More and more research is suggesting that it isn't safe to text or even talk on our phones hands-free while driving,
No it isn't. Plenty of research is showing that driving distracted can lead to worse driving, but there is no data to suggest that this is leading to more accidents. This is basic statistics. If you want to show talking on phones hands free while driving is contributing to accidents, you need to compare the four cases {distracted, not-distracted}X{journey-completed-without-accident, journey-ended-with-accident} over many journeys randomly selected from the population of all journeys.
The so call 'research' is finding univariate associations with things that you can't string together into multivariate conclusions. Put simply, there is nothing to address how people adapt to driving distracted. Just as drunk drivers tend to drive within the speed limit and stop at stop signs to avoid being pulled over, phone users typically know they are distracted and compensate by picking the time they do it to lower complexity driving situations. This adaption hopelessly confounds any broad conclusions about the data that has been collected.
Road accidents have been going down while phone use has been increasing.
I saw the Martian 3 hours ago. At no point did I look at the screen and consider the resolution of the images. Whatever it was, it was sufficient to see people and stuff. The people and stuff interacted in way that rendered an adequate but not awesome movie.
The United States will accept the standard when the rest of the world ditches that stupid metric system and go back to real units.
You mean European units? Either way, metric or imperial, the units came from Europe. The internationally standardized units are the SI units and the US is signed up to that. It is what engineers of many disciplines use since it's the standard.
As far as the article is telling me, a dual core cpu is keeping up with a 8 core cpu.
That's pretty damn impressive, but only if this doesn't come at a price (decrease in battery life).
This is normal. The rationale for 8 slow cores is you can turn off the other 7 when you aren't using them. The aim is to win on power consumption. The world has not declared a winner yet.
>In Geekbench, the iPhone 6s Plus performed second only to Samsung's newest Galaxy models
So it came in second! Yay!
I'm not sure where you got your figures (since there is no citation, Yay!); but this article claims that the iPhone 6s "Obliterates" the competition. And the GeekBench 3 scores in that article would tend to support that claim.
I got my figures from the article. I see a headline proclaiming product A to be the best and I scroll down and the first figures I find are of product B being better.
4996, 4952, 4824 and 4799 are all bigger numbers than 4379, yet they put the 4379 first in the chart, whereas all the other entries in that chart are ranked by geekbench score. This was not an objective exposition of the data. Tufte would been spinning in his grave if he was dead.
The one I remember in particular was in Coed Rhedyn, so it's probably still there. Taking over 60 years to clear an eyesore is nothing the the local council.
I like Excel. It has a good user interface for editing tables. It has good cell formatting for representing currency. It has enough statistics to do basic stuff. Even when I'm pouring data into R or a python numpy script, I'll usually run it through excel to get the csv right.
I would like it a lot more if it had serious statistical functions, unlimited integers, GF arithmetic and a proper scripting language.
>Further, there is no indication that secp256k1 is compromised.
Check out the finding on fields of low characteristic presented to EuroCrypt in January. secp256k1 uses a binary curve, so the characteristic is 2. So it's bang in line for improvements in the attacks based on those findings.
Nope, the blockchain is tamper proof because: - the data is distributed to a large number of nodes - the validity of that data is determined by consensus between those nodes - the data is cryptographically signed
To spoof a blockchain record, you'd need to either: - control a majority of the nodes - break the cryptographic sign -> even with all the computing horsepower of the world at your disposal, this would take literally forever
What I was arguing before, is that to USE blockchain transactions, you don't need that much computing power. It's tied to mining in Bitcoin, but doesn't need to be.
Don't be so confident about the signing part. It's one IACR paper away from doom.
ignores almost everything that makes Bitcoin so secure.
Bitcoin uses secp256k1. It is barely secure today. It will likely fall in due course. That's a really stupid curve to use. That's why no one uses it for anything - except bitcoin.
I am a cryptography goon for a big company. I suspect it's more than just normal probe attempts and someone thinks there's more in my servers than there really is.
Slashdot Mirror
Yup. Pu. You can stand next to it. But don't breath it into your lungs.
More and more research is suggesting that it isn't safe to text or even talk on our phones hands-free while driving,
No it isn't. Plenty of research is showing that driving distracted can lead to worse driving, but there is no data to suggest that this is leading to more accidents. This is basic statistics. If you want to show talking on phones hands free while driving is contributing to accidents, you need to compare the four cases {distracted, not-distracted}X{journey-completed-without-accident, journey-ended-with-accident} over many journeys randomly selected from the population of all journeys.
The so call 'research' is finding univariate associations with things that you can't string together into multivariate conclusions. Put simply, there is nothing to address how people adapt to driving distracted. Just as drunk drivers tend to drive within the speed limit and stop at stop signs to avoid being pulled over, phone users typically know they are distracted and compensate by picking the time they do it to lower complexity driving situations. This adaption hopelessly confounds any broad conclusions about the data that has been collected.
