Yes, they should have submitted it to a similar number of similarly ranked closed-access journals and seen if there's any difference due to the open access policies specifically. As it stands it's sort of interesting, but doesn't tell us squat about open access.
Problem is, at least in this trial they're reviewing already published code, when it's too late to gain much benefit from the review on the part of the original writer. A research project is normally time-limited after all; by the time the paper and data is public, the project is often done and people have moved on.
There's nobody with the time or inclination to, for instance, create and release a new improved version of the code at that point. And unless there's errors which lead to truly significant changes in the analysis, nobody would be willing to publish any kind of amended analysis either.
I'm very likely over-reading this, but my first reaction when seeing this was: don't learn to use Python and code your own stuff. Learn to use JS and code for the Google platform instead. Learn to become a tenant farmer.
As I'm not a US citizen and do not live in the US, it's all but certain that any political solution there will do nothing for me. And as you say, the NSA is not the only one listening anyhow. Making it too costly and difficult, and encouraging as many people as possible to do the same, is the way to go.
"Human spies will get around that as they always have."
Security has never been about _absolute_ security, but simply about making it too expensive, dangerous or time consuming for an adversary to bother. We don't all live in bank vaults, after all; we don't need that much security for the kind of possessions we keep at home.
Schneiers point is the same: we don't need so much security the NSA could never get to our data. We just need enough security - and need enough of us to use it - that the effort to routinely record what we all are up to exceeds their capability of doing so. They do not have an infinite budget, or infinite man-hours.
Make routine surveillance not impossible but too expensive, that's the name of the game.
Once you reach your late fourties, looking closer is not longer an option. Progressive glasses or bifocals are only a partial solution (as your eyes get worse you can't compensate fully), and a phone is something you're supposed to be using while out and about; switching to your close-up glasses to see your email is not a realistic option on your subway or anywhere.
Samsung has the best combination of price-performance and advertising cachet at the moment. This can all easily change; they have nothing to give them a stable lock on the market.
That's like saying almost all sex they've ever had was consensual and legal, so we really shouldn't blame them for the few cases of rape they committed.
What I'd really want is a small ARM-based board that's good for a low-power server; something that can run a simple web site, Tiny Tiny RSS and keep a few git repositories.
The boards we're seeing now are getting close; they have 1-2GB memory, networking and SATA interfaces. What's really missing is the software support over time. Unlike an embedded system you do want security updates and OS updates over time, so you really want a platform that is a regular target for a major distro, whether Red Hat, Ubuntu or someone else.
In what political universe do they imagine the people of the UK would be interested in giving operational control of a nuclear reactor in Somerset to a foreign government,
In the kind of universe where the one who pays for something also gets a say in it. But of course, the UK is free to pick up the tab in their stead and pony up the needed investment.
You can easily have an abundance of STEM people overall, and yet have a shortage of people in specific fields. The shortage is of course most likely in new and in growing fields, while surpluses are most likely in old and settled, or declining areas.
So, mismatch can easily explain the discrepancy without ascribing malicious intent to anybody (which is not to say there is none). Instead the problem really is the tension between learning a field and training for a specific job.
Seems US and European corporations are more and more insistent on finding workers that fit right into a specific job with little to no training*. Which seems good in the short term, but people with mostly job-specific training will have a much harder time retraining for a different kind of job when the winds inevitably change. They'll act as anchors for their employers, and collectively reduce the pool of qualified replacements if or when their employers decide to kick them to the curb.
I suspect that this practice is in fact bad in the short term as well; but since the effects across the life cycle of an employee are felt in very different parts of an organization it's not a waste that any one person will normally notice.
* Japanese corporations, on the other hand, go overboard in the other direction. They hire mostly or only new graduates for any career jobs, and you - and the company - generally don't even know what you will actually be doing once you start. They want to hire blank slates they can train and mold as they see fit.
Before you ask, my hands on IT job prevents me from working from home. No matter how hard you try you can't rack a server from your house...
But a maintenance robot could do it for you. That kind of light industrial environment is probably among the first to become robot-equipped. You have excellent control over the area and can adapt it as needed. The people you replace are skilled and well-paid. Unlike semi-public places such as hospitals there's (or will be) few to no humans, and especially no unauthorized humans, around. And unlike heavy industries, making errors won't lead to catastrophe.
