Looking at the bigger picture, deficit spending is still higher than pre-2008, and those 3-years you're comparing to were not typical due to things like TARP. Spending is continuing to balloon, yes the deficit is still projected to hover in the 0.8 trillion/year range. http://www.usgovernmentspending.com/download_multi_year_1960_2018USb_15s2li101mcn_G0f
Continuing to run up the credit card, paying only interest is a sure-fire way to bankrupt the household.
Cutting anything just to cut is stupid. DO you think just random cutting will get rid of pork? It will not, pork is always the last to go. What we need is targeted and precise cuts on a project basis.
But no, lets just leave all our allies dangling, and leave our defense wide open. Cause that's what happens when you just slash and burn.
Exactly, in fact projects like the SR-72 with no actual mission and the new F35 fighter that no-one wants should be at the top of that list of cuts. The F35 contract is past $60B now with no working aircraft so far. All the DOD departments wanted was a tech refresh of the existing fighters and not something brand new, but all the Congressional pork barreling got in the way of fiscal responsibility.
It's actually worse than that. The topic of study is on the impact of climate change on Nebraska, but the bill says they're only supposed to look at "cyclical" changes.
I think it's more like asking biologists to study the effects of antibiotic resistance, but they're not allowed to use evolution and must assume that the DNA of the bacteria doesn't significantly change over time.
I'd agree with that analogy. Except I see this as commissioning a study to focus on other possible causes of antibiotic resistance outside of DNA changes. Or are you assuming that evolution of DNA is the only possible means of acquiring resistance to antibiotics? That perhaps the host environment plays no role?
The politics here should not be discounted. The group that's refusing is a politically appointed commission, and they taking a political position before they've even read the study proposal. They've gone to the media claiming that the study proposal is rigged because it contains the single word "cyclical". I would think this is not exactly the unbiased group that should be doing this research.
Indeed, it should be entirely privately funded. Thus we can focus on the research that matters: only that which can be monetized within the next 4 quarters or sooner!
Define privately funded. By the oil industry? It's pretty hard to find finding from unbiased sources, much less sources who will fund you in the future if you produce results contrary to what they expects. Certainly politicians and the govt are hardly unbiased, as they want to fund studies that side with their campaign contributors.
A grant to investigate the sources of the observed cyclical changes in the global climate is not unreasonable given some of the conflicting evidence and studies. I don't see that it's in direct conflict with evidence suggesting that man-made pollution is a driver of climate change. The real crime is that both sides of the argument want an all-or-nothing answer, saying either man is entirely responsible or man has no influence. The real answer is that climate change is driven by many factors including elevated CO2 levels because we are pumping carbon of the the ground and putting it in the air (burning fossil fuels), cyclical changes in solar output, etc.
"But it also raises the question of whether targeting anonymity services to hunt out fraudsters could have chilling effects for harmless Tor users trying to protect their privacy online"
Umm.. the user is ordering something using their name, credit card, and address. They are not going to use Tor to protect their anonymity.
But you certainly have a crowd that likes the idea of tor and has their browser always configured to use it. I don't think that raising the risk level associated with a transaction based on the client using tor is unreasonable. If this were a brick and mortar store, they'd probably be a little bit wary of doing a credit card sale to someone wearing a disguise that covered their face.
Also realize that this would only be one of many sanity checks employed. Is the shipping address to the address listed on the cc for example. The credit card company also checks where the card was used, for things like buying gas at 1pm and then buying it again at 2pm 100 miles away. They also consider the type of merchandise as online purchase of electronics is rife with fraud, but very few people use a stolen card to buy socks.
The legal briefs filed so far look like they are about to hand the government its own ass in respect to seizing SSL keys.
How so? The filing is only asking that the contempt charges be reversed, and the govt to turn over any SSL keys in their possession (like you trust them not to keep a copy?) They might get the contempt charge reversed, but I wouldn't expect any other sort of compensation. Setting a precedence that the original demand for the encryption keys was unlawful is about all they can hope for.
