This is known as the availability heuristic, and is a cognitive bias. http://en.wikipedia.org/wiki/Availability_heuristic It's related to "I hit every red light on the way here" and "this is a lucky casino, look at all these people winning".
Then Apple took ownership, trimmed it to three letters, and within months the word 'app' became synonymous with small widgets of code for smartphones.
Until the Xerox Alto UI came along, the word 'widget' largely meant a self-contained device, item or unit of production. Then Xerox took ownership, made longer terms such as "widget toolkits" and "widget APIs", and within months the word 'widget' became synonymous with small icons in apps coded for desktop computers. Now, Microsoft is pushing the boundaries of the 'widget' definition even further. IE users will have seen a new addition to their desktop wallpaper recently: the embedded IE browser. Here, you'll find dozens of 'widgets' to install and run directly from a handy icon on the desktop's preferences screen. Except, these aren't 'widgets' at all. They're webpages. Microsoft's idea of 'widgets' are what we quaintly referred to in the good old days as 'DHTML.' Does the word 'widget' mean anything at all any more?
"An installable web app can be a normal website with a bit of extra metadata; this type of app is called a hosted app. Alternatively, an installable web app can bundle all its content into an archive that users download when they install the app; this is a packaged app."
Also Chrome apps have an authentication and billing API that lets developers charge per access, by time or per install. This means apps don't have to do their own authentication or billing, they just use the API.
Apps also blur the line between extensions and websites, see the above link for more info.
The laser doesn't actually "move" on the moon. Different photons that are all moving at c hit the moon at different times, it's just that the point where they hit is changing. If you used a conical beam that was as wide as the moon at the distance of the moon from the earth, the individual photons in your beam will all hit the moon at different places at different times, but there is no motion inherent in one photon hitting in one place followed by another photon hitting in another place.
The assumptions in the article (and much of Big Bang theory) are wrong -- the only reason we can't see past 14 light years away from us is that the galaxies themselves are moving away from us at close to the speed of light at that distance -- the redshift tends to 1.0 at a distance of slightly less than 14 light years. We can't see past that point because the light will never reach us.
That doesn't make any sense either. c is defined as the speed of light in space. So if space expands relative to some imaginary non-expanding absolute reference frame, then c would have been traveling _slower_ relative to the non-expanding reference frame before the expansion -- so you could still only see the same distance before the expansion.
The second problem is the concept of a non-expanding absolute reference frame. There should be no such thing under the Big Bang model -- space didn't exist before the Big Bang. An observer can't observe space from a reference frame outside of space itself. So in fact it is impossible for there to be any expansion of space itself -- there can only be acceleration of matter within the space. (And that's all we observe now -- matter is accelerating away from other matter in the universe, for as-yet unknown reasons.)
I hate spammers, but I almost as strongly hate the fact that statutory damages can me several orders of magnitude higher than actual damages. (viz Jammie Thomas...)
I had to name friends one time for some stupid facebook game that I installed. I couldn't name more than half of them from photos. Probably 1/3rd were people I didn't know that well who friended me ("sure, whatever -- click") and 1/3rd were people I knew but whom I couldn't identify based on their profile photos.
=> All in all, a novel but (in practice) rather stupid idea.
So it's OK to protect the information of individuals from the government but it's not OK to protect the information of the government from individuals?
I mean, I agree with all that, but I don't really get the logic when I really think about it.
They have the authentication backwards. All anyone ever needs to do is take a screenshot of your account screen and plaster it all over the Internet, and anyone can order coffee on your tab. (Unless of course your app generates a different unique QR code for each transaction, which is unlikely...)
The way it should work is the cash register should display a QR code that contains the amount and the vendor account details to pay to, and you should scan it to initiate payment from your phone using some (not quite yet existant) secure mobile payment mechanism.
No, your graphics processor was not another 486 core. But the baseband proc on most modern phones is a locked ARM core that is perfectly capable of running non-baseband code if it were to be unlocked. May not be a good idea to mess with the radio firmware, but the point is most phones already have more than one core. (Some probably have other embedded ARM cores that are not as easy to unlock or get to as the baseband, such as in the phone's GPU, like the 486 example you give -- but the baseband is not too special.)
Phones have always been dual core, at least recently. The G1, for example, had (at least) two ARM cores. The only problem was, you couldn't actually run anything on the baseband processor "for wireless security reasons" and so that calls were smooth. But it was sitting idle most of the time. If anyone hacked the radio image they could probably have produced dual-core capability.
Of *course* ChromeOS will eventually run Android apps. There is no good reason not to. And Dalvik runs on both ARM and x86 today. Also Android phones will be able to use Chrome OS apps because they are HTML5 or Flash.
Well -- with Qt you trade off the royal pain of the API for the royal pain of Qt's native language. (Yes, I know there are bindings for other languages.)
Slashdot Mirror
Exactly. Or do you count the times where it rings without you looking at it? http://en.wikipedia.org/wiki/Availability_heuristic
This is known as the availability heuristic, and is a cognitive bias. http://en.wikipedia.org/wiki/Availability_heuristic It's related to "I hit every red light on the way here" and "this is a lucky casino, look at all these people winning".
http://www.metalev.org/2011/02/reality-check-on-future-of-ai-and.html
Then Apple took ownership, trimmed it to three letters, and within months the word 'app' became synonymous with small widgets of code for smartphones.
