25% acknowledged that their organization has been breached. What's worse, 42% did not know and could not exclude a breach
No, that's not worse. That's _better_. Those 42% are being realistic. Realistically, unless you're one of a tiny percentage of people who either (a) receives so little traffic they can audit it all or (b) can be 100% certain of the security of all the software they're running, you should be in one of those two categories: breached, or don't know whether you've been breached but can't exclude it.
What's _actually_ worrying is that 33% of respondents think they are in one of these two categories, when in actual fact I'd suspect the figure is less than 1%.
(FTR: my company is in the 'breached' category. We had a worm infect one of our servers via a BIND bug back in 2000 or so, although the infection was apparently unsuccessful... it seemed to rely on there being a line feed on the end of the last line of/etc/inetd.conf, and our file didn't have one. I can't, obviously, rule out any breaches since then, but am reasonably confident there haven't been any.)
If you are in a jurisdiction where it isn't legal to defend yourself then the fact that you were defending yourself is irrelevant.
Not necessarily true. The law doesn't treat all acts with the same outcome as being indistinguishable.
Here in the UK, there is a somewhat limited scope for self-defence as a defence from a murder charge. It wouldn't work in the case where my response was disproportionate to the threat. For instance, if an unarmed man attempts to mug me and I pull a gun and shoot him, even though I can reasonably say I feared for my life, I would probably still be convicted of murder.
Consider as a contrast, though, a case where I'm walking down the street, see somebody I don't like, pull a gun and shoot them.
In the latter case, I could expect to spend 20-30 years in prison for my offence. In the former case, I'd be unlikely to be inside for 10.
Are you 14? I ask only out of curiosity. If you are, carry on. If you're older, for shame sir
I'm not the original AC, but I share his opinion. I'm 31. The point is, anyone who wants a pirated copy of almost any popular game just has to haul their ass over to the pirate bay, look for a popular torrent of it and download. The DRM will have been stripped and everything will work just perfectly.
People actually paying for games will be presented with games that might not work with their hardware (I've seen games that only work with CD drives on a specific list, and if your drive isn't one of those you're SOL... time for a downgrade), or might not run if they have certain perfectly legitimate software installed (I've heard of games that refuse to run if you have debuggers, virtual CD drivers or even fucking Process Explorer installed). Even if they get the game to run, there's a significant change it'll stop working if they upgrade their hardware too many times.
Yes, DRM is an encumbrance on the paying customers but not on the pirates. This seems perfectly clear to me. So when you see people complaining about DRM, generally speaking they have paid.
I have security software which can prevent the DRM from doing harmful things to my computer.
I don't tend to buy modern games (my computer just about runs Doom3, I don't think it could manage anything much more recent than that), but my understanding was that the main issue is the DRM prevents the game from running if you have security software.
I've heard stories about games that scan your computer for stuff they think might be used to crack them (debuggers, CD mastering tools, etc.) and refuse to run while they're installed.
Yes, but what about (a) cost per mile and (b) energy used per mile. Who cares how much volume your fuel takes? Remember that diesel is both more and expensive and higher energy density than petrol.
I'm more worried about the Lithium batteries recycling themselves explosively while I'm driving the vehicle!
Probably not worth worrying about. (1) Lithium batteries only tend to explode under charge. (2) The batteries used in cars are generally different chemistries to the ones used in laptops/phones etc, and are more stable.
The possibility of a toxic electrolyte spill is much more worrying, IMO.
Interesting article on businessweek.com. Ford is selling 65mpg cars in Europe, but not the US.
Well, sure. I don't see a 65mpg diesel as a particular achievement, though. Not in a small car. I mean, I get 40+ out of my Citroen XM, which is (a) over 10 years old and (b) probably weighs about 3 times what a Fiesta does.
Don't forget, diesel has a higher energy density than petrol, so you can't compare them gallon-for-gallon.
