Statistical probabilities tell you EVERYTHING ABOUT EVERYONE and yet NOTHING ABOUT ANYONE.
Yes, if you take 1,000 people like your daughter then of those who live to age 65 or whatever, just about 550 of them are going to get Alzheimer's.
But each individual will have an ACTUAL rate of the disease of exactly 0% or 100%, and that 55% chance actually gives you NO information about which you will be.
And without those gene variations, she still might have a 10% say chance of getting the disease.
Behind the screen the DM rolls the dice. You don't get to see the results of the roll. Anyone who has played a D&D like game, or something like World of Warcraft which is so dependent on dice (random number generators) for the outcome of events, will know that it's hopeless to think too much about what the next roll of the dice will bring, because when you're rolling a lot of numbers between 1 and 100 on a regular basis, you're going to get numbers like 1,2,3 and, 98, 99, and 100 all the time, and you absolutely cannot develop any sort of intuition based on probability.
So lots of people with 55% chances will not get the disease, and lots of people with 10% chances will.
Even when your risk factor is 95%, there's no guarantee, and you should not be "surprised" if you turn out to have it happen to you after you thought you could just round the probability to the nearest value of 0 or 100.
People hate uncertainty. Given a probability most people will NEED to decide at that point whether the event will happen or not, because they can't stand to go through life in suspense. They will ask you "Ok, so that means I will get the disease?" when it means nothing of the sort.
A 55% chance to get terrible disease by age 65 is just NOT a reason to change your lifestyle IMHO.
For a good dose of reality, take 100 people age 65 and have them get their 23 and Me tests done and watch while they laugh at all the things they were at higher risk for that they DIDN'T get and all the things they were at low risk for that they DID get. If you do this (even for one sample, give a 23 and Me gift certificate to an older relative and see how much their results make you worry less) chances are it will make you worry about probability a lot less.
Statistics are great for determining insurance rates and public health policy, but they DO NOT ACTUALLY GIVE YOU ANY INFORMATION about whether YOU as a single individual will come out one way or the other.
There are a few "completely penetrant" genetic diseases (hemophilia, etc.) where if you have the gene then you WILL GET the disease. But almost everything 23 and Me tells you is about probabilities which are much less than certain and honestly nothing to get too worked up about.
I've been out of the FPGA world for a while, but I would expect anything interesting in terms of a 3D chip to require a largeish FPGA chip, and if you have to use a $200-$2000 FPGA (with exotic surface-mount and power requirements) then this seems like maybe not that much of a win.
This is not about you and me and our health data, it's about some billionaires who (understandably) don't want to see what happened to Steve Jobs happen to them.
You can sit around and say this is Evil and whatnot, but honestly would any of us here behave any differently if we were in their position?
Note that the president of the new company Calico is Art Levinson, who is currently also Chairman of Genentech and *Apple* so there's a direct Steve Jobs link here and even a quote in the press release from Tim Cook:
Tim Cook, Chief Executive Officer of Apple, said: “For too many of our friends and family, life has been cut short or the quality of their life is too often lacking. Art is one of the crazy ones who thinks it doesn't have to be this way. There is no one better suited to lead this mission and I am excited to see the results.”
So this is all very science-fiction (both from a technology *and* a social point of view) and it will be fascinating to watch.
From a Science point of view this is really interesting because at the moment there are relatively few sources of funding for basic life-extension research, since much research money comes from the government, and making people live longer is no more politically acceptable than talking about population control. So if you're a fan of applied biological research and specifically life-extension technologies then you should be pretty excited by Google's move here.
Interesting times.
G.
P.S. I for one welcome our new genetically enhanced immortal geek overlords.
Given the option of taking point on some really hazardous operation yourself and letting a machine do it? I think it's stretching it to think that soldiers are going to start treating the machines as equals.
Besides, you can backup the robot's state onto a USB key, air-drop in a new one and restore its "personality" (such as it is) and your "friend" is back from the dead.
Very little requirement for sympathetic emotional attachment.
Ignore the parts that are differently visible and the color differences, and focus on the parts that are the same in both images.