Would that replace all admins? Of course not. But how many fewer admins would you need if people mostly didn't need to be on-site for each and every data center? You could concentrate most admin work for lots of data centers to a single site, with perhaps one or two roaming admins to go on-site when needed.
There's absolutely a lot of places where changing the lighting in sensible ways would make a major difference. Just having street lights and others shining downwards only can have a large impact, and save energy and money in the process.
But in large cities this particular battle is lost. Dense urban areas will be too bright no matter what you do, short of a war-like imposed blackout. Have street lights point downwards and there's still enough street area to lighten up any dust or particles in the air (and all that activity makes sure there's plenty of dust and stuff to reflect off as well).
Where I live (central Osaka), I can usually see a few of the brightest stars at night from our balcony, but it's a close thing. A long-exposure shot of the sky will pick up a sprinkling of bright stars, but the sky itself is bright enogh that it drowns out anything else.
False positives is as you say a huge issue. It ties in to the ridiculous airport screening programs, to the daft idea of facial recognition for street-side cameras, as well as recent developments in common screening programs such as for breast cancer.
Mostly I agree wholeheartedly. A close relative has survived colon cancer (one of the most survivable cancers today), in no small part because it was diagnosed and removed in time.
But, it seems that screening has some pretty tough limits. You have some cancers such as glioblastoma that seem to still be pretty much uniformly fatal no matter how early you find them. And breast cancer seems to be less promising for screening than it seemed at first; the aggressive type seems to be liable to have metastised almost no matter how early it is detected, while the other types are fairly unlikely to do so, even with late detection.
This seems to partly explain why breast cancer survival hasn't budged nearly as much as expected with the advent of wide-spread screening.
Screeining _is_ important. Surgery is the main means of cancer cure. And both have improved hugely over time. But for all that, cancers still collectively comprise the second most common cause of death, behind cardiovascular issues. And arguably a much more difficult and prolongued death for most sufferers. I'd say any improvement in treatment is both urgent and welcome.
"Hosting in Europe is exactly the same as hosting in the US!"
Not really. A US citizen does have legal rights regarding US spying that non-citizens do not, and can at least nominally ask for restitution, and use their vote and political action to influence the practice. Note how the widespread surveillance being exposed is illegal for targeting US citizens, but perfectly legal - and thus with no recourse available - for non-citizen targets.
In the same way, if you're a European citizen you really should prefer using hosting, email and other providers based in EU jurisditions. That will in the same way give rights and leverage to influence privacy issues that you are not given by US authorities.
So yes, the rational response to the widespread surveillance is unfortunately a balkanized internet, where we all mostly use and pay for services within our own legal bloc and avoid any companies based elsewhere.
Firefox for Android does reflow. Usually works well, but a few js-heavy sites can fail, and you end up with sidescrolling anyway. I just never visit such sites again; the internet is too big and life too short to put up with the annoyance.
I've seen gaming headsets that mix two audio sources (is yours the PS3 headset?), but I haven't found wireless earbuds available that do that (thinking more about it, it seems my problems are only with wireless earbuds (I've had sets from two manufacturers)).
I use the Sony SBH20 wireless earbuds: http://www.sonymobile.com/global-en/products/accessories/stereo-bluetooth-headset-sbh20/ Seems to work fine. I haven't had any problem with mixing up devices, but I'm pretty careful about turning off BT whenever I don't use it to save battery. So I turn on the headset (or keyboard), then turn on BT on the device I want to use. Then turn both off once I'm done.
I've got two Bluetooth devices; A Microsoft Wedge portable keyboard, and a Sony headset. Both trivially paired with both my Nexus 7 and Tablet Z (and my phone, for the headset) and are instantly available whenever I turn BT on on either device. The headset can connect to two devices at once too, which I guess is useful if you're, say, taking a call while playing a game on the tablet.
This is the first time I've heard of bluetooth devices that pair with only one device at a time.
Slashdot Mirror
Yes, they should have submitted it to a similar number of similarly ranked closed-access journals and seen if there's any difference due to the open access policies specifically. As it stands it's sort of interesting, but doesn't tell us squat about open access.