So basically, they provide a VPN-to-Internet service, but do not record the data required for a pen register order. Based on the EFF information on pen registers, that would constitute connection data (i.e. incoming and outgoing connections - IP addressed, ports, etc, but not content of the connections).
The NSA already sees the data when it hits the internet, and the VPN "pen data" only provides some anonymity by making the NSA have to work a little to attribute the data to specific customers. Seeing how they managed to make attributions with bit-torrent, I doubt they have to work very hard.
I agree this was more of a PR stunt than out of real concerns.
I still see a number of issues that need resolved. A human driver is capable of reacting to a much wider range of problems and required responses, although I grant that human drivers have slower response times and sometimes react badly to surprises. What will a driverless car do if the road is blocked? Will is sit there confused, will it attempt a 5-pt turn in the middle of the road, etc. Will it drive around the construction guy holding up a stop sign? Does it know that it's supposed to yield to people waiting at a cross-walk, etc. All these are situations that would probably involve the human having to take over.
Laws and liability issues are a major hurdle, and it's likely to be a messy one as each state has it's own laws and opinions.
These cars are also limited by the quality of road map data. As an example, all the major map companies still don't show the circle by my house as one-way, so we're constantly getting large delivery trucks going the wrong way around a 1-way circle. A simple error on a map could potentially have cars driving the wrong way down a freeway. All the google tests so far have been on very groomed map data. I don't believe any of the tests have been through areas with spotty gps coverage like long tunnels either. So again, I think the totally driverless part is quite a ways off.
Someone who users MS Word more than anyone reading this site has spotted some problems and you are bitching that he is "non-technical"? I suppose if he used the program less and did some technical stuff instead you would dismiss him as not using the program enough to comment on it? We shouldn't blame the users for usability problems. That's where we have to lift our game instead of throwing insults back.
Not bitching, just stating the obvious. He's trying to argue that MS Word sucks on technical merits, but doesn't seem to have a good grasp of those technical details. Did you bother to read his article? The bulk of the article was a history lesson of MS Word file format changes. The wikipedia articles he seems to be paraphrasing are mostly correct about how Word used to store the files, but not real accurate for the current format. He only spends a single paragraph to state the features he doesn't like in very general terms, and then says MS Word must die because he can't avoid using it.
The sensationalistic title of "Why Microsoft Word must Die" is a bit over the top, and smacks of a cheap tabloid rag. Not something worthy of someone with a few Hugos for novels.
I don't think that the liability will be a barrier for the implementation of autonomous cars.
Seat belts are not a good comparison as seatbelts can't cause multi-vehicle accidents.
Liability absolutely will be a factor in our overly litigious society . The first time there is an accident with a fatality, the lawsuit against the manufacturer will be enormous. I don't think Google isn't interested in actually building the cars, but more into developing the tech and then licensing it to the automakers. Essentially what they do with Android. That gets them out of the loop for liability as they can blame the automaker for their implementation if something goes wrong. The automakers will of course attempt to shift any blame to the operator.
We are still a long, long way away from a totally driver-less car. Call me back when the car can navigate from my house to my work and park without some sort of human interaction. Watch out for detours, deer, kids in the road, etc.
I just tested this in Word 2007. Yes, I'd consider it a bug that if the table is the first object, you can't select in front of it in either draft or print view. The work around is pretty simply though. Add a return before you add the table as a place holder. Or to fix afterwards - select the table, cut the table, type your text and repaste the table.
If the table is the only thing on the page, then Word assumes if you're dragging it then what you really wanted is to adjust the margins. Otherwise dragging makes no sense as there is nothing to flow around the table as it's moving.
Also non-technical. He's a writer who dislikes Word and managed to find a few technical sound-bits to support his argument, but doesn't seem to understand them. For example style sheets require the use of control codes. It's just that the control codes can specify the change directly, or they can refer to the style sheet.