You'll see that the elements from the earlier photo have moved away from the center of the nebula and this is visible relative to all the background stars.
True. Depending on the processes involved, the chance that a live intact HIV virus inadvertently makes it through the system into a patient is probably greater than getting home from the rug store to find out that a live bear made it through the rug making process.
But on the other hand one of the researchers involved (or even TFA) might be able to explain to your satisfaction that the chances of these two different events are actually quite similar due to the methods being employed to produce the synthetic biology product.
"Session" was the wrong term to use. It's the second time they're offering the class, and it's the complete class from the beginning, not a "part 2" as might have been suggested by my poor word choice:)
MITx is offering the second session of their free massively open course 7.00x on Introduction to Biology - The Secret of Life taught by one of the best teachers I have ever listened to, Eric Lander of MIT, which starts on Sept 10th:
This class is mostly about the molecular biology machinery that makes cells work, and it should be fascinating to anyone who finds the way computers work interesting because most of what goes on at the cellular level is actually information processing and digital operations (though based on stochastic principles).
Warning: this class might make you want to (or wish you could) change your career path...
These stories always play off the "we're using scary HIV to cure you" angle, but they're just using components stolen from the HIV virus as tools or building-blocks to make something useful.
You might as well write a story that portrays bear-skin rugs as scary and dangerous because they were once part of a whole live bear.
And actually pretty much all of our recombinant DNA tools as well as many drugs like antibiotics are simply ancient things we stole from bacteria and other life forms. Somewhat annoyingly, Nature and 3+ billion years of evolution are still a lot better than we are at inventing things.
My definition of modern biology that I use to introduce it to computer people is: Hacking into ancient alien computer systems (stochastic digital computers not designed by the mind of man) to look for technology we can steal to cure cancer, solve world hunger, and produce renewable energy as well as whatever else we discover along the way.
In almost any skill that has to be learned, there's often a fairly rapid and abrupt transition from "I can't do that" to "I CAN do that and since I now know how to it's actually easy".
I think a lot of people get discouraged when they're unable to get through that transition on their own the first time they try it, and "I can't do that right" can be appear to be an impossible mountain to climb, even if you're not far from the top.
I think we need to be challenging kids from an early age to learn things that are "hard" so that they become intimately familiar with this progression from impossible to trivial. Too often I see kids these days try something that looks interesting to them a couple times and then decide "nah, that's too hard" and quit.
It's not specifically teaching perseverance, but more about learning to recognize that progress is almost never linear toward a goal and many times you won't recognize you've reached your goal until you're actually there.
Additionally, we ought to be able to get better at helping people fight through these places they get stuck, rather than just leaving them with a failing grade in a math class and a feeling that that they're not up to the task. Early recognition of students who are having difficulty and focused tutoring and other help getting through the hard parts to the point that they achieve their needed breakthrough.
I don't think any undergraduate subject should be so inherently difficult that anyone who can get into the university in the first place shouldn't be able to do well in it.
When you have 600 billion dollars in sales, an exotic technology that gives you a tenth of a percent more profit is worth $600,000,000, so honestly the thinner your profit margins are, the higher your cash-flow must be (or you wouldn't bother with the business to begin with) so such businesses will probably always be early tech / efficiency adopters and will always be pushing the boundaries of what's possible.
If I had some new tech that was applicable, these are exactly the customers I would seek out.
I think his results could be explained if the calculation of the final mark in a subject area involve some dodgy math to scale the result such that some intermediate step compresses the possible result to a discrete range of say 50 or so values which are then scaled / normalized to a 0-100 range. This expansion will result in every other final score value being impossible to obtain.
They may be scoring different parts of the exams with different weights, and then combining and scaling the results together, and I could imagine that process could produce the distribution he's seeing even without malicious intent of some sort.
This seems much more likely than some conspiracy to adjust grades which managed to produce so specific a set of results. The testing board may be playing with the overall weighting and projection of the raw scores onto a final normalized 0-100 range, but then that's what such organizations do to try to account for variations in the test questions from year to year, and I think in the US the SAT people do very similar things.
In other words, his data don't immediately indicate any per-student grade manipulation that I can see. The author is also a bit too proud of his accomplishment and indicates that he's clever, but perhaps also a bit young and naive.