I think the Chevy Volt will have a Lithium Ion battery, the kind that has caused problems in laptops. I've heard they are working on reliability and safety issues with the batteries.
They're currently considering two different techs, Lithium-Manganese Oxide and Lithium-Iron Phosphate (aka LiFePO4). Neither of these are the same chemistry as the Li-ions used in laptops, and don't have the problem of producing excess gases during charging, so are (without any further work necessary) substantially safer.
I think it's more like replacing your car's fuel tank.
In terms of expense, it's definitely more like replacing the engine. Think ~$10,000 if you ever need to do it.
Of course, it's hard to see why you ever would. A hybrid is not like a battery electric; it uses the same sort of battery technology that's specced to last (typically) 10,000 deep charge-discharge cycles. But you're unlikely to ever see more than 50% discharge on a hybrid's battery pack (probably not even that), so you'll get a lot more cycles. 20,000 or more, I would guess. That equates to more than 200,000 miles with anything like a normal usage pattern. If you're doing long journeys it's probably more like 400,000 miles.
I think 65-100A delivering 21.5kW-33kW at least would be required to push the car around at a decent speed
Probably not, no. At 21.5kW a good electric motor typically delivers something like 100Nm (70 lbs.ft) at 4000rpm (or nearly-proportionally higher torque at lower speeds), which is more than enough to accelerate a reasonable weight vehicle quite quickly.
I agree that 6.5A@330V wouldn't be enough, but I don't think as much as 65 would be necessary. Looking at the specs for the Prius, its motor produces 450Nm at 400RPM, which probably equates to approximately 10kW. The _motor_ (not the battery pack) is described as max 6.5A, which suggests there's a voltage step-up somewhere in the control system.
3ms is still quite a bit slower than the NYSE, who (I think) claim 1ms for their linux-based system. But I imagine there are other factors here, like physical distance, the precise definition of 'transaction' and whether that's the guaranteed or typical speed.
I don't know about NYSE, but this system was supposed to have redundant implementations at two different locations with a 1-second failover between them. Obviously, in the event of a failure, that didn't work, but I presume they did at least implement it. This would likely have a substantial effect on transaction latency. I dunno if NYSE has a similar system or not?
I see no difference in Oracle performance on Windows vs Linux. Performance is usually more about architecture than the platform. You can make highly available networks with Windows, people do it all the time, much more often than a lot of Slashdotters seem to think.
Sure you can. The problem is that when you see a high profile Windows deployment like this (and, in fact, a lot of smaller deployments as well), you'll frequently find that MS has paid the consultants and told them what technology to use. So you often end up with a compromised architecture that's designed to show off a fancy new feature of the latest MS server software, rather than a properly engineered architecture that's as simple as possible (but no simpler), which is the way architecture should be done.
An example I know about in more detail: my company was hired about 6 or 7 years back to fix a catastrophic web site fuckup... the existing consultants hadn't finished the site in about a year and didn't seem to be capable of doing so, so the business owner came to us with 4 weeks before it was due to launch. It was a fairly simple web system, nothing fancy, a reverse auction system that allowed suppliers to bid using a back-end system on requests for supply that customers entered in the front end. The problem? Microsoft had paid the consultants to implement the back end with ASP and Access (!) while Macromedia had paid them to implement the front end entirely in flash version 4 (that's before flash supported actionscript). Every minor change to the user interface (e.g., "no I don't want to show 10 results on that list, let's go for 15") required hours of work including negotiations backwards and forwards between the ASP programmers and the Flash designers concerning the data structures (flat by necessity, so filled with items like 'result13title' and 'result13description' and stuff like that) that were passed between the client and the server.
Needless to say we ripped it all out, reimplemented in HTML & Javascript, and they launched with a much better site. But this is a cautionary tale: if you aren't careful, you can end up with a solution paid for by somebody who's trying to sell something, rather than one that's actually designed to fit your needs. And that solution may do something particularly well that you don't need, while it utterly fails to supply something you do need. In this case, it seems reliability (I would guess LSE's top requirement) has been sacrificed for performance (probably on the list, but I doubt it was top of it).
