Why would you want to connect your brain to the Internet? Far too great a risk of an NSA virus.
What you want to do is place your brain in a networked, Earthquake-proof, fire-proof enclosure, with an Infiniband connection to a Linux server. This would then be linked, via an OpenBSD firewall, to the Internet and also to some sort of ROV that can act as a relay between brain and body.
Meanwhile, your skull would contain an embedded computer, a massive multiplexer/demultiplexer to link up the nervous system and a very high bandwidth microwave link to the ROV.
It would reduce bandwidth requirements and latency if the motor neurons remained in the skull, with the rest of the brain transmitting only executive instructions and not specific nerve impulses. Those could be generated more locally. Split brains are found in squid, so we know this kind of isolation is possible. You'd also need a smaller computer, as it would largely be relegated to providing network security, error correction and data compression/decompression. This means you could increase the number of motor neurons, increasing the sophistication of muscle response.
From the brain-at-home POV, removing motor neurons means reducing distances between the other parts of the brain, reducing response time. (Since microwaves are a fair bit faster than electrochemical chains for transmitting data, latency due to distance is insignificant. You would be limited by signal strength and error correction codes, but the radius would be far beyond typical distances travelled by westerners anyway.)
Dissecting the brain further, with no executive functions in the skull to contend with, eyes could be larger, giving you higher resolution at the same number of distinguishable colours, or more colours at the same resolution. This would require the optic centres to be cut out of the brain and enlarged accordingly. Unlike those parts dealing with memory, the structure of the visual cortex probably won't vary much. By replacing the synapses with optic fibre, you can reduce latency, reduce errors, increase resilience to aging, reduce space requirements and eliminate tau protein knots. The reduced space means more visual cortex for the same response time, letting you process every scrap of available data rather than wasting it as the brain currently does.
Much the same applies to hearing and sense of balance. You could probably double your frequency range and your ability to distinguish tilt.
You may be more restricted in movement, but you would exchange it for superhuman senses and superhuman reflexes.
Well, on the basis that malloc() and free() are opposites, and writing to free()d memory is delusional, delusional == free.
On the other hand, look at what America has been fine with imposing on other nations, then what it has been fine living with within itself. America has never been about upholding the standards it claims.
As for freedom, from pretty much the very beginning, Constitutional "rights" have been revealed as merely permissions granted to whoever those in power happened to like. (Hence the amazing numbers of prisoners promising to vote in favour of those in power at the time magically getting dispensation, and the amazing problems minorities with a habit of voting the wrong way have.) There are no "rights" in America, never have been. You can't remove, grant or even waive rights - they exist or they don't and that's the end of it. (On matters of rights, I can be complete Kant.) Permissions are the only things you can ever add or subtract, and since everything in the Constitution has been put on hold or removed at some point, everything in the Constitution is a permission.
But that's fine with Americans, provided it's some other person getting these pseudo-rights pulled, ideally if they're from the wrong political party or the wrong ethnicity, or (in this case) testifying for the wrong side. People want THEIR freedoms and the freedoms of "their side" defended, "the other side" gets whatever crumbs are thrown their way. And that is why Americans are generally elitist snobs of an order pre-WW1 Europe could only ever have dreamed of aspiring to. Even the worst of the empires treated "non-conformists" better than that. Admittedly, in many cases, by shooting them, but that's still treating them as equals of dogs. The US treats undesirables (ie: those who don't think the same) as closer to pond scum.
Judges have never been taken seriously, when decisions run against the government. There have been Supreme Court rulings awarding the Black Hills to the Sioux. The government, for 50? 60? years has managed to ignore the ruling. That covers a lot of Presidents in both parties. But it includes an awful lot more voters, who (by re-electing anyone at all) are active supporters of the Government's refusal to follow SCOTUS orders. You are putting people in power who you know, full well, will not obey judge rulings they do not like. The upshot of that is that I have bugger-all pity when those same politicians disregard (or creatively interpret) rulings that happen to affect you or your "side". You're discovering the price of giving them that power.
The only downside is that everyone in the world, free or otherwise, must pay the price for these games. You have robbed us all of what liberty we had, just so that the right people were oppressed.
If notices are going to be sent and processed (then accepted/rejected automatically), followed by verbal fighting, why not simplify the whole exercise? Have a two-way corewars battle between the service and the complainant, winner decides if the link stays. Since the complainant is rarely the copyright holder and the service rarely provides the material, you can extend this to a four-way battle using crobots.