I note that the little digital clock on my desk does not need a 1TB disk drive full of software in order to operate either.
Constructing a large mammalian brain complete with things like "instincts" might well make use of non-protein-coding information of some sort.
One thing about biology and the functioning of cells that you learn pretty quick is that "if it can happen then it probably does", and this is a very strong argument against writing off anything that appears to be conserved as "useless".
Simply finding an organism that itself has no need of other information simply says that it's not a universal requirement, and doesn't really tell you anything about whether other organisms might have found a use for it.
The very-long-term future of mobile human/computer interaction is ultimately some kind of implant that can feed your visual system directly, since people want more and more information but a 30" screen won't fit into a phone.
But along the road to that science-fiction future we're going to go through a lot of external projection devices like Google Glass, since they can give you (eventually) that big display without the big form-factor.
The problem with Glass is that it comes from Google who are pushing its social and camera/augmented reality features, and nobody likes the idea of dorky-looking people wearing dorky glasses pushing a camera in their face all the time, and so there is a good chance that this will stigmatize all similar devices for years to come, at least until a device can be made unobtrusive (and probably without a camera, or at least with a big red "recording" light to make people comfortable).
Or I guess we may somehow become comfortable with the idea that cameras are just everywhere and you always have to assume you're being recorded from multiple angles at all times.
Glass will be similar to the Segway which never really caught on partly because it's too dorky and draws too much attention to the user to the point that they feel too self-conscious to leave the house with it.
While walking through the Old Orchard Apple Store recently I was interested to see the variety of non-Apple products on display, which included things like Philips remote-controlled LED mood lighting systems, and the Nike Fuel Bands.
Because of the presence of the Fuel products, as well as Apple's previous integration of Nike stuff in their iPods, I am slightly more inclined to believe in the idea of an Apple "wrist" device than most of the Apple rumors that I hear.
So I haven't RTFA but the "best programmer in the building" thing, and the fact that it's Verizon, a telecommunications company who would be a prime target for foreign intelligence, makes me think that maybe we're giving the employee too much credit for being a genius here.
Which is more likely, that he sought out and found a good, cheap, reliable programmer, and then went on to expand his scheme into multiple companies, or instead perhaps that someone sought him out, and suggested the scheme to him, and maybe later said "Hey, I have more free time and some friends here, so maybe if you applied for jobs at some other companies that your resume matches we could make you even more money!". But I'm sure he would have been made to feel like it was all his idea as much as possible.
...why high speed trading is a good idea for anyone? It seems like the equivalent of slash-and-burn agriculture where you're destroying a resource (in this case basically sanity) in exchange for a one-time benefit of briefly being faster than your competitors.
So can someone explain how the world is a better place than if, say, you could only issue one trade per second?
The linked SecurityWeek articles includes the quote:
“We were able to immediately fix the hole and upgraded our hashing mechanisms from sha-256 with random salts to bcrypt to fortify security."
Which suggests that they were indeed salting the passwords. Assuming this was actually done, and done in a reasonable manner, then in theory there should actually be little or no risk from this breach I would think. But then I don't know why they would feel the need to immediately replace their hashing mechanism...
Unfortunately, playing GW1 will tell you almost nothing about GW2.
GW1 was a very unique game, designed before WoW basically defined what the standard MMOG interface was going to look like, and it works very differently compared to most other games, and especially when compared to GW2 which is much closer to WoW (in good ways rather than just copying stuff as most other games have done in the last ten years).
GW1 was an annoying, frustrating game for many people who tried it over the years, so I would NOT recommend it for someone who now is thinking they might be interested in GW2, because GW2 is almost totally different in every way.
Slashdot Mirror
This, lol. I would mod you funny by have no points today, sorry :)
G.
Clearly they need to hire someone named Gil Bates, that would be awesome.
G.
Statistical probabilities tell you EVERYTHING ABOUT EVERYONE and yet NOTHING ABOUT ANYONE.
Yes, if you take 1,000 people like your daughter then of those who live to age 65 or whatever, just about 550 of them are going to get Alzheimer's.
