies developing GPS-less solutions for positioning too. The idea is that GSM cells are fairly small, especially in cities, so by determining what cell (and possibly what cells has recently been used) a user is in you can also fairly well pin-point where that user is.
The applications mentioned for this kind of stuff is emergency call support, taxi finding out where a caller really is, or keeping track of employees in the field.
There are interesting privacy/security issues here though!
Lars
__
Salon.com published a review recently.
Lars
__
This is not an accurate picture of broadband roll-out in Sweden. After a long debate about government funding of high-speed networks, a decision was taken last spring (summer?), but I don't think they have actually started working. Meanwhile, corporations are speeding like mad to connect people in (at least) densly populated areas. In my neighbourhood we'll get broadband access in the next few weeks, and it will cost me $50 initially (which also buys me a network card) + $20 per month. The neighbourhood will own the LAN hardware, and we are locked in for two or three years to the same ISP. I don't see how they can spew ads on us though, except by distributing special browsers.
What the government should do now is to make sure rural areas are connected, but I am not sure they have understood that. Also, many small rural towns are pointing out that they are already connected, so the money could be spent on better things.
Lars
__
Lars
__
Lars
__
Are you implying that I would need stuff like that? Take that back you miserable bastard! I have been on the Net for 13 years and know bloody well how to communicate with others. I don't need a stupid "AI" program to tell ME how to write damn it! You are of course an exception, huh? Thinking "Yeah, let me give those slashdotters a hint --- there is way to much flaming there" right? Well let me tell you something: you suck! Just because there are stypid minors out there that loves to spew there guts over things at /. to make themselves look cool to their friends does not make all of us spammers. SO THERE!
And Hemos, you better find an app for reading /. submissions and writing article blurbs, because what you do today stinks and I am sure a program could do a better job!
Cordially,
Lars
__
I have been told that the situation is a little different in the US, where DNA fingerprinting was adopted in the crime labs much earlier. A set of sites to cut the DNA has been chosen and the lengths of the fragments produced are used. This method has been critisized for not being understood well enough. This may have changed since I heard about it though. Anyone with details?
The exclusion capacity of DNA fingerprinting is something like 99.999%. That is, the probability of two people sharing the same pattern is 1 in a 100000. This means that if you can reduce the number of people that could have been involved in the crime significantly, then you can quite confidently convict someone. However, the strength of DNA fingerprinting is of course exclude people from an investigation.
Lars
__
For example, this weeks issue of Nature has an article about constructing a model of Parkinsons disease in the fruitfly. This is probably a great tool to test drug candidates on. Fruitflies are somewhat cheaper and more convenient to experiment on than human beings. ;-)
Lars
__
Second, the fact that HGP has been going on for 10 years (how long has Celera been going) means nothing when the sequencing capacity seems to double every year. This means that you in one year can recover what has accumulated over several years of work!
Third, HGP is making their data public within 24 hours. You think Celera doesn't make use of that data?
I am also uncertain about your oversampling claims. If Celera is content with 4x and do not use public sequences (which they can't if they are supposed to be ahead), then they will have serious problem of actually connecting the pieces. Granted, you can still go gene-mining and make important discoveries. Also, I think (due to the competition) that the HGP has settled for 5x oversampling to get a rough draft available later this year. Whether there is one or more individuals sampled doesn't really matter. You are fighting statistics which says that you need 10x if you are going to have any hope of connecting the pieces. Notice that they used 14x for Drosophila. Actually, this came up in an earlier /. discussion where it was claimed that the HGP uses a single individual as well.
Lars
__
Lars
__
However, I think you are wrong. The FDL could very well be posted under itself due to the modification clause, point A:
So either the derived licence can not be called the FDL, or you have received permission from the original author, in which case the content cannot have changed distinctively.
Cheers,
Lars
__
From the document:
Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed.
This is certainly not allowed by the FDL...
Lars
__
There was a special issue on digital watermarking in Communications of the ACM two years ago: The citation is CACM vol 41, no 7, July 1998.
I never read the articles on the subject very carefully, but I don't think they presented more than ideas there (CACM is rarely technical nowadays), but maybe it is possible to get some pointers to modern research. The articles discuss many types of watermarking, including what you probably want: cryptograhpically secure which is robust against operations such as compression, filtering, and reduction.
The field of digital watermarking seems quite undeveloped, so I wouldn't hold my breath for free software for it.
Lars
__
In a Standard article (found and cited by Yardley in post #58) there is a quote from a content provider:
"Streambox has been stealing our streams for quite some time," says Opher Mizrahi, CEO of MovieFlix.com. "It's costing us bandwidth and we don't get the benefit of the registration. They can link to our movies illegally, and there's no reason to come to the MovieFlix Web site," he adds.
