Functional MRI is able to indirectly measure brain activity via the blood oxygenation level dependent signal, so no, she's talking about MRI, specifically fMRI. Except what she's really talking about sounds like fNIRS, which is functional near infrared spectroscopy, which measures the BOLD signal using infrared diodes on the scalp.
But honestly I just have to laugh at the claims. fMRI is nowhere near being able to reliably read minds and fNIRS is even more limited in terms of spatial resolution. And that's totally ignoring that you can't input anything with fMRI or fNIRS, so you need some other technology to be able to stimulate the brain to actually have telepathy and that technology is even farther away than read-only.
Nope. I thought so as well, because that's what the statement in their press materials for it indicated to me. But the truth is it looks like the nag is based on a torque sensor or something on the steering wheel, so it only really nags you when it's going around curves above some threshold. There are videos of people filming their commute where they spend most of the time without their hands on the wheel.
The summary is incorrect. Only one of the 3 crashes had any fatalities. And so far as I'm aware is the only one yet to have confirmed that autopilot was engaged at the time of the accident. The second two accidents the drivers indicated they were using autopilot, but Tesla has not been able to confirm that yet (or at least has not publicly confirmed it).
None of the links in the summary even support the claim that there have been two fatalities.
All of the software packages tested in the article (AFNI, FSL, SPM) are open source, including the package the authors built to do massively parallel non-parametric permutation tests (BROCCOLI).
The paper has been available as a preprint for awhile now, and my lab has discussed it internally and I've also paid attention to outside coverage. The key issue that the paper reports is that false positive rates are two high for most existing software WHEN using a specific type of test under a specific set of conditions. They show that voxelwise familywise error (FWE) correction actually seems to work reasonably or even conservatively. Cluster level FWE correction (looking for groups of voxels that are active) fails when using a very liberal cluster-defining threshold, but works reasonably well when using a more stringent cluster defining threshold. It also says nothing about the performance of another very common correction method that is frequently used in fMRI studies (false discovery rate or FDR).
I'm not really sure how extensive the group of findings that these issues actually affect is, but it's certainly not 40,000 as is claimed in the paper's significance section. Many of the earlier papers (and even more recent) likely used uncorrected statistical tests, so are suspect for entirely different reasons from this issue. Of the ones that use correction, the findings in this paper only call into question the results for those that are using FWE cluster correction with a cluster defining threshold that is too liberal (likely > 0.001, the paper's findings suggest that at 0.001 the familywise error rate is in the ballpark of the desired 5%). Those using a cluster defining threshold of p=0.001 or lower are likely fine, and those using a different correction method like FDR are unknown as to my knowledge there isn't currently any similar paper on that correction method.
You can also check out this technical report by some other big names in imaging that basically says that this result is known and expected for overly liberal cluster defining thresholds: http://www.fil.ion.ucl.ac.uk/s...
You might want to read the article (or even the summary). Google is saying that their results suggest that these computers are NOT just doing simulated annealing, but rather true quantum annealing.
Though as one of the questions during the briefing asked, they don't actually know the mass. They're pretty much guessing based on the size by using the distribution of a smaller pool of past planets that they do have mass estimates for.
From the caption for figure 4 from their presentation:
"While its mass and composition are not yet determined, previous research suggests that planets the size of Kepler-452b have a better than even chance of being rocky."
Production of nanoparticle titanium dioxide is definitely hazardous. Titanium dioxide dust, present during production, is a possible carcinogen. Nanoparticle titanium dioxide has also been shown to cause genetic damage in mice. That said, I don't think there's evidence that the nanoparticles or powder present in paint pigment and sunscreen, for example, actually cause adverse effects in humans to this point, so it's unclear whether these coatings would be a concern or not.
Or, based on the timer going on the livestream now, they were streaming. But there was no apparent link from either the front page or yesterday's or today's story about watching it live.
That was essentially their support response to me when I recently tried to use Google Wallet. A particular feature was not showing up in the app. After a week of back and forth with their support the response was that my versions of Android and Google Wallet were too old and unsupported. Despite it being the version of Google Wallet I had just downloaded, and with no mention anywhere in their online help that the particular feature wasn't available in some versions.
Except that they usually don't deliver on Saturday, so ordering on Thursday means you won't get it until Monday. And ordering on Friday evening, Saturday, or Sunday means it doesn't show up until Wednesday because it gets processed as if it were a Monday order.
Considering how poor their shipping has been for me and a few others I've talked to over the past 6-10 months, I definitely would not be renewing if the price goes up pretty much at all. I never really use the streaming service, because the selection and ease of use on my devices is inferior to Netflix, so if their shipping gets any worse, I don't think it'll even be worth $79/year.
I can't tell if you're trying to be funny or not. The person reporting on his experiences is one of the developers of Aaaaaculus! which is essentially an Oculus compatible version of AaaaaAAaaaAAAaaAAAAaAAAAA!!!. In addition, they've worked on several sequels and ports of that game with the original developers.
It does however stipulate that the device can't display "a television broadcast or video signal that produces entertainment or business applications."
