Studies that aren't blinded properly (and this one looks like it wasn't) are shitty. Improper blinding makes anything that comes after meaningless, and has been responsible for some high profile, and very expensive mistakes.
Did you read the comments at the end? The reviewer echos most of my comments. A couple of the most serious:
The blinding was poorly done. I think the reviewer actually went to easy on them here. The pathology raters knew which rats were in group A and group B even if they didn't know what A and B referred to. The paper only says that the pathology reviewers were blinded... the veterinarians that performed the necropsies and the study leaders who selected slides for the pathologists are NOT claimed to be blinded at all. Those are serious issues, easily capable of causing the very difference between groups.
The study is very underpowered, and it seems unlikely a power calculation was performed.
Downtown in cities, particularly in the ones this thing was designed for, there usually isn't much direct sunlight. People somehow manage to survive driving past buildings that block the sun. And tunnels. My city built a bunch of tunnels with concrete sunshades, I guess to ease the transition into and out of them. Then the shades started falling on the traffic so they removed them. It wasn't a problem.
When you change lanes you need to check for cars in the lane you're changing into. Checking for a big ass 1200 person bus while you're at it shouldn't be an issue.
This thing is two lanes wide, which probably means at least four times the capacity per unit length. Plus it doesn't have to get stuck in traffic like regular busses so you're unlikely to be waiting for long.
Did you miss the part where he says WISE estimates agree with radar measurements much better than Myhrvold predicts they should? It's in the summary....
Someone who just got finished ripping researchers a new one for improperly using causal language probably should avoid absolute statements. I couldn't actually find the source paper, but it certainly doesn't sound like it was a quantitative meta-analysis. "The evidence is a bit of a mess" is a statement that could be defended with this kind of study, but that's not the same as saying that breakfast isn't important.
Characterizing javascript as object oriented is a bit of a stretch. It's object oriented in much the same way that C is. You can do it, but the language doesn't really do much to help out.
It is now, yes. In the future it's almost certain to be much easier. Releasing the complete genome seems to be a remarkable example of a lapse in judgement.
The argument works out the same though: right now it's hard to manufacture smallpox from a genome, so keeping a bit of actual virus around in case of need is probably a good idea. In the future it will probably be pretty easy to manufacture so not keeping it in easy to steal electronic form is probably a good idea. Except it looks like the latter option is now closed.
The entire story is based on a false premise anyway. "Scientists" don't control the fate of smallpox. The American, Russian, and probably a few other, militaries do. So the question should be "why don't the world's militaries destroy the last remaining smallpox samples?" To which the answer is pretty obvious.
No. We have no way of synthesizing life, or anything like it, from scratch. It is possible to create a synthetic virus from a genome sequence but you need some host cell to actually manufacture it for you.
Or maybe you just don't want to give Google extra help prosecuting psychological warfare on you.
Google's business is to get you to buy stuff their advertisers want you to buy. Both Google and the advertisers employ cutting edge psychology to try and manipulate you into do that.
IMDB benefits from a lot of people who think they're film critics. Some of them do give balanced reviews. Some of them are even good. Yes, they can be hard to find among the masses of people who are as you describe.
An even more interesting question is how does does IMDB and/or 538 know the gender of reviewers? The account setup doesn't seem to ask for that information. Maybe they demand it later?
Doesn't matter. Characterizing one demographic choosing to review a particular set of TV shows as "sabotage" is clickbait sensationalism at best. This is an opportunity for sites like IMDB to detect a gender based split and report it:
"'Sex and the City' received an overall 3.8 stars, but female reviewers rated it 4.7 stars."
Or you could get even more useful: "General Hospital received an overall 1.0 stars but the distribution of ratings is non-normal, suggesting that while most viewers thought it was a waste of airwaves, a small demographic really enjoys the program."
Plague wasn't wiped out centuries ago. There were 10 cases of it in the US in 2014, and anywhere from a few to tens of thousands of cases in other countries, depending on the country. It's rare in the first world because of public health measures, and few people in the west die from it because it's usually fairly easily treatable with antibiotics.
That's not the only reason your example is silly. Cost-benefit analysis is done on vaccines openly by independent scientists. A pharma company could try and push one through, but there are quite a few people who are fairly good at the simple math required to determine whether a vaccine is likely to be beneficial.
The tetanus vaccine is good for somewhere around 10 years (I think if you travel they recommend a booster every 5 to make sure). The Ontario vaccination schedule calls for a tetanus vaccine in infancy and a booster when you're 15. Adults are recommended to get one every 10 years.
Counter examples are trivially easy to find. Look no further than the bible, which is full of preventative injunctions, ranging from the inane to the potentially useful. Complete with sanctioned punishments for disobedience.
Slashdot Mirror
Studies that aren't blinded properly (and this one looks like it wasn't) are shitty. Improper blinding makes anything that comes after meaningless, and has been responsible for some high profile, and very expensive mistakes.
Did you read the comments at the end? The reviewer echos most of my comments. A couple of the most serious:
The blinding was poorly done. I think the reviewer actually went to easy on them here. The pathology raters knew which rats were in group A and group B even if they didn't know what A and B referred to. The paper only says that the pathology reviewers were blinded... the veterinarians that performed the necropsies and the study leaders who selected slides for the pathologists are NOT claimed to be blinded at all. Those are serious issues, easily capable of causing the very difference between groups.