Both Word and HTML suck for physical publishing purposes, as neither truly describes a page layout. Anyone who's been frustrated by paragraphs suddenly flowing over a page break, or tried to view html on different browsers has experienced that problem. Sending the publisher a Word doc gives you no guarantee that why you get back looks like what you had on your screen. PDF and PS at least nail down the exact page layout.
There are great uses for flexible standards like ebook format, because they aren't restricted to a particular page layout. Ebooks can flow to the readers screen size and whatever font size they picked. PDFs suck for digital books because they don't reflow.
After 30 years of automakers blatantly providing theoretical and incredibly optimistic EPA estimates for gas mileage, you'd think that dealers would be willing to give a little on another car maker fudging some other numbers on their site.
The blame for that lies with the EPA, not the automakers. You should also blame the EPA for letting the automakers make totally irrelevant mpg claims for plug-in hybrids.
You are just repeating the fallacy. Having the code open to the world does little to make that code more secure, particularly when its something as complex as a kernel. The examples of a white hat noticing a vulnerability from reviewing the code and reporting it are very few compared with the number of vulnerabilities found by a black hat and turned into a working exploit. When a white hat reports a vulnerability, it usually found by observing the behavior of the code and then digging through the code, not the other way around.
There are many people that work on the kernel, and even more students that study it. The kernel is of little concern. What is a concern is the thousands and thousands of little executables that are in so many distros. Worse still, how many people look through all the code from an average everyday apt-get?
Doesn't really matter in the end as there is always the Underhanded C Contest to think about.
Linux had 14 kernel vulnerabilities this year versus 7 Windows kernel-mode vulnerabilities this year. (Just going by MS announcements for Windows 7, there may have been more unannounced issues)
Like everyone else on slashdot, I only run Debian and must say I smile when I see reports such as country sponsored malware strikes like this. But it does make me ask an honest question:
How can we be sure that the Linux kernel isn't compromised? I don't really have the time to go through all lines of code and I doubt my security analysis and development skills are up to the task anyway.
But, but, aren't there "many eyes" reviewing the code, making it perfectly safe? At least you recognize the fallacy of open source software being more secure. In reality there really isn't an entire community reviewing and proofing the code. Just a handful of hackers pouring through it looking for exploits. It's less likely that it was intentionally compromised, but in some respects the linux kernel and distros are more vulnerable because their code is published.
Being a decent coder has little to do with CS. It's a very valuable skill in its own right, but quite different.
Very true. Unfortunately, many employers haven't a clue what the difference is. I see too many jobs ads looking for a CS degree when what they want is a good programmer. They end up with a CS major who hasn't a clue how to design or write good code. Or vice-versa, they get a programmer to do software engineering and wonder why they end up with a crap program that doesn't meet their needs.
They're not suffering yet if the persistent rumors about what is about to get released are anywhere near the reality. Apparently the next batch will tell the story of industrial espionage as conducted by NSA and then provided as a service to US companies. If that gets confirmed, you can start seeing actual government sanctions from large trading partners like EU, China, Russia and so on. And it's then when the meaning of "suffering" for US companies will start to really be realized in this particular case.
Right now it's just cloud companies that start getting squeezed. That's nothing major when you look at US-dominated IT industry as a whole.
Still nothing new, as other countries do this. Even countries we consider to be our military allies, can still be our economic adversaries and spy on us to gain economic or industrial advantages. Embarrassing to the US on the international stage maybe, but you won't see any real sanctions beyond scolding.
Nobody voted for Ron Paul because he wasn't expected to have a real chance at getting elected. Instead of voting for a losing proposition, most people voted against the republican or democrat that they didn't want to win.
The reason he gave was one step ahead of that: even if you get apps that can use all eight cores, it's going to be murder on battery life, and most of the cost will be wasted (because most apps won't be using all that core power still.)
Qualcom basically called them out for building an 8-core as a marketing ploy ("mine goes to eleven!"). Consumers have been focusing on the number of cores without understanding that more cores doesn't translate to better (e.g. all the Chinese crap tablets with slow dual-cores). Right or wrong about the benefit of an 8-core SOC, this is a marketing war and Qualcom just fired a viciously public volley.