So a cell phone is definitely out unless:
(A) The equipment has an interlock device that, when the motor vehicle is driven, disables the equipment for all uses except as a visual display as described in paragraphs (1) to (4), inclusive.
(B) The equipment is designed, operated, and configured in a manner that prevents the driver of the motor vehicle from viewing the television broadcast or video signal while operating the vehicle in a safe and reasonable manner.
The bit at the end of the summary that suggests anyone with a PubMed/NCBI account can sign up are either wrong or they fixed it. I have an account and a number of publications, but no grants from the sources they're searching, and it doesn't allow me to request an invite or register with my email.
Make sure you really enjoy the research you'll be doing, because moving from a private programming position into an academic research programming position is going to come with a hefty pay cut.
Unfortunately for my wallet, I *do* enjoy the research quite a bit and have co-authored more publications than most junior faculty PhDs, so it can be very rewarding in a non-monetary way.
Though as others have said, if you're goal is to be conceiving of and performing your own research, you need to go through the typical channels of getting a PhD, doing a post-doc, and eventually finding a position somewhere as a professor. The research programmer will always be a research programmer and won't be running their own research lab or anything like that. Though doing a few years of research in a field can be a great way to get experience and figure out what you're interested in before going to grad school if you do want to go the academic route.
* Trade-in, lend, resell, gift, and rent disc based games just like you do today – There will be no limitations to using and sharing games, it will work just as it does today on Xbox 360.
The ability to do all those things just like it works now pretty clearly means they won't be using the one-time codes that they were planning on previously. Individual publishers still could try it, but it won't be the universal method that it sounds like Microsoft was planning originally.
A neuron doesn't "fire" the dendrites. The dendrites are inputs. It sends signal out through the axon, and each neuron only has a single axon (though it has multiple terminating ends that can connect to numerous other neurons).
If he gave an anonymous tip the NDA would still hold.
I just want to say that this is ridiculous. Reporting something anonymously does not mean he wouldn't have been violating the NDA. If that were the case, all NDAs ever would be completely pointless. Now, it might mean he doesn't get caught for violating the NDA, but it would still be a violation of it.
You assume that formatting refers only to whitespace, but given the larger context of the conversation, I believe he's using formatting to refer to the whole of the standards (variable name conventions, spacing, brace position, etc).
Slashdot Mirror
Yeah, that's pretty much how I read it too.
What could possibly go wrong... https://www.theregister.co.uk/...
Probably imperceptible to people, but clearly not imperceptible to machines since it was picked up as a magnitude 2 earthquake.
Functional MRI is able to indirectly measure brain activity via the blood oxygenation level dependent signal, so no, she's talking about MRI, specifically fMRI. Except what she's really talking about sounds like fNIRS, which is functional near infrared spectroscopy, which measures the BOLD signal using infrared diodes on the scalp. But honestly I just have to laugh at the claims. fMRI is nowhere near being able to reliably read minds and fNIRS is even more limited in terms of spatial resolution. And that's totally ignoring that you can't input anything with fMRI or fNIRS, so you need some other technology to be able to stimulate the brain to actually have telepathy and that technology is even farther away than read-only.
Nope. I thought so as well, because that's what the statement in their press materials for it indicated to me. But the truth is it looks like the nag is based on a torque sensor or something on the steering wheel, so it only really nags you when it's going around curves above some threshold. There are videos of people filming their commute where they spend most of the time without their hands on the wheel.
The summary is incorrect. Only one of the 3 crashes had any fatalities. And so far as I'm aware is the only one yet to have confirmed that autopilot was engaged at the time of the accident. The second two accidents the drivers indicated they were using autopilot, but Tesla has not been able to confirm that yet (or at least has not publicly confirmed it).
None of the links in the summary even support the claim that there have been two fatalities.
All of the software packages tested in the article (AFNI, FSL, SPM) are open source, including the package the authors built to do massively parallel non-parametric permutation tests (BROCCOLI).
The paper has been available as a preprint for awhile now, and my lab has discussed it internally and I've also paid attention to outside coverage. The key issue that the paper reports is that false positive rates are two high for most existing software WHEN using a specific type of test under a specific set of conditions. They show that voxelwise familywise error (FWE) correction actually seems to work reasonably or even conservatively. Cluster level FWE correction (looking for groups of voxels that are active) fails when using a very liberal cluster-defining threshold, but works reasonably well when using a more stringent cluster defining threshold. It also says nothing about the performance of another very common correction method that is frequently used in fMRI studies (false discovery rate or FDR).
I'm not really sure how extensive the group of findings that these issues actually affect is, but it's certainly not 40,000 as is claimed in the paper's significance section. Many of the earlier papers (and even more recent) likely used uncorrected statistical tests, so are suspect for entirely different reasons from this issue. Of the ones that use correction, the findings in this paper only call into question the results for those that are using FWE cluster correction with a cluster defining threshold that is too liberal (likely > 0.001, the paper's findings suggest that at 0.001 the familywise error rate is in the ballpark of the desired 5%). Those using a cluster defining threshold of p=0.001 or lower are likely fine, and those using a different correction method like FDR are unknown as to my knowledge there isn't currently any similar paper on that correction method.