And there have been, what, 6 or so cases worldwide of death from Mad Cow? The threat is of the same order of magnitude.
Since 1990 there have been 1334 in the UK alone. Baseline figures for the US (in absence of any contaminated food sources, etc.) are 1 per 9,000 adults aged over 55, which based on a back-of-envelope calculation means somewhere in the region of 200 per annum.
More generally, unless you're a worker in the field where a risk occurs, a good rule of thumb is that it's a waste of time to worry about any danger that you hear about on the news. If it happens rarely enough to be newsworthy, it's not a danger to you
No different then what can happen on a unix box I suppose.
Note that the current system is built around a large cluster of 2.2GHz servers, while the unix-based system it replaced (which coped perfectly happily with a substantial portion of the same traffic) ran from a smaller cluster of much slower servers.
The primary purpose for the new system, introduced less than a year ago, was to expand capacity. For it to have failed within a year due to lack of capacity basically means that it has failed in that objective.
So prions can make proteins into prions in other species, e.a. cross the species barrier? Big deal.
The summary's wrong. Yes, we've known this for ages. What's new is we have a nice, fast, reliable way of studying without actually infecting animals, which could help us answer the important questions, like 'can this particular prion infect humans?'
So the public is donating a lot of computing time and electrical energy. What does the public get back?
If Folding@Home goes towards lining the pockets of a university endowment or a drug company's coffers, count me out. If the research product is required to be free from patents, and available for public good...full speed ahead. Somehow, I seriously doubt that any successful results will be freely available.
The project is run by a non-profit company. Results are routinely made public. I don't see any reason to be concerned here.
Uhuh. The summary is just wrong, is what's going on here. If you read the article, what it's actually about is a new way to study prions in vitro and test them to determine whether there's a possibility of them crossing a particular species barrier.
Which is, when you think about it, a very useful test to have available.
The good news is that prion-based diseases are vanishingly rare. Avoid cannibalism and you're safe: so far meteor strikes are about as dangerous a threat, along with heart-attack induced by the stress of winning the lottery.
Don't overestimate the threat of meteor strikes. While there have been a lot of near misses and property damage, there are apparently no recorded instances of somebody being killed by a meteorite since 1929.
Slashdot Mirror
Is.
"Data" is used in this context as a collective noun.
25% acknowledged that their organization has been breached. What's worse, 42% did not know and could not exclude a breach
No, that's not worse. That's _better_. Those 42% are being realistic. Realistically, unless you're one of a tiny percentage of people who either (a) receives so little traffic they can audit it all or (b) can be 100% certain of the security of all the software they're running, you should be in one of those two categories: breached, or don't know whether you've been breached but can't exclude it.
What's _actually_ worrying is that 33% of respondents think they are in one of these two categories, when in actual fact I'd suspect the figure is less than 1%.
(FTR: my company is in the 'breached' category. We had a worm infect one of our servers via a BIND bug back in 2000 or so, although the infection was apparently unsuccessful... it seemed to rely on there being a line feed on the end of the last line of /etc/inetd.conf, and our file didn't have one. I can't, obviously, rule out any breaches since then, but am reasonably confident there haven't been any.)
If you are in a jurisdiction where it isn't legal to defend yourself then the fact that you were defending yourself is irrelevant.
Not necessarily true. The law doesn't treat all acts with the same outcome as being indistinguishable.
Here in the UK, there is a somewhat limited scope for self-defence as a defence from a murder charge. It wouldn't work in the case where my response was disproportionate to the threat. For instance, if an unarmed man attempts to mug me and I pull a gun and shoot him, even though I can reasonably say I feared for my life, I would probably still be convicted of murder.