Road accidents have been going down while phone use has been increasing.
I saw the Martian 3 hours ago. At no point did I look at the screen and consider the resolution of the images. Whatever it was, it was sufficient to see people and stuff. The people and stuff interacted in way that rendered an adequate but not awesome movie.
A small plug for Tahoe-LAFS.
It doesn't matter where it is. It uses cryptography to give you what you want. Mirror in many places including on your own machines for redundancy.
https://www.tahoe-lafs.org/tra...
And yet they fiddled the rank sorting. That's a VW move.
The United States will accept the standard when the rest of the world ditches that stupid metric system and go back to real units.
You mean European units?
Either way, metric or imperial, the units came from Europe. The internationally standardized units are the SI units and the US is signed up to that. It is what engineers of many disciplines use since it's the standard.
I didn't hold the geekbench. I was pointing out what the article said.
Or we could accept that tiny microphones and speakers don't work well in flat phones. Get rid of them and require the phone to be used with a headset.
As far as the article is telling me, a dual core cpu is keeping up with a 8 core cpu.
That's pretty damn impressive, but only if this doesn't come at a price (decrease in battery life).
This is normal. The rationale for 8 slow cores is you can turn off the other 7 when you aren't using them. The aim is to win on power consumption. The world has not declared a winner yet.
>In Geekbench, the iPhone 6s Plus performed second only to Samsung's newest Galaxy models
So it came in second! Yay!
I'm not sure where you got your figures (since there is no citation, Yay!); but this article claims that the iPhone 6s "Obliterates" the competition. And the GeekBench 3 scores in that article would tend to support that claim.
I got my figures from the article. I see a headline proclaiming product A to be the best and I scroll down and the first figures I find are of product B being better.
4996, 4952, 4824 and 4799 are all bigger numbers than 4379, yet they put the 4379 first in the chart, whereas all the other entries in that chart are ranked by geekbench score. This was not an objective exposition of the data. Tufte would been spinning in his grave if he was dead.
>In Geekbench, the iPhone 6s Plus performed second only to Samsung's newest Galaxy models
So it came in second! Yay!
The one I remember in particular was in Coed Rhedyn, so it's probably still there. Taking over 60 years to clear an eyesore is nothing the the local council.
>In the 50's, there used to be police call boxes. Also in the 50's, the TARDIS was a police call box. The police have since stopped using those boxes.
I remember seeing actual police call boxes in the 70s. Maybe it just took them over 20 years to remove them.
Doing GF arithmetic is probably an effective way of repelling girlfriends.
GF Arithmetic is Galois Field Arithmetic.
It's been the convention to mark an arithmetic expression with = since Visicalc. Visicalc is older than many Slashdotters.
I like Excel. It has a good user interface for editing tables. It has good cell formatting for representing currency. It has enough statistics to do basic stuff.
Even when I'm pouring data into R or a python numpy script, I'll usually run it through excel to get the csv right.
I would like it a lot more if it had serious statistical functions, unlimited integers, GF arithmetic and a proper scripting language.
According to the source the original article has been corrected. The Skylace-C was killed not the broadwell-c.
But they seem to have done a search and replace. The logic of the article evaporated.
>Further, there is no indication that secp256k1 is compromised.
Check out the finding on fields of low characteristic presented to EuroCrypt in January. secp256k1 uses a binary curve, so the characteristic is 2.
So it's bang in line for improvements in the attacks based on those findings.
There shouldn't be critical cables. There should be redundant paths to make the network tolerant to any individual cut.
Nope, the blockchain is tamper proof because:
- the data is distributed to a large number of nodes
- the validity of that data is determined by consensus between those nodes
- the data is cryptographically signed
To spoof a blockchain record, you'd need to either:
- control a majority of the nodes
- break the cryptographic sign -> even with all the computing horsepower of the world at your disposal, this would take literally forever
What I was arguing before, is that to USE blockchain transactions, you don't need that much computing power. It's tied to mining in Bitcoin, but doesn't need to be.
Don't be so confident about the signing part. It's one IACR paper away from doom.
ignores almost everything that makes Bitcoin so secure.
Bitcoin uses secp256k1. It is barely secure today. It will likely fall in due course. That's a really stupid curve to use. That's why no one uses it for anything - except bitcoin.
Try googling "koblitz curves are bad".
Burn in hell HOA Nazi
While I tend to agree. I'm the anti-HOA Nazi. I work to prevent the HOA doing anything beyond cutting the grass and maintaining insurance.
I am a cryptography goon for a big company. I suspect it's more than just normal probe attempts and someone thinks there's more in my servers than there really is.
Well strong passwords solve the primary problem.
The connection can handle it. But Fail2Ban is just one of the things in my life I haven't got around to yet.