Good points all. Just a small addendum/correction: replace
"Ken Burns style"
With
"although not Ken Burns style, as that has been overused lately, and makes it harder to appreciate the image"
Problem is, at least in this trial they're reviewing already published code, when it's too late to gain much benefit from the review on the part of the original writer. A research project is normally time-limited after all; by the time the paper and data is public, the project is often done and people have moved on.
There's nobody with the time or inclination to, for instance, create and release a new improved version of the code at that point. And unless there's errors which lead to truly significant changes in the analysis, nobody would be willing to publish any kind of amended analysis either.
But unless the U.S. takes a huge turn to the left, disaster will result.
At which point the survivors|revolutionaries|a guy named Hank can pick up the pieces and build a new state that is better suited to the situation.
I'm very likely over-reading this, but my first reaction when seeing this was: don't learn to use Python and code your own stuff. Learn to use JS and code for the Google platform instead. Learn to become a tenant farmer.
How would you know they don't have that already?
As I'm not a US citizen and do not live in the US, it's all but certain that any political solution there will do nothing for me. And as you say, the NSA is not the only one listening anyhow. Making it too costly and difficult, and encouraging as many people as possible to do the same, is the way to go.
Security has never been about _absolute_ security, but simply about making it too expensive, dangerous or time consuming for an adversary to bother. We don't all live in bank vaults, after all; we don't need that much security for the kind of possessions we keep at home.
Schneiers point is the same: we don't need so much security the NSA could never get to our data. We just need enough security - and need enough of us to use it - that the effort to routinely record what we all are up to exceeds their capability of doing so. They do not have an infinite budget, or infinite man-hours.
Make routine surveillance not impossible but too expensive, that's the name of the game.
"Welcome to middle age! Here's your bifocals!"
Once you reach your late fourties, looking closer is not longer an option. Progressive glasses or bifocals are only a partial solution (as your eyes get worse you can't compensate fully), and a phone is something you're supposed to be using while out and about; switching to your close-up glasses to see your email is not a realistic option on your subway or anywhere.
Samsung has the best combination of price-performance and advertising cachet at the moment. This can all easily change; they have nothing to give them a stable lock on the market.
That's like saying almost all sex they've ever had was consensual and legal, so we really shouldn't blame them for the few cases of rape they committed.
Nobody said it'd be the user of the device that employs those circumvention methods.
The Beagle Boards look pretty good, but they lack a SATA controller and don't really carry enough memory. I'll keep an eye on them though.
What I'd really want is a small ARM-based board that's good for a low-power server; something that can run a simple web site, Tiny Tiny RSS and keep a few git repositories.
The boards we're seeing now are getting close; they have 1-2GB memory, networking and SATA interfaces. What's really missing is the software support over time. Unlike an embedded system you do want security updates and OS updates over time, so you really want a platform that is a regular target for a major distro, whether Red Hat, Ubuntu or someone else.
In the kind of universe where the one who pays for something also gets a say in it. But of course, the UK is free to pick up the tab in their stead and pony up the needed investment.
You can easily have an abundance of STEM people overall, and yet have a shortage of people in specific fields. The shortage is of course most likely in new and in growing fields, while surpluses are most likely in old and settled, or declining areas.
So, mismatch can easily explain the discrepancy without ascribing malicious intent to anybody (which is not to say there is none). Instead the problem really is the tension between learning a field and training for a specific job.
Seems US and European corporations are more and more insistent on finding workers that fit right into a specific job with little to no training*. Which seems good in the short term, but people with mostly job-specific training will have a much harder time retraining for a different kind of job when the winds inevitably change. They'll act as anchors for their employers, and collectively reduce the pool of qualified replacements if or when their employers decide to kick them to the curb.
I suspect that this practice is in fact bad in the short term as well; but since the effects across the life cycle of an employee are felt in very different parts of an organization it's not a waste that any one person will normally notice.
* Japanese corporations, on the other hand, go overboard in the other direction. They hire mostly or only new graduates for any career jobs, and you - and the company - generally don't even know what you will actually be doing once you start. They want to hire blank slates they can train and mold as they see fit.