Streambox denies the theft of course, but what is the guy talking about? Deep linking? Or does the RA player provide them with some kind of statistics?
I have not used these players, but I can't see why an RA user would not be able to do the same thing as with a Streambox player.
Lars
__
The AC said:
The optimal performance increase with 2 CPU SMP is about 45%.
Linux is below 20%, Solaris is close to 40%.
Huh? Why 45%? Is this Quake performance or some kind of general claim? It sounds awfully low.
I have made many CPU intensive tasks on a 4 CPU Solaris machine that I have shared with others (also doing intensive computations) and have not noticed anything like this. I have not made any timings, but each CPU is faster than what I have on my desk and it certainly looks like a linear speedup until the machine runs out of processes.
Lars
__
I don't think the idea of a microwave oven cooperating with the frozen dinner will happen. Sure, the technology works (and I actually can't see why a university has to be involved to make it happen), but it needs cooperation from too many entities: more than one.
My parents still have a VCR with a bad code reader. The Swedish version of TV guide carried full page ads with bar codes for a couple of years, making it possible to drag the bar code pen across a few lines and then beam the data over to the VCR using infra-red light. It was cool, but it required someone to publish the bar codes for the most popular TV programming every week and this is expensive, so naturally it died a premature death.
Similarly with a cheaper coding scheme. Some papers used to have ~10 digit code by the program title which encoded the essential data for the VCR. But who wanted to make sure that those codes were right? Seems that idea died too.
I believe that successful technology development evolves from existing technologies. For instance, I discovered this fall that my new VCR could set the time all by itself. It simply use the time info sent out in the text-TV data. That is an excellent use of an existing technology!
Lars
__
Actually, there are those who wants to see this happen soon (except for the brain-fry). They want a web-appliance on your fridge. I still have a hard time seeing why I would need this, but hey, who knows what people wants?
Lars
__
I note with interest that the Slazenger Institute at Cambridge UK has published the first batch of sequenced DNA (Nr. 23) on the Net so everyone can have it. Are US researchers doing the same?
I presume you mean the Sanger Institute?
It was researchers from several institutes/universties, including US researchers involved in that feat. The foremost private initiative, Celera Genomics, have promised to donate humage genome data early this year. Many academics worry that they will not stand to their promise though. However, that is independent of the issue of patents.
Lars
__
I think a better analogy is with prospecting. A company can buy the rights to prospect an area and get first dibs on a gold ore within that area.
Some gene patenting seems reasonable to me, because they require a large investment in time, money and development in general, but there are certainly many cases that are downright trivial. The concern among many academic researchers is that the patent offices are giving away gold that rightfully belongs to humanity.
New species have, sort of, been in patent applications. I have heard of a case where a company tried to patent a compound found in a plant which would make a good ingredient for toothpaste. Indigenous people have chewed on these plants for generations... The patent was supposedly turned down. Anyone heard more of this story?
There is also a village in Italy where the locals have barely no heart problems. It turns out that they carry a mutated form of a protein that is very good at breaking down colesterol (sp?). This protein has been patented and is now developed to become a drug. It remains to be seen if the company is successful.
Lars
__
No, this is most likely not about "synthesized genes" (is this really a used concept? Synthesized proteins makes sense to me, but synthesizing a gene seems like a strange concept.) but so called Expressed Sequence Tags (EST). These are short DNA sequences (in the range of 100-200 bases I believe) that are used to map the genome. They bind to positions on the genome that code for proteins (they are "expressed") and are simply a kind of marker. Notice also that an EST is a part of a gene, not a full gene.
EST are useful for detecting whether a gene is present and can therefore be used to piece together the big puzzle of DNA strings that the genome projects produce, but can also be used to detect a disease I presume (I don't fully understand why companies try to patent EST:s).
See this policy paper for an interesting view on the patent issue.
Lars
__
Yep. IIRC Reading Football Club (...). Still, last I checked, Reading FC weren't in any danger of threatening the Premier division... Or even anywhere near even being in that division, either...
Didn't the English national team also use Geller's "talents" during the last world cup? I remember reading that the former English manager (don't remember his name now, I am not big on football) had a thing for the paranormal...
And of course, they did not do too well...
Lars
__
"..., and sequencing the insert"
And how do you do that without PCR?
Lars
--
They are doing the MDMI (multiple data many individuals) approach. What they get is basically an average genome. And certainly a good map of the genome.
The next step is to hunt for SNP:s, which stands for Single Nuceleotide Polymorphisms. And the race has already started. Both companies and universities are hunting for them. They are expected to be useful for things like identifying inheritedable deseases.
Lars
--