You can also check out this technical report by some other big names in imaging that basically says that this result is known and expected for overly liberal cluster defining thresholds:
http://www.fil.ion.ucl.ac.uk/s...
You might want to read the article (or even the summary). Google is saying that their results suggest that these computers are NOT just doing simulated annealing, but rather true quantum annealing.
Though as one of the questions during the briefing asked, they don't actually know the mass. They're pretty much guessing based on the size by using the distribution of a smaller pool of past planets that they do have mass estimates for.
From the caption for figure 4 from their presentation: "While its mass and composition are not yet determined, previous research suggests that planets the size of Kepler-452b have a better than even chance of being rocky."
http://www.nasa.gov/keplerbrie...
Production of nanoparticle titanium dioxide is definitely hazardous. Titanium dioxide dust, present during production, is a possible carcinogen. Nanoparticle titanium dioxide has also been shown to cause genetic damage in mice. That said, I don't think there's evidence that the nanoparticles or powder present in paint pigment and sunscreen, for example, actually cause adverse effects in humans to this point, so it's unclear whether these coatings would be a concern or not.
Or, based on the timer going on the livestream now, they were streaming. But there was no apparent link from either the front page or yesterday's or today's story about watching it live.
So it's 11:04 EST now, and no live feed. I'm taking this to mean the clock was advanced past midnight, and we're all actually dead now.
That was essentially their support response to me when I recently tried to use Google Wallet. A particular feature was not showing up in the app. After a week of back and forth with their support the response was that my versions of Android and Google Wallet were too old and unsupported. Despite it being the version of Google Wallet I had just downloaded, and with no mention anywhere in their online help that the particular feature wasn't available in some versions.
Except that they usually don't deliver on Saturday, so ordering on Thursday means you won't get it until Monday. And ordering on Friday evening, Saturday, or Sunday means it doesn't show up until Wednesday because it gets processed as if it were a Monday order. Considering how poor their shipping has been for me and a few others I've talked to over the past 6-10 months, I definitely would not be renewing if the price goes up pretty much at all. I never really use the streaming service, because the selection and ease of use on my devices is inferior to Netflix, so if their shipping gets any worse, I don't think it'll even be worth $79/year.
I can't tell if you're trying to be funny or not. The person reporting on his experiences is one of the developers of Aaaaaculus! which is essentially an Oculus compatible version of AaaaaAAaaaAAAaaAAAAaAAAAA!!!. In addition, they've worked on several sequels and ports of that game with the original developers.
It does however stipulate that the device can't display "a television broadcast or video signal that produces entertainment or business applications." So a cell phone is definitely out unless: (A) The equipment has an interlock device that, when the motor vehicle is driven, disables the equipment for all uses except as a visual display as described in paragraphs (1) to (4), inclusive. (B) The equipment is designed, operated, and configured in a manner that prevents the driver of the motor vehicle from viewing the television broadcast or video signal while operating the vehicle in a safe and reasonable manner.
Every reference I've seen in regards to that talks only about the average Police response time being 58 minutes, not the EMS.
58 minutes, really? Funny then that the official Detroit website lists it as 17. http://www.detroitmi.gov/DetroitPerformanceDashboard.aspx
The bit at the end of the summary that suggests anyone with a PubMed/NCBI account can sign up are either wrong or they fixed it. I have an account and a number of publications, but no grants from the sources they're searching, and it doesn't allow me to request an invite or register with my email.
Oh, right, because travertine tile and free steak and lobster meals for non-patients is totally a good trade-off for how much we spend on healthcare.
Make sure you really enjoy the research you'll be doing, because moving from a private programming position into an academic research programming position is going to come with a hefty pay cut.
Unfortunately for my wallet, I *do* enjoy the research quite a bit and have co-authored more publications than most junior faculty PhDs, so it can be very rewarding in a non-monetary way.
Though as others have said, if you're goal is to be conceiving of and performing your own research, you need to go through the typical channels of getting a PhD, doing a post-doc, and eventually finding a position somewhere as a professor. The research programmer will always be a research programmer and won't be running their own research lab or anything like that. Though doing a few years of research in a field can be a great way to get experience and figure out what you're interested in before going to grad school if you do want to go the academic route.
The ability to do all those things just like it works now pretty clearly means they won't be using the one-time codes that they were planning on previously. Individual publishers still could try it, but it won't be the universal method that it sounds like Microsoft was planning originally.
A neuron doesn't "fire" the dendrites. The dendrites are inputs. It sends signal out through the axon, and each neuron only has a single axon (though it has multiple terminating ends that can connect to numerous other neurons).
If he gave an anonymous tip the NDA would still hold.
I just want to say that this is ridiculous. Reporting something anonymously does not mean he wouldn't have been violating the NDA. If that were the case, all NDAs ever would be completely pointless. Now, it might mean he doesn't get caught for violating the NDA, but it would still be a violation of it.
You assume that formatting refers only to whitespace, but given the larger context of the conversation, I believe he's using formatting to refer to the whole of the standards (variable name conventions, spacing, brace position, etc).