But a maintenance robot could do it for you. That kind of light industrial environment is probably among the first to become robot-equipped. You have excellent control over the area and can adapt it as needed. The people you replace are skilled and well-paid. Unlike semi-public places such as hospitals there's (or will be) few to no humans, and especially no unauthorized humans, around. And unlike heavy industries, making errors won't lead to catastrophe.
Would that replace all admins? Of course not. But how many fewer admins would you need if people mostly didn't need to be on-site for each and every data center? You could concentrate most admin work for lots of data centers to a single site, with perhaps one or two roaming admins to go on-site when needed.
There's absolutely a lot of places where changing the lighting in sensible ways would make a major difference. Just having street lights and others shining downwards only can have a large impact, and save energy and money in the process.
But in large cities this particular battle is lost. Dense urban areas will be too bright no matter what you do, short of a war-like imposed blackout. Have street lights point downwards and there's still enough street area to lighten up any dust or particles in the air (and all that activity makes sure there's plenty of dust and stuff to reflect off as well).
Where I live (central Osaka), I can usually see a few of the brightest stars at night from our balcony, but it's a close thing. A long-exposure shot of the sky will pick up a sprinkling of bright stars, but the sky itself is bright enogh that it drowns out anything else.
False positives is as you say a huge issue. It ties in to the ridiculous airport screening programs, to the daft idea of facial recognition for street-side cameras, as well as recent developments in common screening programs such as for breast cancer.
Mostly I agree wholeheartedly. A close relative has survived colon cancer (one of the most survivable cancers today), in no small part because it was diagnosed and removed in time.
But, it seems that screening has some pretty tough limits. You have some cancers such as glioblastoma that seem to still be pretty much uniformly fatal no matter how early you find them. And breast cancer seems to be less promising for screening than it seemed at first; the aggressive type seems to be liable to have metastised almost no matter how early it is detected, while the other types are fairly unlikely to do so, even with late detection.
This seems to partly explain why breast cancer survival hasn't budged nearly as much as expected with the advent of wide-spread screening.
Screeining _is_ important. Surgery is the main means of cancer cure. And both have improved hugely over time. But for all that, cancers still collectively comprise the second most common cause of death, behind cardiovascular issues. And arguably a much more difficult and prolongued death for most sufferers. I'd say any improvement in treatment is both urgent and welcome.
"No one shall be subjected to arbitrary interference "
Nothing arbitrary about the mass surveilance. It's all quite deliberate and systematic. Your rights are well protected.
Not really. A US citizen does have legal rights regarding US spying that non-citizens do not, and can at least nominally ask for restitution, and use their vote and political action to influence the practice. Note how the widespread surveillance being exposed is illegal for targeting US citizens, but perfectly legal - and thus with no recourse available - for non-citizen targets.
In the same way, if you're a European citizen you really should prefer using hosting, email and other providers based in EU jurisditions. That will in the same way give rights and leverage to influence privacy issues that you are not given by US authorities.
So yes, the rational response to the widespread surveillance is unfortunately a balkanized internet, where we all mostly use and pay for services within our own legal bloc and avoid any companies based elsewhere.
Firefox for Android does reflow. Usually works well, but a few js-heavy sites can fail, and you end up with sidescrolling anyway. I just never visit such sites again; the internet is too big and life too short to put up with the annoyance.
I use the Sony SBH20 wireless earbuds: http://www.sonymobile.com/global-en/products/accessories/stereo-bluetooth-headset-sbh20/ Seems to work fine. I haven't had any problem with mixing up devices, but I'm pretty careful about turning off BT whenever I don't use it to save battery. So I turn on the headset (or keyboard), then turn on BT on the device I want to use. Then turn both off once I'm done.
I've got two Bluetooth devices; A Microsoft Wedge portable keyboard, and a Sony headset. Both trivially paired with both my Nexus 7 and Tablet Z (and my phone, for the headset) and are instantly available whenever I turn BT on on either device. The headset can connect to two devices at once too, which I guess is useful if you're, say, taking a call while playing a game on the tablet.
This is the first time I've heard of bluetooth devices that pair with only one device at a time.