That's not enough, to a large degree. It must also be designed so that no peripheral outside of the CPU is trusted, if you're going that far. Hard drives, network peripherals,... all today have CPUs of their own, usually with entirely secret firmware, and often access to the bus.
If you're not doing a design where you cost the various design alternatives, you're doing it wrong, and may even not minimise the thing you're 'caring most' about.
It is always - for example - possible to improve safety of cars through expensive technical means. More complex construction with better crash absorbtion properties,... But, if your vehicle is twice as safe as the rest of the fleet, and yet due to your safety upgrades costs eight times more than one that is only 1.1* as safe, you may actually end up costing lives with your 'safer' design.
Similarly - weight generally means poorer gas mileage, which means more pollution, which means more deaths locally due to pollution.
Quite - it's protocols and very, very simple engineering which kills transmission of many diseases, not medicine. Not drinking fecal contaminated water is a _huge_ one. First estimate I found said 2.2 million deaths/year in 1998. Washing hands after defecating/urinating. Proper storage of food. Not using in-home unvented fires to cook. Not personally burying or touching the dead. The willingness/ability to go to hospital if ill. Social distances - if it is socially unacceptable to be bumping up against people.
All of these turn (for example) Ebola from in parts of Africa things that spread extremely rapidly to something that has limited spread, and is much more controllable even without modern medicine.
Add proper nutrition, shoes, and you're a large part of the way to a 'modern' lifestyle.
http://arstechnica.co.uk/scien... - twice in the last thousand years or so, there has been an event around ten to twenty times larger, with a _much_ more energetic and destructive (to orbital things) spectrum.
Especially fun is that you usually don't get to disagree with the judges findings. That is he can make a claim - and as long as it is arguably reasonably supported by the facts, and not so unreasonable that 'no reasonable person' would make the same decision (very, very far from 'was the decision reasonable), you can't challenge that at any appelate level.
Err - no. http://ark.intel.com/products/... - 18 core. " a 36 total core dual E5-2699v3 we see performance 29x to 51x that of llvmpipe" Two processors, with 18 cores each.
' On a 36 total core dual E5-2699v3 we see performance 29x to 51x that of llvmpipe. ' Clearly some improvement happens going from single to multithreaded, but I suspect very few desktops have >4 cores, and a vanishingly small number >16.
I was an otherwise healthy child until about age 12, when one month to the next I became extremely fatigued, with other symptoms. This is not depression - I can be doing something I enjoy for 15 minutes, and my thinking gets gradually muddier. I enjoy cooking. Often I cannot successfully make a bacon omelette due to fatigue and irrationality that comes on when doing a task as small as this.
Imagine a 15 minute task tires you as much as a 36 hour one. Not everyone is so affected, and there are likely many other related syndromes all lumped together with different etiologies.
Severe CFS/ME patients differed from controls and moderate CFS/ME patients over time and expressed significant alterations in iNKT cell phenotypes, CD8(+)T cell markers, NK cell receptors and T cells at 6 months. This highlights the importance of further assessing these potential immune biomarkers longitudinally in both moderate and severe CFS/ME patients.
I wish something rather easier to bear had happened at age 12, for example losing both legs.
Even the best testing panels have only found markers for CFS in a statistical manner. That is - you take a hundred people with CFS, a hundred people without CFS, and you can with certainty tell which group is which. However, you can't with any useful result test a single individual. The false positive rate is 45%, and the false negative rate is 45% or so.
CFS is not one disease, it is almost certainly many.
Err - no. Modern satellites have many ways of denying GPS to areas or turning off the 'civilian' mode totally. As well as high power 'spot-beams' to make jamming much, much harder.
'$10000/lb - in fact it costs a lot more'. This is misleading - the costs quoted are for an entire developed system and sending it to ISS where it will also dock and return cargo.
This is the difference between someone delivering 10 tons of sand to a dock on the other side of the country, and bagging it nicely in their own truck, and delivering it to your desk.
Not quite. This is not yet a full attack on SHA-1. It cannot - yet - be used to generate a collision for any known hash. It is an indication that you should move away from sha-1 as fast as you can.
Weeel. This whole argument became a hell of a lot less compelling now that even crappy SSDs will read random files at a couple of hundred meg a second.
The number of workloads where you actually need to reread files at over a couple of hundred meg as second, and have that working set be between 2 and 10 gigabytes or so, I suspect is going to equate to almost zero.
Do you donate to 'Dog Rescue Cyprus', or 'Penelopies Pantyhose' ? (I note the latter is a wecomic, not actual goods.) There are some very, very questionable and for some life-damaging things you could support. In addition, private messages asking for specific content to be created could be obviously damaging.
My first computer was a ZX81. This had a 16K RAM pack, with 250ns RAM. This could read a random byte of RAM in 470ns. My current system can access a byte of RAM in around 50ns.
That is a factor of ten improvement, when RAM size has increased a million fold. Caches ammeliorate this. But they do not eliminate it. #cachemissesmatter
Slashdot Mirror
in wood?
That's not enough, to a large degree. ... all today have CPUs of their own, usually with entirely secret firmware, and often access to the bus.
It must also be designed so that no peripheral outside of the CPU is trusted, if you're going that far.
Hard drives, network peripherals,
Exactly why should you (as an employee) have any rights to privacy on a computer you do not own, and agree to being monitored on?