The study is very underpowered, and it seems unlikely a power calculation was performed.
Doesn't seem like that big a problem.
Downtown in cities, particularly in the ones this thing was designed for, there usually isn't much direct sunlight. People somehow manage to survive driving past buildings that block the sun. And tunnels. My city built a bunch of tunnels with concrete sunshades, I guess to ease the transition into and out of them. Then the shades started falling on the traffic so they removed them. It wasn't a problem.
When you change lanes you need to check for cars in the lane you're changing into. Checking for a big ass 1200 person bus while you're at it shouldn't be an issue.
This thing is two lanes wide, which probably means at least four times the capacity per unit length. Plus it doesn't have to get stuck in traffic like regular busses so you're unlikely to be waiting for long.
Sure, just like summer in the northern hemisphere on Earth is the one day we call the summer solstice.
A bit more useful definition might be some period of time surrounding the equinoxes. Say from equinox to solstice, like we do it on Earth.
Did you miss the part where he says WISE estimates agree with radar measurements much better than Myhrvold predicts they should? It's in the summary....
Someone who just got finished ripping researchers a new one for improperly using causal language probably should avoid absolute statements. I couldn't actually find the source paper, but it certainly doesn't sound like it was a quantitative meta-analysis. "The evidence is a bit of a mess" is a statement that could be defended with this kind of study, but that's not the same as saying that breakfast isn't important.
Characterizing javascript as object oriented is a bit of a stretch. It's object oriented in much the same way that C is. You can do it, but the language doesn't really do much to help out.
It's javascript. Humor them.
Yeah right. Payroll NEVER makes any mistakes.
Glad you agree with me. The stats on food and agriculture weren't really necessary.
Yes, you liked to the smallpox genome before. That's why I said it was a remarkable lapse in judgement to release it.
It is now, yes. In the future it's almost certain to be much easier. Releasing the complete genome seems to be a remarkable example of a lapse in judgement.
The argument works out the same though: right now it's hard to manufacture smallpox from a genome, so keeping a bit of actual virus around in case of need is probably a good idea. In the future it will probably be pretty easy to manufacture so not keeping it in easy to steal electronic form is probably a good idea. Except it looks like the latter option is now closed.
The entire story is based on a false premise anyway. "Scientists" don't control the fate of smallpox. The American, Russian, and probably a few other, militaries do. So the question should be "why don't the world's militaries destroy the last remaining smallpox samples?" To which the answer is pretty obvious.
No. We have no way of synthesizing life, or anything like it, from scratch. It is possible to create a synthetic virus from a genome sequence but you need some host cell to actually manufacture it for you.
"There's no reason we couldn't just store smallpox on floppy disk."
Sure. Data is way harder to steal from secure facilities than physical samples.
Trained BSL-4 lab staff != random physician.
Sounds like a good deal. Commit fraud, get away with it, you're golden. Get caught, break even.
Or maybe you just don't want to give Google extra help prosecuting psychological warfare on you.
Google's business is to get you to buy stuff their advertisers want you to buy. Both Google and the advertisers employ cutting edge psychology to try and manipulate you into do that.
IMDB benefits from a lot of people who think they're film critics. Some of them do give balanced reviews. Some of them are even good. Yes, they can be hard to find among the masses of people who are as you describe.
An even more interesting question is how does does IMDB and/or 538 know the gender of reviewers? The account setup doesn't seem to ask for that information. Maybe they demand it later?
Doesn't matter. Characterizing one demographic choosing to review a particular set of TV shows as "sabotage" is clickbait sensationalism at best. This is an opportunity for sites like IMDB to detect a gender based split and report it:
"'Sex and the City' received an overall 3.8 stars, but female reviewers rated it 4.7 stars."
Or you could get even more useful: "General Hospital received an overall 1.0 stars but the distribution of ratings is non-normal, suggesting that while most viewers thought it was a waste of airwaves, a small demographic really enjoys the program."
A tensor is an array, possibly with dimension > 2 if you want to be picky. TensorFlow absolutely does use tensors.
Can't. They were vaccinated when they were children.
Plague wasn't wiped out centuries ago. There were 10 cases of it in the US in 2014, and anywhere from a few to tens of thousands of cases in other countries, depending on the country. It's rare in the first world because of public health measures, and few people in the west die from it because it's usually fairly easily treatable with antibiotics.
That's not the only reason your example is silly. Cost-benefit analysis is done on vaccines openly by independent scientists. A pharma company could try and push one through, but there are quite a few people who are fairly good at the simple math required to determine whether a vaccine is likely to be beneficial.
The tetanus vaccine is good for somewhere around 10 years (I think if you travel they recommend a booster every 5 to make sure). The Ontario vaccination schedule calls for a tetanus vaccine in infancy and a booster when you're 15. Adults are recommended to get one every 10 years.
Counter examples are trivially easy to find. Look no further than the bible, which is full of preventative injunctions, ranging from the inane to the potentially useful. Complete with sanctioned punishments for disobedience.
Or else you can't make use of public services. Like public schools.