This is just as logical as the current system and has a higher probability of protecting original content from abusive/malign notices.
...has never changed. People should never be in a position where they can be pressured, cajoled or manipulated into suicide; doctors are improving but are still not very competent at figuring out the long-term prognosis for patients, the error rates are way too high. This should ALWAYS be balanced against indisputably terminal conditions, conditions where the person ceases to be meaningfully human, conditions of extreme suffering where the patient makes unconditionally clear (not just once but over time) that pain has to cease, one way or another, and any similar situation.
In other words, I don't trust doctors, patients or families to make perfect decisions. There has to be hard evidence that this case is typical, not atypical, hard evidence (within reason - coma patients don't talk much) that the patient has firmly decided on that path and it isn't the moment-to-moment feeling everyone gets on a regular basis, and hard evidence that conceding to fate is the last realistic untried option.
But if that evidence is beyond dispute, doctors are oath-bound to do least harm and painless death is less harm than agonizing death.
(I would never have made a good doctor. The older and moodier I get, the more I feel that someone invent hell, just so that idiots have somewhere to go.)
June 4th would be symbolically better, and it would piss off those responsible for a date being needed. Which just makes that day even better. Or we could combine it with veteran's day, as I imagine the veterans are a little unhappy at winning the battles but losing the war. Hey, if you fight for freedom and then have none, you lost.
I do have a plan, a plan built on solid foundations (although I must admit some of the support beams need work). A plan I have posted a few times, so am unsure why I should bother now. It is also a plan that would require a lot of people joining in, which is never going to happen these days - experimental governance was all the rage in the 1600s and 1800s, but people are too comfortable now. Better the devil you know, after all.
Nonetheless, if it could be started, I can absolutely guarantee it would give superior results to your approach.
The US has been expelled from UN bodies in the past, due to violations of international law. Although 95% symbolic, it hurts them in things like trade negotiations and bilateral agreements. In turn, that makes equally abusive but unsanctioned nations more attractive to business. That, in turn, hurts donations to politicians and tax revenues. Not necessarily by a lot, but name a politician who wants to spend less.
(Note: Tea Partiers and Libertarians want other people to spend less. They, themselves, are by far the worst of the pork barrel spenders.)
For other things to be thankful for, The Guardian has an excellent review of a documentary + book on the death squads the US military and intelligence run. For example, I'm thankful there are attempts to colonize Mars. I have better odds of a life away from such idiots, if I'm on a civilized world and they're stuck on Earth.
Direct proof? Yes. I have had sections of my DNA sequenced by 3 different groups, using 3 different sequencers, including the whole of my mitochondrial DNA and various chromosomes from my nucleic DNA (Y is the easiest one to get sequenced multiple ways).
So far, differences detected: Zero.
So they are giving me valid SNPs everywhere I can double check.
Can I confirm SNPs I haven't had tested elsewhere? No, but sequencers only cost around $15 million, so if you could loan me the money, I can easily verify the rest. I would even be happy to verify results for anyone else who asked, at cost in electricity to run a batch divided by the number in that batch. I'll deal with the other costs.
Now, how fair is that? Total verification of the raw data results from any genome research group.
In fact, I don't care who supplies the cash. The FDA can, for all I care. I will then give you verification of any of these start-ups for slightly under the cost of the run itself, complete with statistical confidence limits because no test is 100% perfect.
I'm willing to sacrifice my time and effort on building an ultraclean room and buying the rest of the hardware and enzymes needed to take you from one anecdotal account to a full, absolutely independent (including of any existing lab) facility that can provide an accurate scientific assessment in a manner that is compliant with my ethos of science verified by scientists and observed by the public, nothing hidden, nothing obscured, no trade secrets, no concealing statistics by hiding behind certainties.
The Gates Foundation could provide the initial funding, for all I care. And as a long time Slashdotter, I don't say that lightly.
So who's game? Reply with who you sent the grant/funding request to and how I can check the progress. Anyone?
Like I said, those people should travel to the top of the nearest tall building and jump. Preferably yesterday. I have no pity for such people, no compassion, no time. The world would be far better off if such people offed themselves. The world needs thinkers, it does not need reactors. Well, other than fusion reactors. We could do with a lot of those.