But each individual will have an ACTUAL rate of the disease of exactly 0% or 100%, and that 55% chance actually gives you NO information about which you will be.
And without those gene variations, she still might have a 10% say chance of getting the disease.
Behind the screen the DM rolls the dice. You don't get to see the results of the roll. Anyone who has played a D&D like game, or something like World of Warcraft which is so dependent on dice (random number generators) for the outcome of events, will know that it's hopeless to think too much about what the next roll of the dice will bring, because when you're rolling a lot of numbers between 1 and 100 on a regular basis, you're going to get numbers like 1,2,3 and, 98, 99, and 100 all the time, and you absolutely cannot develop any sort of intuition based on probability.
So lots of people with 55% chances will not get the disease, and lots of people with 10% chances will.
Even when your risk factor is 95%, there's no guarantee, and you should not be "surprised" if you turn out to have it happen to you after you thought you could just round the probability to the nearest value of 0 or 100.
People hate uncertainty. Given a probability most people will NEED to decide at that point whether the event will happen or not, because they can't stand to go through life in suspense. They will ask you "Ok, so that means I will get the disease?" when it means nothing of the sort.
A 55% chance to get terrible disease by age 65 is just NOT a reason to change your lifestyle IMHO.
For a good dose of reality, take 100 people age 65 and have them get their 23 and Me tests done and watch while they laugh at all the things they were at higher risk for that they DIDN'T get and all the things they were at low risk for that they DID get. If you do this (even for one sample, give a 23 and Me gift certificate to an older relative and see how much their results make you worry less) chances are it will make you worry about probability a lot less.
Statistics are great for determining insurance rates and public health policy, but they DO NOT ACTUALLY GIVE YOU ANY INFORMATION about whether YOU as a single individual will come out one way or the other.
There are a few "completely penetrant" genetic diseases (hemophilia, etc.) where if you have the gene then you WILL GET the disease. But almost everything 23 and Me tells you is about probabilities which are much less than certain and honestly nothing to get too worked up about.
G.
I've been out of the FPGA world for a while, but I would expect anything interesting in terms of a 3D chip to require a largeish FPGA chip, and if you have to use a $200-$2000 FPGA (with exotic surface-mount and power requirements) then this seems like maybe not that much of a win.
G.
This is not about you and me and our health data, it's about some billionaires who (understandably) don't want to see what happened to Steve Jobs happen to them.
You can sit around and say this is Evil and whatnot, but honestly would any of us here behave any differently if we were in their position?
Note that the president of the new company Calico is Art Levinson, who is currently also Chairman of Genentech and *Apple* so there's a direct Steve Jobs link here and even a quote in the press release from Tim Cook:
Tim Cook, Chief Executive Officer of Apple, said: “For too many of our friends and family, life has been cut short or the quality of their life is too often lacking. Art is one of the crazy ones who thinks it doesn't have to be this way. There is no one better suited to lead this mission and I am excited to see the results.”
So this is all very science-fiction (both from a technology *and* a social point of view) and it will be fascinating to watch.
From a Science point of view this is really interesting because at the moment there are relatively few sources of funding for basic life-extension research, since much research money comes from the government, and making people live longer is no more politically acceptable than talking about population control. So if you're a fan of applied biological research and specifically life-extension technologies then you should be pretty excited by Google's move here.
Interesting times.
G.
P.S. I for one welcome our new genetically enhanced immortal geek overlords.
Given the option of taking point on some really hazardous operation yourself and letting a machine do it? I think it's stretching it to think that soldiers are going to start treating the machines as equals.
Besides, you can backup the robot's state onto a USB key, air-drop in a new one and restore its "personality" (such as it is) and your "friend" is back from the dead.
Very little requirement for sympathetic emotional attachment.
G.
Ignore the parts that are differently visible and the color differences, and focus on the parts that are the same in both images.
You'll see that the elements from the earlier photo have moved away from the center of the nebula and this is visible relative to all the background stars.
G.
...and post on Facebook!
G.