The problem is that workloads people regularly do simply don't use 8 cores.
The problem is people who don't do much on their computers who then claim that more computing power isn't necessary because nobody does much on their computers.
I work in an environment where modelers are using quad-chip hex-core systems and could easily use more in a heartbeat. Load averages of greater than 100 on a regular basis. Small input, medium output, and lotsa lotsa CPU time.
Where did you get the idea that everyone is expected to buy an 8 core system and so 8 core systems aren't justified because some people don't need them?
Okay, so you'd pointed out that there is a high-end workstation market that benefits from multiple cores. How does this translate into needing them in a phone where multiple threads or parallel processing are not the normal? The other consideration is that multiple cores can be less efficient that fewer cores that total the same flops rating. In that respect Qualcom is absolutely right that battery life is a very important consideration.
The story talks about a *wired* port by the parking brake. That would mean the attacker was in the car, or a remote device was attached, which investigators would (or at least could) find. It also only addresses a specific Ford vehicle, which has no relationship to a Mercedes.
Yes, they basically plugged into the can-bus and started spewing bogus data and commands onto the network. Easiest access point would be the ODB-II connector which is required to be under the dash within a few feet of the steering wheel. So yeah, as you describe this isn't some sort of remote drive-by hack that can be inflicted onto any car on the highway.
One amusing thing is that most radios listen to the can-bus for things like vehicle speed now. Wonder if it's possible to create a virus to compromise the radio via flash-drive and have it randomly spew similar potentially dangerous data onto the bus?
Slashdot Mirror
Looking at the bigger picture, deficit spending is still higher than pre-2008, and those 3-years you're comparing to were not typical due to things like TARP. Spending is continuing to balloon, yes the deficit is still projected to hover in the 0.8 trillion/year range.
http://www.usgovernmentspending.com/download_multi_year_1960_2018USb_15s2li101mcn_G0f
Continuing to run up the credit card, paying only interest is a sure-fire way to bankrupt the household.
Cutting anything just to cut is stupid. DO you think just random cutting will get rid of pork? It will not, pork is always the last to go.
What we need is targeted and precise cuts on a project basis.
But no, lets just leave all our allies dangling, and leave our defense wide open. Cause that's what happens when you just slash and burn.
Exactly, in fact projects like the SR-72 with no actual mission and the new F35 fighter that no-one wants should be at the top of that list of cuts. The F35 contract is past $60B now with no working aircraft so far. All the DOD departments wanted was a tech refresh of the existing fighters and not something brand new, but all the Congressional pork barreling got in the way of fiscal responsibility.
It's actually worse than that. The topic of study is on the impact of climate change on Nebraska, but the bill says they're only supposed to look at "cyclical" changes.
I think it's more like asking biologists to study the effects of antibiotic resistance, but they're not allowed to use evolution and must assume that the DNA of the bacteria doesn't significantly change over time.
I'd agree with that analogy. Except I see this as commissioning a study to focus on other possible causes of antibiotic resistance outside of DNA changes. Or are you assuming that evolution of DNA is the only possible means of acquiring resistance to antibiotics? That perhaps the host environment plays no role?
The politics here should not be discounted. The group that's refusing is a politically appointed commission, and they taking a political position before they've even read the study proposal. They've gone to the media claiming that the study proposal is rigged because it contains the single word "cyclical". I would think this is not exactly the unbiased group that should be doing this research.
Indeed, it should be entirely privately funded. Thus we can focus on the research that matters: only that which can be monetized within the next 4 quarters or sooner!
Define privately funded. By the oil industry? It's pretty hard to find finding from unbiased sources, much less sources who will fund you in the future if you produce results contrary to what they expects. Certainly politicians and the govt are hardly unbiased, as they want to fund studies that side with their campaign contributors.