If you're not doing a design where you cost the various design alternatives, you're doing it wrong, and may even not minimise the thing you're 'caring most' about.
It is always - for example - possible to improve safety of cars through expensive technical means. ...
More complex construction with better crash absorbtion properties,
But, if your vehicle is twice as safe as the rest of the fleet, and yet due to your safety upgrades costs eight times more than one that is only 1.1* as safe, you may actually end up costing lives with your 'safer' design.
Similarly - weight generally means poorer gas mileage, which means more pollution, which means more deaths locally due to pollution.
Quite - it's protocols and very, very simple engineering which kills transmission of many diseases, not medicine.
Not drinking fecal contaminated water is a _huge_ one.
First estimate I found said 2.2 million deaths/year in 1998.
Washing hands after defecating/urinating.
Proper storage of food.
Not using in-home unvented fires to cook.
Not personally burying or touching the dead.
The willingness/ability to go to hospital if ill.
Social distances - if it is socially unacceptable to be bumping up against people.
All of these turn (for example) Ebola from in parts of Africa things that spread extremely rapidly to something that has limited spread, and is much more controllable even without modern medicine.
Add proper nutrition, shoes, and you're a large part of the way to a 'modern' lifestyle.
It's metastasised.
I initially read it as $1B
'Free' may not quite mean free.
It means you can now advertise this in your resume, for example.
http://arstechnica.co.uk/scien... - twice in the last thousand years or so, there has been an event around ten to twenty times larger, with a _much_ more energetic and destructive (to orbital things) spectrum.
Especially fun is that you usually don't get to disagree with the judges findings.
That is he can make a claim - and as long as it is arguably reasonably supported by the facts, and not so unreasonable that 'no reasonable person' would make the same decision (very, very far from 'was the decision reasonable), you can't challenge that at any appelate level.
Err - no.
http://ark.intel.com/products/... - 18 core.
" a 36 total core dual E5-2699v3 we see performance 29x to 51x that of llvmpipe"
Two processors, with 18 cores each.
Hyperthreading does not meaningfully improve performance of compute-bound tasks.
' On a 36 total core dual E5-2699v3 we see performance 29x to 51x that of llvmpipe. '
Clearly some improvement happens going from single to multithreaded, but I suspect very few desktops have >4 cores, and a vanishingly small number >16.
Yes, it's not lyme. I've had all the tests regrettably.
There are issues around CFS and the DSM.
I was an otherwise healthy child until about age 12, when one month to the next I became extremely fatigued, with other symptoms.
This is not depression - I can be doing something I enjoy for 15 minutes, and my thinking gets gradually muddier. I enjoy cooking.
Often I cannot successfully make a bacon omelette due to fatigue and irrationality that comes on when doing a task as small as this.
Imagine a 15 minute task tires you as much as a 36 hour one.
Not everyone is so affected, and there are likely many other related syndromes all lumped together with different etiologies.
http://www.ncbi.nlm.nih.gov/pu... - for example.
CONCLUSIONS:
Severe CFS/ME patients differed from controls and moderate CFS/ME patients over time and expressed significant alterations in iNKT cell phenotypes, CD8(+)T cell markers, NK cell receptors and T cells at 6 months. This highlights the importance of further assessing these potential immune biomarkers longitudinally in both moderate and severe CFS/ME patients.
I wish something rather easier to bear had happened at age 12, for example losing both legs.
Even the best testing panels have only found markers for CFS in a statistical manner.
That is - you take a hundred people with CFS, a hundred people without CFS, and you can with certainty tell which group is which.
However, you can't with any useful result test a single individual.
The false positive rate is 45%, and the false negative rate is 45% or so.
CFS is not one disease, it is almost certainly many.
Err - no.
Modern satellites have many ways of denying GPS to areas or turning off the 'civilian' mode totally.
As well as high power 'spot-beams' to make jamming much, much harder.
VR twister!
'$10000/lb - in fact it costs a lot more'.
This is misleading - the costs quoted are for an entire developed system and sending it to ISS where it will also dock and return cargo.
This is the difference between someone delivering 10 tons of sand to a dock on the other side of the country, and bagging it nicely in their own truck, and delivering it to your desk.
And this is precisely why the advertising industry doesn't exist.
Not quite.
This is not yet a full attack on SHA-1.
It cannot - yet - be used to generate a collision for any known hash.
It is an indication that you should move away from sha-1 as fast as you can.
Weeel.
This whole argument became a hell of a lot less compelling now that even crappy SSDs will read random files at a couple of hundred meg a second.
The number of workloads where you actually need to reread files at over a couple of hundred meg as second, and have that working set be between 2 and 10 gigabytes or so, I suspect is going to equate to almost zero.
Do you donate to 'Dog Rescue Cyprus', or 'Penelopies Pantyhose' ?
(I note the latter is a wecomic, not actual goods.)
There are some very, very questionable and for some life-damaging things you could support.
In addition, private messages asking for specific content to be created could be obviously damaging.
My first computer was a ZX81.
This had a 16K RAM pack, with 250ns RAM.
This could read a random byte of RAM in 470ns.
My current system can access a byte of RAM in around 50ns.
That is a factor of ten improvement, when RAM size has increased a million fold.
Caches ammeliorate this.
But they do not eliminate it.
#cachemissesmatter
I recommend fbreader.