Anyone who thinks a 0.1% increased risk in a single SNP when dozens are involved, all of which have to show compatible mutations, means they will surely die of, well, anything is too stupid to live. Frankly, if that's your best objection, I think the test should be provided free across America and Britain. Get rid of at least some of the fools, rather than have them be a burden on society by running banks or whatever.
Good riddance to them. Hope the jumped-up twits die slowly.
(23&Me cannot tell you if you have a disease, or if you will ever get it, it can only give you very rudimentary statistical information on correlations. Anyone confused about correlations and causation should be provided with a HOWTO on suicide and appropriate implements.)
As for the studies, published peer-reviewed studies on genetics do vary in quality. That is why nobody considers them. Only metastudies matter. I would argue a metastudy with a sample size of 2 is low, but even getting a repeat study published is a major feat and genetics hasn't been cheap enough for long enough for the numbers of studies to grow. I like that 23&Me makes an effort to run their own study, giving them a metastudy size of 3, still low but getting into respectable territory.
But, again, what does it matter? Let us say that there are 10 SNPs required, and one SNP has a 0.1% increased risk but no others do, should you worry? One in a thousand chance above normal for anything is so close to normal that you would have to be one of the deserving suicides to think it worth concerning yourself with.
What does it diagnose? It doesn't tell you that you have condition X, it doesn't tell you that you will ever have condition X, and since correlation is NOT causation, it doesn't tell you that condition X is even possible.
I see no diagnosis, I see raw data and I see correlations (most of which are borderline noise level) but I see no statement to the effect that anything specific has happened, is happening or will ever happen. Where is the diagnosis?
That should be a simple one to answer, even for you.
I would agree with that. I would even agree to 23&Me putting more emphasis on the fact that it measures probabilities, not certainties. I certainly think they should quantify their confidence in a marker better than they do.
I also absolutely agree with you that information on life-threatening reactions to medications should be unhindered. If it is wrong, then the doctors either used an equally effective alternative (so, no loss) or gave you unnecessarily extensive monitoring (no harm done and everyone likes being fussed over occasionally). If it is right, it saved your life. There is no obvious downside.
Besides, doctors aren't going to go by your printouts for personalized medicine. They will use your raw data files, but any hospital with that level of expertise (and there won't be many) will have their own databases built from the existing literature and the hospital's experience. They will have actual geneticists on-hand or readily available via strong links with the local university. That is how these things work - and, yes, I am very aware of how regular (non-teaching) hospitals work with universities for mutual benefit. I've seen it, even been a part of the process.
So, who is 23&Me potentially misinforming? If not the patients or the practitioners, who is left?
A high level of confidence of a correlation between an SNP and a condition means that two independent genetic studies have produced the same results AND members of the site who have agreed to allow their data to be used for research purposes (not all do) show the same correlation.
Ok, I wouldn't do it that way, I would indicate reliability as a percentage asymptotic to 100% (you can never be certain) where their high confidence would be somewhere around 50%. But it's a good indication of reliability. Most associations are not marked as high confidence. They have either not been in a repeat study or there is no basis in the "natural" sample set from users. Such results should be ignored. They are given because there is a vague possibility and therefore interesting to those looking at the research process and trends. Users know this.
There is nothing to evaluate, except perhaps the lab conditions. The SNPs do not diagnose a condition, 23&Me do not offer any diagnosis, all that is given is a set of percentages relating to the correlation of an SNP with an effect. Correlation is NOT causation.
The FDA actively work with drug companies known to be carrying out unauthorized tests on unwilling human subjects in Africa. They have accepted results from US prisons, where again no consent was ever given. They encourage the suppression of negative results in journals. Corrupt? They are better than certain historical medical authorities, but not significantly.
(No, I don't give a damn that 1960s and 1970s America had no restrictions on non-consensual biological and chemical experiments for medical purposes. They'd been declared a crime against humanity two decades. And outsourcing to Africa is no excuse. Sorry, they are criminals. Whatever right they do should be seen in isolation from their wrongs. They are guilty, the heads should stand trial along with the heads of pharmaceutical companies indulging in data suppression or illegal human experimentation, and a medically sound alternative should be created.)
Then you're an idiot, because that is the only way to make any claim regarding effectiveness (positive or negative) at the individual gene level when we don't know all the genes, all the interactions of genes or the impact of "Junk DNA" (which we now know isn't junk but metadata).