A grant to investigate the sources of the observed cyclical changes in the global climate is not unreasonable given some of the conflicting evidence and studies. I don't see that it's in direct conflict with evidence suggesting that man-made pollution is a driver of climate change. The real crime is that both sides of the argument want an all-or-nothing answer, saying either man is entirely responsible or man has no influence. The real answer is that climate change is driven by many factors including elevated CO2 levels because we are pumping carbon of the the ground and putting it in the air (burning fossil fuels), cyclical changes in solar output, etc.
"But it also raises the question of whether targeting anonymity services to hunt out fraudsters could have chilling effects for harmless Tor users trying to protect their privacy online"
Umm.. the user is ordering something using their name, credit card, and address. They are not going to use Tor to protect their anonymity.
But you certainly have a crowd that likes the idea of tor and has their browser always configured to use it. I don't think that raising the risk level associated with a transaction based on the client using tor is unreasonable. If this were a brick and mortar store, they'd probably be a little bit wary of doing a credit card sale to someone wearing a disguise that covered their face.
Also realize that this would only be one of many sanity checks employed. Is the shipping address to the address listed on the cc for example. The credit card company also checks where the card was used, for things like buying gas at 1pm and then buying it again at 2pm 100 miles away. They also consider the type of merchandise as online purchase of electronics is rife with fraud, but very few people use a stolen card to buy socks.
Donate to Lavabit legal fund
The legal briefs filed so far look like they are about to hand the government its own ass in respect to seizing SSL keys.
How so? The filing is only asking that the contempt charges be reversed, and the govt to turn over any SSL keys in their possession (like you trust them not to keep a copy?) They might get the contempt charge reversed, but I wouldn't expect any other sort of compensation. Setting a precedence that the original demand for the encryption keys was unlawful is about all they can hope for.
So basically, they provide a VPN-to-Internet service, but do not record the data required for a pen register order. Based on the EFF information on pen registers, that would constitute connection data (i.e. incoming and outgoing connections - IP addressed, ports, etc, but not content of the connections).
The NSA already sees the data when it hits the internet, and the VPN "pen data" only provides some anonymity by making the NSA have to work a little to attribute the data to specific customers. Seeing how they managed to make attributions with bit-torrent, I doubt they have to work very hard.
I agree this was more of a PR stunt than out of real concerns.
Not posting it publicly since it's literally next to my house. Plenty of examples here though http://wiki.openstreetmap.org/wiki/Catalog_of_Errors.
I still see a number of issues that need resolved. A human driver is capable of reacting to a much wider range of problems and required responses, although I grant that human drivers have slower response times and sometimes react badly to surprises. What will a driverless car do if the road is blocked? Will is sit there confused, will it attempt a 5-pt turn in the middle of the road, etc. Will it drive around the construction guy holding up a stop sign? Does it know that it's supposed to yield to people waiting at a cross-walk, etc. All these are situations that would probably involve the human having to take over.
Laws and liability issues are a major hurdle, and it's likely to be a messy one as each state has it's own laws and opinions.
These cars are also limited by the quality of road map data. As an example, all the major map companies still don't show the circle by my house as one-way, so we're constantly getting large delivery trucks going the wrong way around a 1-way circle. A simple error on a map could potentially have cars driving the wrong way down a freeway. All the google tests so far have been on very groomed map data. I don't believe any of the tests have been through areas with spotty gps coverage like long tunnels either. So again, I think the totally driverless part is quite a ways off.
Someone who users MS Word more than anyone reading this site has spotted some problems and you are bitching that he is "non-technical"? I suppose if he used the program less and did some technical stuff instead you would dismiss him as not using the program enough to comment on it?
We shouldn't blame the users for usability problems. That's where we have to lift our game instead of throwing insults back.
Not bitching, just stating the obvious. He's trying to argue that MS Word sucks on technical merits, but doesn't seem to have a good grasp of those technical details. Did you bother to read his article? The bulk of the article was a history lesson of MS Word file format changes. The wikipedia articles he seems to be paraphrasing are mostly correct about how Word used to store the files, but not real accurate for the current format. He only spends a single paragraph to state the features he doesn't like in very general terms, and then says MS Word must die because he can't avoid using it.