(though it was more fun to light up the carrier detect indicator on old 300 baud modems this way)
True. Depending on the processes involved, the chance that a live intact HIV virus inadvertently makes it through the system into a patient is probably greater than getting home from the rug store to find out that a live bear made it through the rug making process.
But on the other hand one of the researchers involved (or even TFA) might be able to explain to your satisfaction that the chances of these two different events are actually quite similar due to the methods being employed to produce the synthetic biology product.
G.
"Session" was the wrong term to use. It's the second time they're offering the class, and it's the complete class from the beginning, not a "part 2" as might have been suggested by my poor word choice :)
G.
MITx is offering the second session of their free massively open course 7.00x on Introduction to Biology - The Secret of Life taught by one of the best teachers I have ever listened to, Eric Lander of MIT, which starts on Sept 10th:
https://www.edx.org/course/mit/7-00x/introduction-biology-secret-life/1014
This class is mostly about the molecular biology machinery that makes cells work, and it should be fascinating to anyone who finds the way computers work interesting because most of what goes on at the cellular level is actually information processing and digital operations (though based on stochastic principles).
Warning: this class might make you want to (or wish you could) change your career path...
G.
These stories always play off the "we're using scary HIV to cure you" angle, but they're just using components stolen from the HIV virus as tools or building-blocks to make something useful.
You might as well write a story that portrays bear-skin rugs as scary and dangerous because they were once part of a whole live bear.
And actually pretty much all of our recombinant DNA tools as well as many drugs like antibiotics are simply ancient things we stole from bacteria and other life forms. Somewhat annoyingly, Nature and 3+ billion years of evolution are still a lot better than we are at inventing things.
My definition of modern biology that I use to introduce it to computer people is: Hacking into ancient alien computer systems (stochastic digital computers not designed by the mind of man) to look for technology we can steal to cure cancer, solve world hunger, and produce renewable energy as well as whatever else we discover along the way.
G.
In almost any skill that has to be learned, there's often a fairly rapid and abrupt transition from "I can't do that" to "I CAN do that and since I now know how to it's actually easy".
I think a lot of people get discouraged when they're unable to get through that transition on their own the first time they try it, and "I can't do that right" can be appear to be an impossible mountain to climb, even if you're not far from the top.
I think we need to be challenging kids from an early age to learn things that are "hard" so that they become intimately familiar with this progression from impossible to trivial. Too often I see kids these days try something that looks interesting to them a couple times and then decide "nah, that's too hard" and quit.
It's not specifically teaching perseverance, but more about learning to recognize that progress is almost never linear toward a goal and many times you won't recognize you've reached your goal until you're actually there.
Additionally, we ought to be able to get better at helping people fight through these places they get stuck, rather than just leaving them with a failing grade in a math class and a feeling that that they're not up to the task. Early recognition of students who are having difficulty and focused tutoring and other help getting through the hard parts to the point that they achieve their needed breakthrough.
I don't think any undergraduate subject should be so inherently difficult that anyone who can get into the university in the first place shouldn't be able to do well in it.
G.
...what a shitty storage medium DVDs are these days.
A cheap 3TB disk drive burned to DVDs will produce a rather unwieldy tower of disks as well.
G.
When you have 600 billion dollars in sales, an exotic technology that gives you a tenth of a percent more profit is worth $600,000,000, so honestly the thinner your profit margins are, the higher your cash-flow must be (or you wouldn't bother with the business to begin with) so such businesses will probably always be early tech / efficiency adopters and will always be pushing the boundaries of what's possible.
If I had some new tech that was applicable, these are exactly the customers I would seek out.
G.
I think his results could be explained if the calculation of the final mark in a subject area involve some dodgy math to scale the result such that some intermediate step compresses the possible result to a discrete range of say 50 or so values which are then scaled / normalized to a 0-100 range. This expansion will result in every other final score value being impossible to obtain.
They may be scoring different parts of the exams with different weights, and then combining and scaling the results together, and I could imagine that process could produce the distribution he's seeing even without malicious intent of some sort.
This seems much more likely than some conspiracy to adjust grades which managed to produce so specific a set of results. The testing board may be playing with the overall weighting and projection of the raw scores onto a final normalized 0-100 range, but then that's what such organizations do to try to account for variations in the test questions from year to year, and I think in the US the SAT people do very similar things.