23&Me do not diagnose. They identify SNPs that have a given (often very low) correlation with a condition. They state the current best guess at that correlation. That is it. There are no "tests" except for the SNPs themselves. And those will be accurate.
They ARE NOT telling you you have X, Y or Z. They are telling you that gene G is ASSOCIATED WITH an N% increased risk.
For chrissakes (or, as my kindle spellchecker insists, cheesecakes), at least bother to do basic research. Oh, that's right, the serious COMPETENT Slashdotters abandoned the site because of the overwhelming sheer stupidity out here.
I don't know why I bother. Anyone with an ounce of brain capacity can figure out that raised odds of 0.23% do not equal an absolute certainty of instant death. But, clearly, you and the FDA lack even that.
They're limiting it to images on the Wayback Machine where the levels of pink or black do not indicate things that might cause it to suddenly decide the human race needs obliterating.
Deadly mutagenesis! Happens naturally all the time. See any number of stories of fungi becoming deadly for no apparent reason.
We use it ourselves, as others have noted.
Actually, "deadly" is just code for a failure of your microbiome to process the organics. If you have crappy microflora, don't blame the UV. Blame your diet. If you have flexible, powerful microbes, you're going to be fine.
Translated: Eat well or die, Earthling scum! (Sorry, channelling my inner Sontaran.)
Router poisoning is an old attack, used many times. Sometimes deliberately, sometimes when a router goes bad.
BGP supports encryption, via certificates or shared secrets. People were supposed to have made the switch years ago, but I have posted time and again that this hasn't happened in practice.
I'd fix the problems myself, but that would mean risking being sued to oblivion by patent trolls. Most of what I outlined could be prototyped and demo'd in very short order, but if the return for my investment is financial death, what is the point?
I'd be happy to give action, not words, in a rational, fair environment. We don't have one right now, and short of setting up an office in Antarctica (which has not signed any patent treaties), I don't see how I can make it viable to operate in the system we do have.
Slashdot Mirror
Why would you want to connect your brain to the Internet? Far too great a risk of an NSA virus.
What you want to do is place your brain in a networked, Earthquake-proof, fire-proof enclosure, with an Infiniband connection to a Linux server. This would then be linked, via an OpenBSD firewall, to the Internet and also to some sort of ROV that can act as a relay between brain and body.
Meanwhile, your skull would contain an embedded computer, a massive multiplexer/demultiplexer to link up the nervous system and a very high bandwidth microwave link to the ROV.
It would reduce bandwidth requirements and latency if the motor neurons remained in the skull, with the rest of the brain transmitting only executive instructions and not specific nerve impulses. Those could be generated more locally. Split brains are found in squid, so we know this kind of isolation is possible. You'd also need a smaller computer, as it would largely be relegated to providing network security, error correction and data compression/decompression. This means you could increase the number of motor neurons, increasing the sophistication of muscle response.
From the brain-at-home POV, removing motor neurons means reducing distances between the other parts of the brain, reducing response time. (Since microwaves are a fair bit faster than electrochemical chains for transmitting data, latency due to distance is insignificant. You would be limited by signal strength and error correction codes, but the radius would be far beyond typical distances travelled by westerners anyway.)
Dissecting the brain further, with no executive functions in the skull to contend with, eyes could be larger, giving you higher resolution at the same number of distinguishable colours, or more colours at the same resolution. This would require the optic centres to be cut out of the brain and enlarged accordingly. Unlike those parts dealing with memory, the structure of the visual cortex probably won't vary much. By replacing the synapses with optic fibre, you can reduce latency, reduce errors, increase resilience to aging, reduce space requirements and eliminate tau protein knots. The reduced space means more visual cortex for the same response time, letting you process every scrap of available data rather than wasting it as the brain currently does.
Much the same applies to hearing and sense of balance. You could probably double your frequency range and your ability to distinguish tilt.
You may be more restricted in movement, but you would exchange it for superhuman senses and superhuman reflexes.
Well, on the basis that malloc() and free() are opposites, and writing to free()d memory is delusional, delusional == free.
On the other hand, look at what America has been fine with imposing on other nations, then what it has been fine living with within itself. America has never been about upholding the standards it claims.