The sensationalistic title of "Why Microsoft Word must Die" is a bit over the top, and smacks of a cheap tabloid rag. Not something worthy of someone with a few Hugos for novels.
I don't think that the liability will be a barrier for the implementation of autonomous cars.
Seat belts are not a good comparison as seatbelts can't cause multi-vehicle accidents.
Liability absolutely will be a factor in our overly litigious society . The first time there is an accident with a fatality, the lawsuit against the manufacturer will be enormous. I don't think Google isn't interested in actually building the cars, but more into developing the tech and then licensing it to the automakers. Essentially what they do with Android. That gets them out of the loop for liability as they can blame the automaker for their implementation if something goes wrong. The automakers will of course attempt to shift any blame to the operator.
We are still a long, long way away from a totally driver-less car. Call me back when the car can navigate from my house to my work and park without some sort of human interaction. Watch out for detours, deer, kids in the road, etc.
I just tested this in Word 2007. Yes, I'd consider it a bug that if the table is the first object, you can't select in front of it in either draft or print view. The work around is pretty simply though. Add a return before you add the table as a place holder. Or to fix afterwards - select the table, cut the table, type your text and repaste the table.
If the table is the only thing on the page, then Word assumes if you're dragging it then what you really wanted is to adjust the margins. Otherwise dragging makes no sense as there is nothing to flow around the table as it's moving.
This critic just comes across as whiny to me.
Also non-technical. He's a writer who dislikes Word and managed to find a few technical sound-bits to support his argument, but doesn't seem to understand them. For example style sheets require the use of control codes. It's just that the control codes can specify the change directly, or they can refer to the style sheet.
Both Word and HTML suck for physical publishing purposes, as neither truly describes a page layout. Anyone who's been frustrated by paragraphs suddenly flowing over a page break, or tried to view html on different browsers has experienced that problem. Sending the publisher a Word doc gives you no guarantee that why you get back looks like what you had on your screen. PDF and PS at least nail down the exact page layout.
There are great uses for flexible standards like ebook format, because they aren't restricted to a particular page layout. Ebooks can flow to the readers screen size and whatever font size they picked. PDFs suck for digital books because they don't reflow.
After 30 years of automakers blatantly providing theoretical and incredibly optimistic EPA estimates for gas mileage, you'd think that dealers would be willing to give a little on another car maker fudging some other numbers on their site.
The blame for that lies with the EPA, not the automakers. You should also blame the EPA for letting the automakers make totally irrelevant mpg claims for plug-in hybrids.
You are just repeating the fallacy. Having the code open to the world does little to make that code more secure, particularly when its something as complex as a kernel. The examples of a white hat noticing a vulnerability from reviewing the code and reporting it are very few compared with the number of vulnerabilities found by a black hat and turned into a working exploit. When a white hat reports a vulnerability, it usually found by observing the behavior of the code and then digging through the code, not the other way around.
There are many people that work on the kernel, and even more students that study it. The kernel is of little concern. What is a concern is the thousands and thousands of little executables that are in so many distros. Worse still, how many people look through all the code from an average everyday apt-get?
Doesn't really matter in the end as there is always the Underhanded C Contest to think about.
So please explain the number of kernel exploits over the past year.
http://www.cvedetails.com/vulnerability-list/vendor_id-33/product_id-47/cvssscoremin-7/cvssscoremax-7.99/Linux-Linux-Kernel.html
http://www.zdnet.com/linux-trailed-windows-in-patching-zero-days-in-2012-report-says-7000011326/
Linux had 14 kernel vulnerabilities this year versus 7 Windows kernel-mode vulnerabilities this year. (Just going by MS announcements for Windows 7, there may have been more unannounced issues)
Like everyone else on slashdot, I only run Debian and must say I smile when I see reports such as country sponsored malware strikes like this. But it does make me ask an honest question:
How can we be sure that the Linux kernel isn't compromised? I don't really have the time to go through all lines of code and I doubt my security analysis and development skills are up to the task anyway.