In other words, his data don't immediately indicate any per-student grade manipulation that I can see. The author is also a bit too proud of his accomplishment and indicates that he's clever, but perhaps also a bit young and naive.
G.
So what's the best alternative to gmail thse days?
Actually it's prey will simply evolve tiny little towels to wrap around their heads.
G.
I note that the little digital clock on my desk does not need a 1TB disk drive full of software in order to operate either.
Constructing a large mammalian brain complete with things like "instincts" might well make use of non-protein-coding information of some sort.
One thing about biology and the functioning of cells that you learn pretty quick is that "if it can happen then it probably does", and this is a very strong argument against writing off anything that appears to be conserved as "useless".
Simply finding an organism that itself has no need of other information simply says that it's not a universal requirement, and doesn't really tell you anything about whether other organisms might have found a use for it.
G.
The very-long-term future of mobile human/computer interaction is ultimately some kind of implant that can feed your visual system directly, since people want more and more information but a 30" screen won't fit into a phone.
But along the road to that science-fiction future we're going to go through a lot of external projection devices like Google Glass, since they can give you (eventually) that big display without the big form-factor.
The problem with Glass is that it comes from Google who are pushing its social and camera/augmented reality features, and nobody likes the idea of dorky-looking people wearing dorky glasses pushing a camera in their face all the time, and so there is a good chance that this will stigmatize all similar devices for years to come, at least until a device can be made unobtrusive (and probably without a camera, or at least with a big red "recording" light to make people comfortable).
Or I guess we may somehow become comfortable with the idea that cameras are just everywhere and you always have to assume you're being recorded from multiple angles at all times.
Glass will be similar to the Segway which never really caught on partly because it's too dorky and draws too much attention to the user to the point that they feel too self-conscious to leave the house with it.
G.
While walking through the Old Orchard Apple Store recently I was interested to see the variety of non-Apple products on display, which included things like Philips remote-controlled LED mood lighting systems, and the Nike Fuel Bands.
Because of the presence of the Fuel products, as well as Apple's previous integration of Nike stuff in their iPods, I am slightly more inclined to believe in the idea of an Apple "wrist" device than most of the Apple rumors that I hear.
G.
So I haven't RTFA but the "best programmer in the building" thing, and the fact that it's Verizon, a telecommunications company who would be a prime target for foreign intelligence, makes me think that maybe we're giving the employee too much credit for being a genius here.
Which is more likely, that he sought out and found a good, cheap, reliable programmer, and then went on to expand his scheme into multiple companies, or instead perhaps that someone sought him out, and suggested the scheme to him, and maybe later said "Hey, I have more free time and some friends here, so maybe if you applied for jobs at some other companies that your resume matches we could make you even more money!". But I'm sure he would have been made to feel like it was all his idea as much as possible.
G.
...why high speed trading is a good idea for anyone? It seems like the equivalent of slash-and-burn agriculture where you're destroying a resource (in this case basically sanity) in exchange for a one-time benefit of briefly being faster than your competitors.
So can someone explain how the world is a better place than if, say, you could only issue one trade per second?
G.
The linked SecurityWeek articles includes the quote:
“We were able to immediately fix the hole and upgraded our hashing mechanisms from sha-256 with random salts to bcrypt to fortify security."
Which suggests that they were indeed salting the passwords. Assuming this was actually done, and done in a reasonable manner, then in theory there should actually be little or no risk from this breach I would think. But then I don't know why they would feel the need to immediately replace their hashing mechanism...
G.
Unfortunately, playing GW1 will tell you almost nothing about GW2.
GW1 was a very unique game, designed before WoW basically defined what the standard MMOG interface was going to look like, and it works very differently compared to most other games, and especially when compared to GW2 which is much closer to WoW (in good ways rather than just copying stuff as most other games have done in the last ten years).
GW1 was an annoying, frustrating game for many people who tried it over the years, so I would NOT recommend it for someone who now is thinking they might be interested in GW2, because GW2 is almost totally different in every way.
Not even just for picking up the lore.
G.