As for freedom, from pretty much the very beginning, Constitutional "rights" have been revealed as merely permissions granted to whoever those in power happened to like. (Hence the amazing numbers of prisoners promising to vote in favour of those in power at the time magically getting dispensation, and the amazing problems minorities with a habit of voting the wrong way have.) There are no "rights" in America, never have been. You can't remove, grant or even waive rights - they exist or they don't and that's the end of it. (On matters of rights, I can be complete Kant.) Permissions are the only things you can ever add or subtract, and since everything in the Constitution has been put on hold or removed at some point, everything in the Constitution is a permission.
But that's fine with Americans, provided it's some other person getting these pseudo-rights pulled, ideally if they're from the wrong political party or the wrong ethnicity, or (in this case) testifying for the wrong side. People want THEIR freedoms and the freedoms of "their side" defended, "the other side" gets whatever crumbs are thrown their way. And that is why Americans are generally elitist snobs of an order pre-WW1 Europe could only ever have dreamed of aspiring to. Even the worst of the empires treated "non-conformists" better than that. Admittedly, in many cases, by shooting them, but that's still treating them as equals of dogs. The US treats undesirables (ie: those who don't think the same) as closer to pond scum.
Judges have never been taken seriously, when decisions run against the government. There have been Supreme Court rulings awarding the Black Hills to the Sioux. The government, for 50? 60? years has managed to ignore the ruling. That covers a lot of Presidents in both parties. But it includes an awful lot more voters, who (by re-electing anyone at all) are active supporters of the Government's refusal to follow SCOTUS orders. You are putting people in power who you know, full well, will not obey judge rulings they do not like. The upshot of that is that I have bugger-all pity when those same politicians disregard (or creatively interpret) rulings that happen to affect you or your "side". You're discovering the price of giving them that power.
The only downside is that everyone in the world, free or otherwise, must pay the price for these games. You have robbed us all of what liberty we had, just so that the right people were oppressed.
If notices are going to be sent and processed (then accepted/rejected automatically), followed by verbal fighting, why not simplify the whole exercise? Have a two-way corewars battle between the service and the complainant, winner decides if the link stays. Since the complainant is rarely the copyright holder and the service rarely provides the material, you can extend this to a four-way battle using crobots.
This is just as logical as the current system and has a higher probability of protecting original content from abusive/malign notices.
I'll agree on the -ert part, but think that ov-, cov- and perv- apply just about equally.
I've even seen some sort of training video on the dangers of being attacked by someone wielding soft fruit.
...has never changed. People should never be in a position where they can be pressured, cajoled or manipulated into suicide; doctors are improving but are still not very competent at figuring out the long-term prognosis for patients, the error rates are way too high. This should ALWAYS be balanced against indisputably terminal conditions, conditions where the person ceases to be meaningfully human, conditions of extreme suffering where the patient makes unconditionally clear (not just once but over time) that pain has to cease, one way or another, and any similar situation.
In other words, I don't trust doctors, patients or families to make perfect decisions. There has to be hard evidence that this case is typical, not atypical, hard evidence (within reason - coma patients don't talk much) that the patient has firmly decided on that path and it isn't the moment-to-moment feeling everyone gets on a regular basis, and hard evidence that conceding to fate is the last realistic untried option.
But if that evidence is beyond dispute, doctors are oath-bound to do least harm and painless death is less harm than agonizing death.
(I would never have made a good doctor. The older and moodier I get, the more I feel that someone invent hell, just so that idiots have somewhere to go.)
June 4th would be symbolically better, and it would piss off those responsible for a date being needed. Which just makes that day even better. Or we could combine it with veteran's day, as I imagine the veterans are a little unhappy at winning the battles but losing the war. Hey, if you fight for freedom and then have none, you lost.
I do have a plan, a plan built on solid foundations (although I must admit some of the support beams need work). A plan I have posted a few times, so am unsure why I should bother now. It is also a plan that would require a lot of people joining in, which is never going to happen these days - experimental governance was all the rage in the 1600s and 1800s, but people are too comfortable now. Better the devil you know, after all.
Nonetheless, if it could be started, I can absolutely guarantee it would give superior results to your approach.
The US has been expelled from UN bodies in the past, due to violations of international law. Although 95% symbolic, it hurts them in things like trade negotiations and bilateral agreements. In turn, that makes equally abusive but unsanctioned nations more attractive to business. That, in turn, hurts donations to politicians and tax revenues. Not necessarily by a lot, but name a politician who wants to spend less.