But, but, aren't there "many eyes" reviewing the code, making it perfectly safe? At least you recognize the fallacy of open source software being more secure. In reality there really isn't an entire community reviewing and proofing the code. Just a handful of hackers pouring through it looking for exploits. It's less likely that it was intentionally compromised, but in some respects the linux kernel and distros are more vulnerable because their code is published.
Norton 360 that is completely worthless against their root kit?
For all we know, Norton 360 might *be* their root kit.
So you use Kaspersky instead? You realize that it's banned by DOD because of "supply chain concerns".
Being a decent coder has little to do with CS. It's a very valuable skill in its own right, but quite different.
Very true. Unfortunately, many employers haven't a clue what the difference is. I see too many jobs ads looking for a CS degree when what they want is a good programmer. They end up with a CS major who hasn't a clue how to design or write good code. Or vice-versa, they get a programmer to do software engineering and wonder why they end up with a crap program that doesn't meet their needs.
They're not suffering yet if the persistent rumors about what is about to get released are anywhere near the reality. Apparently the next batch will tell the story of industrial espionage as conducted by NSA and then provided as a service to US companies. If that gets confirmed, you can start seeing actual government sanctions from large trading partners like EU, China, Russia and so on. And it's then when the meaning of "suffering" for US companies will start to really be realized in this particular case.
Right now it's just cloud companies that start getting squeezed. That's nothing major when you look at US-dominated IT industry as a whole.
Still nothing new, as other countries do this. Even countries we consider to be our military allies, can still be our economic adversaries and spy on us to gain economic or industrial advantages. Embarrassing to the US on the international stage maybe, but you won't see any real sanctions beyond scolding.
Nobody voted for Ron Paul because he wasn't expected to have a real chance at getting elected. Instead of voting for a losing proposition, most people voted against the republican or democrat that they didn't want to win.
The reason he gave was one step ahead of that: even if you get apps that can use all eight cores, it's going to be murder on battery life, and most of the cost will be wasted (because most apps won't be using all that core power still.)
Qualcom basically called them out for building an 8-core as a marketing ploy ("mine goes to eleven!"). Consumers have been focusing on the number of cores without understanding that more cores doesn't translate to better (e.g. all the Chinese crap tablets with slow dual-cores). Right or wrong about the benefit of an 8-core SOC, this is a marketing war and Qualcom just fired a viciously public volley.
The problem is that workloads people regularly do simply don't use 8 cores.
The problem is people who don't do much on their computers who then claim that more computing power isn't necessary because nobody does much on their computers.
I work in an environment where modelers are using quad-chip hex-core systems and could easily use more in a heartbeat. Load averages of greater than 100 on a regular basis. Small input, medium output, and lotsa lotsa CPU time.
Where did you get the idea that everyone is expected to buy an 8 core system and so 8 core systems aren't justified because some people don't need them?
Okay, so you'd pointed out that there is a high-end workstation market that benefits from multiple cores. How does this translate into needing them in a phone where multiple threads or parallel processing are not the normal? The other consideration is that multiple cores can be less efficient that fewer cores that total the same flops rating. In that respect Qualcom is absolutely right that battery life is a very important consideration.
The story talks about a *wired* port by the parking brake. That would mean the attacker was in the car, or a remote device was attached, which investigators would (or at least could) find. It also only addresses a specific Ford vehicle, which has no relationship to a Mercedes.
Yes, they basically plugged into the can-bus and started spewing bogus data and commands onto the network. Easiest access point would be the ODB-II connector which is required to be under the dash within a few feet of the steering wheel. So yeah, as you describe this isn't some sort of remote drive-by hack that can be inflicted onto any car on the highway.
One amusing thing is that most radios listen to the can-bus for things like vehicle speed now. Wonder if it's possible to create a virus to compromise the radio via flash-drive and have it randomly spew similar potentially dangerous data onto the bus?