Until the Xerox Alto UI came along, the word 'widget' largely meant a self-contained device, item or unit of production. Then Xerox took ownership, made longer terms such as "widget toolkits" and "widget APIs", and within months the word 'widget' became synonymous with small icons in apps coded for desktop computers. Now, Microsoft is pushing the boundaries of the 'widget' definition even further. IE users will have seen a new addition to their desktop wallpaper recently: the embedded IE browser. Here, you'll find dozens of 'widgets' to install and run directly from a handy icon on the desktop's preferences screen. Except, these aren't 'widgets' at all. They're webpages. Microsoft's idea of 'widgets' are what we quaintly referred to in the good old days as 'DHTML.' Does the word 'widget' mean anything at all any more?
A Chrome app is not just a website. From http://code.google.com/chrome/webstore/docs/index.html#concepts:
"An installable web app can be a normal website with a bit of extra metadata; this type of app is called a hosted app. Alternatively, an installable web app can bundle all its content into an archive that users download when they install the app; this is a packaged app."
Also Chrome apps have an authentication and billing API that lets developers charge per access, by time or per install. This means apps don't have to do their own authentication or billing, they just use the API.
Apps also blur the line between extensions and websites, see the above link for more info.
The laser doesn't actually "move" on the moon. Different photons that are all moving at c hit the moon at different times, it's just that the point where they hit is changing. If you used a conical beam that was as wide as the moon at the distance of the moon from the earth, the individual photons in your beam will all hit the moon at different places at different times, but there is no motion inherent in one photon hitting in one place followed by another photon hitting in another place.
The assumptions in the article (and much of Big Bang theory) are wrong -- the only reason we can't see past 14 light years away from us is that the galaxies themselves are moving away from us at close to the speed of light at that distance -- the redshift tends to 1.0 at a distance of slightly less than 14 light years. We can't see past that point because the light will never reach us.
That doesn't make any sense either. c is defined as the speed of light in space. So if space expands relative to some imaginary non-expanding absolute reference frame, then c would have been traveling _slower_ relative to the non-expanding reference frame before the expansion -- so you could still only see the same distance before the expansion.
The second problem is the concept of a non-expanding absolute reference frame. There should be no such thing under the Big Bang model -- space didn't exist before the Big Bang. An observer can't observe space from a reference frame outside of space itself. So in fact it is impossible for there to be any expansion of space itself -- there can only be acceleration of matter within the space. (And that's all we observe now -- matter is accelerating away from other matter in the universe, for as-yet unknown reasons.)
I hate spammers, but I almost as strongly hate the fact that statutory damages can me several orders of magnitude higher than actual damages. (viz Jammie Thomas...)
That's 360 MILLION not thousand...
Doesn't matter. With the root key, his job is done.
Yes, Scala runs fine on Dalvik, as does anything that generates Java bytecodes (due to cross-compilation). Also check out the NDK.
I had to name friends one time for some stupid facebook game that I installed. I couldn't name more than half of them from photos. Probably 1/3rd were people I didn't know that well who friended me ("sure, whatever -- click") and 1/3rd were people I knew but whom I couldn't identify based on their profile photos. => All in all, a novel but (in practice) rather stupid idea.
I mean, I agree with all that, but I don't really get the logic when I really think about it.
They have the authentication backwards. All anyone ever needs to do is take a screenshot of your account screen and plaster it all over the Internet, and anyone can order coffee on your tab. (Unless of course your app generates a different unique QR code for each transaction, which is unlikely...) The way it should work is the cash register should display a QR code that contains the amount and the vendor account details to pay to, and you should scan it to initiate payment from your phone using some (not quite yet existant) secure mobile payment mechanism.
And... cue Slumdog Millionaire jokes...
What if you don't have five fingers?
No, your graphics processor was not another 486 core. But the baseband proc on most modern phones is a locked ARM core that is perfectly capable of running non-baseband code if it were to be unlocked. May not be a good idea to mess with the radio firmware, but the point is most phones already have more than one core. (Some probably have other embedded ARM cores that are not as easy to unlock or get to as the baseband, such as in the phone's GPU, like the 486 example you give -- but the baseband is not too special.)
Phones have always been dual core, at least recently. The G1, for example, had (at least) two ARM cores. The only problem was, you couldn't actually run anything on the baseband processor "for wireless security reasons" and so that calls were smooth. But it was sitting idle most of the time. If anyone hacked the radio image they could probably have produced dual-core capability.
It sucks that the plug fits only one way around.
Similar analysis here.
I posted some basic password statistics and ranked prevalence graphs here, if anyone is interested in seeing what sorts of passwords people use in the wild.
Of *course* ChromeOS will eventually run Android apps. There is no good reason not to. And Dalvik runs on both ARM and x86 today. Also Android phones will be able to use Chrome OS apps because they are HTML5 or Flash.
> 'words on paper can be made secure, electronic archives not.'
What about the Pentagon Papers? They were photocopied. Thousands of pages.
Well -- with Qt you trade off the royal pain of the API for the royal pain of Qt's native language. (Yes, I know there are bindings for other languages.)