(Note: Tea Partiers and Libertarians want other people to spend less. They, themselves, are by far the worst of the pork barrel spenders.)
For other things to be thankful for, The Guardian has an excellent review of a documentary + book on the death squads the US military and intelligence run. For example, I'm thankful there are attempts to colonize Mars. I have better odds of a life away from such idiots, if I'm on a civilized world and they're stuck on Earth.
Direct proof? Yes. I have had sections of my DNA sequenced by 3 different groups, using 3 different sequencers, including the whole of my mitochondrial DNA and various chromosomes from my nucleic DNA (Y is the easiest one to get sequenced multiple ways).
So far, differences detected: Zero.
So they are giving me valid SNPs everywhere I can double check.
Can I confirm SNPs I haven't had tested elsewhere? No, but sequencers only cost around $15 million, so if you could loan me the money, I can easily verify the rest. I would even be happy to verify results for anyone else who asked, at cost in electricity to run a batch divided by the number in that batch. I'll deal with the other costs.
Now, how fair is that? Total verification of the raw data results from any genome research group.
In fact, I don't care who supplies the cash. The FDA can, for all I care. I will then give you verification of any of these start-ups for slightly under the cost of the run itself, complete with statistical confidence limits because no test is 100% perfect.
I'm willing to sacrifice my time and effort on building an ultraclean room and buying the rest of the hardware and enzymes needed to take you from one anecdotal account to a full, absolutely independent (including of any existing lab) facility that can provide an accurate scientific assessment in a manner that is compliant with my ethos of science verified by scientists and observed by the public, nothing hidden, nothing obscured, no trade secrets, no concealing statistics by hiding behind certainties.
The Gates Foundation could provide the initial funding, for all I care. And as a long time Slashdotter, I don't say that lightly.
So who's game? Reply with who you sent the grant/funding request to and how I can check the progress. Anyone?
Like I said, those people should travel to the top of the nearest tall building and jump. Preferably yesterday. I have no pity for such people, no compassion, no time. The world would be far better off if such people offed themselves. The world needs thinkers, it does not need reactors. Well, other than fusion reactors. We could do with a lot of those.
Anyone who thinks a 0.1% increased risk in a single SNP when dozens are involved, all of which have to show compatible mutations, means they will surely die of, well, anything is too stupid to live. Frankly, if that's your best objection, I think the test should be provided free across America and Britain. Get rid of at least some of the fools, rather than have them be a burden on society by running banks or whatever.
Good riddance to them. Hope the jumped-up twits die slowly.
(23&Me cannot tell you if you have a disease, or if you will ever get it, it can only give you very rudimentary statistical information on correlations. Anyone confused about correlations and causation should be provided with a HOWTO on suicide and appropriate implements.)
As for the studies, published peer-reviewed studies on genetics do vary in quality. That is why nobody considers them. Only metastudies matter. I would argue a metastudy with a sample size of 2 is low, but even getting a repeat study published is a major feat and genetics hasn't been cheap enough for long enough for the numbers of studies to grow. I like that 23&Me makes an effort to run their own study, giving them a metastudy size of 3, still low but getting into respectable territory.
But, again, what does it matter? Let us say that there are 10 SNPs required, and one SNP has a 0.1% increased risk but no others do, should you worry? One in a thousand chance above normal for anything is so close to normal that you would have to be one of the deserving suicides to think it worth concerning yourself with.
What does it diagnose? It doesn't tell you that you have condition X, it doesn't tell you that you will ever have condition X, and since correlation is NOT causation, it doesn't tell you that condition X is even possible.
I see no diagnosis, I see raw data and I see correlations (most of which are borderline noise level) but I see no statement to the effect that anything specific has happened, is happening or will ever happen. Where is the diagnosis?
That should be a simple one to answer, even for you.
I would agree with that. I would even agree to 23&Me putting more emphasis on the fact that it measures probabilities, not certainties. I certainly think they should quantify their confidence in a marker better than they do.
I also absolutely agree with you that information on life-threatening reactions to medications should be unhindered. If it is wrong, then the doctors either used an equally effective alternative (so, no loss) or gave you unnecessarily extensive monitoring (no harm done and everyone likes being fussed over occasionally). If it is right, it saved your life. There is no obvious downside.
Besides, doctors aren't going to go by your printouts for personalized medicine. They will use your raw data files, but any hospital with that level of expertise (and there won't be many) will have their own databases built from the existing literature and the hospital's experience. They will have actual geneticists on-hand or readily available via strong links with the local university. That is how these things work - and, yes, I am very aware of how regular (non-teaching) hospitals work with universities for mutual benefit. I've seen it, even been a part of the process.
So, who is 23&Me potentially misinforming? If not the patients or the practitioners, who is left?
A high level of confidence of a correlation between an SNP and a condition means that two independent genetic studies have produced the same results AND members of the site who have agreed to allow their data to be used for research purposes (not all do) show the same correlation.
Ok, I wouldn't do it that way, I would indicate reliability as a percentage asymptotic to 100% (you can never be certain) where their high confidence would be somewhere around 50%. But it's a good indication of reliability. Most associations are not marked as high confidence. They have either not been in a repeat study or there is no basis in the "natural" sample set from users. Such results should be ignored. They are given because there is a vague possibility and therefore interesting to those looking at the research process and trends. Users know this.
There is nothing to evaluate, except perhaps the lab conditions. The SNPs do not diagnose a condition, 23&Me do not offer any diagnosis, all that is given is a set of percentages relating to the correlation of an SNP with an effect. Correlation is NOT causation.
The FDA actively work with drug companies known to be carrying out unauthorized tests on unwilling human subjects in Africa. They have accepted results from US prisons, where again no consent was ever given. They encourage the suppression of negative results in journals. Corrupt? They are better than certain historical medical authorities, but not significantly.
(No, I don't give a damn that 1960s and 1970s America had no restrictions on non-consensual biological and chemical experiments for medical purposes. They'd been declared a crime against humanity two decades. And outsourcing to Africa is no excuse. Sorry, they are criminals. Whatever right they do should be seen in isolation from their wrongs. They are guilty, the heads should stand trial along with the heads of pharmaceutical companies indulging in data suppression or illegal human experimentation, and a medically sound alternative should be created.)
Then you're an idiot, because that is the only way to make any claim regarding effectiveness (positive or negative) at the individual gene level when we don't know all the genes, all the interactions of genes or the impact of "Junk DNA" (which we now know isn't junk but metadata).
23&Me do not diagnose. They identify SNPs that have a given (often very low) correlation with a condition. They state the current best guess at that correlation. That is it. There are no "tests" except for the SNPs themselves. And those will be accurate.
They ARE NOT telling you you have X, Y or Z. They are telling you that gene G is ASSOCIATED WITH an N% increased risk.
For chrissakes (or, as my kindle spellchecker insists, cheesecakes), at least bother to do basic research. Oh, that's right, the serious COMPETENT Slashdotters abandoned the site because of the overwhelming sheer stupidity out here.
I don't know why I bother. Anyone with an ounce of brain capacity can figure out that raised odds of 0.23% do not equal an absolute certainty of instant death. But, clearly, you and the FDA lack even that.
They're limiting it to images on the Wayback Machine where the levels of pink or black do not indicate things that might cause it to suddenly decide the human race needs obliterating.
Deadly mutagenesis! Happens naturally all the time. See any number of stories of fungi becoming deadly for no apparent reason.
We use it ourselves, as others have noted.
Actually, "deadly" is just code for a failure of your microbiome to process the organics. If you have crappy microflora, don't blame the UV. Blame your diet. If you have flexible, powerful microbes, you're going to be fine.
Translated: Eat well or die, Earthling scum! (Sorry, channelling my inner Sontaran.)
Router poisoning is an old attack, used many times. Sometimes deliberately, sometimes when a router goes bad.
BGP supports encryption, via certificates or shared secrets. People were supposed to have made the switch years ago, but I have posted time and again that this hasn't happened in practice.
I'd fix the problems myself, but that would mean risking being sued to oblivion by patent trolls. Most of what I outlined could be prototyped and demo'd in very short order, but if the return for my investment is financial death, what is the point?
I'd be happy to give action, not words, in a rational, fair environment. We don't have one right now, and short of setting up an office in Antarctica (which has not signed any patent treaties), I don't see how I can make it viable to operate in the system we do have.
A surprising amount is FOSS. I routinely get screamed at by irate scientists for listing their stuff of Freshm...freecode.