Consider as a contrast, though, a case where I'm walking down the street, see somebody I don't like, pull a gun and shoot them.
In the latter case, I could expect to spend 20-30 years in prison for my offence. In the former case, I'd be unlikely to be inside for 10.
Are you 14? I ask only out of curiosity. If you are, carry on. If you're older, for shame sir
I'm not the original AC, but I share his opinion. I'm 31. The point is, anyone who wants a pirated copy of almost any popular game just has to haul their ass over to the pirate bay, look for a popular torrent of it and download. The DRM will have been stripped and everything will work just perfectly.
People actually paying for games will be presented with games that might not work with their hardware (I've seen games that only work with CD drives on a specific list, and if your drive isn't one of those you're SOL... time for a downgrade), or might not run if they have certain perfectly legitimate software installed (I've heard of games that refuse to run if you have debuggers, virtual CD drivers or even fucking Process Explorer installed). Even if they get the game to run, there's a significant change it'll stop working if they upgrade their hardware too many times.
Yes, DRM is an encumbrance on the paying customers but not on the pirates. This seems perfectly clear to me. So when you see people complaining about DRM, generally speaking they have paid.
I have security software which can prevent the DRM from doing harmful things to my computer.
I don't tend to buy modern games (my computer just about runs Doom3, I don't think it could manage anything much more recent than that), but my understanding was that the main issue is the DRM prevents the game from running if you have security software.
I've heard stories about games that scan your computer for stuff they think might be used to crack them (debuggers, CD mastering tools, etc.) and refuse to run while they're installed.
However, in real life, cars most often wind up in a scrap yard where usable parts are resold and the remainder is seldom properly recycled.
Seeing as Toyota have offered to pay for the battery packs, I dare say they are resold one way or the other by most scrap yards.
I know its jeremy clarkson and [...]
he's a one-man right-wing we-should-all-be-driving-15mpg-supercars spin machine, so anything he has to say on the subject is irrelevant.
There. Fixed that for you.
VW will beat the Prius in mileage handily
Yes, but what about (a) cost per mile and (b) energy used per mile. Who cares how much volume your fuel takes? Remember that diesel is both more and expensive and higher energy density than petrol.
Look for a car that uses diesel and get a stick shift. I'm not sure the mileage on a Volkswagon Jetta, but I've heard 42-45 city and 60ish highway.
Sounds plausible. But remember that diesel is (a) more expensive and (b) higher energy density than petrol. The figures are not directly comparable.
I'm more worried about the Lithium batteries recycling themselves explosively while I'm driving the vehicle!
Probably not worth worrying about. (1) Lithium batteries only tend to explode under charge. (2) The batteries used in cars are generally different chemistries to the ones used in laptops/phones etc, and are more stable.
The possibility of a toxic electrolyte spill is much more worrying, IMO.
Interesting article on businessweek.com. Ford is selling 65mpg cars in Europe, but not the US.
Well, sure. I don't see a 65mpg diesel as a particular achievement, though. Not in a small car. I mean, I get 40+ out of my Citroen XM, which is (a) over 10 years old and (b) probably weighs about 3 times what a Fiesta does.
Don't forget, diesel has a higher energy density than petrol, so you can't compare them gallon-for-gallon.
I think the Chevy Volt will have a Lithium Ion battery, the kind that has caused problems in laptops. I've heard they are working on reliability and safety issues with the batteries.
They're currently considering two different techs, Lithium-Manganese Oxide and Lithium-Iron Phosphate (aka LiFePO4). Neither of these are the same chemistry as the Li-ions used in laptops, and don't have the problem of producing excess gases during charging, so are (without any further work necessary) substantially safer.
I wrote: "specced to last (typically) 10,000 deep charge-discharge cycles"
Err... I meant 1,000. The rest of the figures remain unchanged.
I think it's more like replacing your car's fuel tank.
In terms of expense, it's definitely more like replacing the engine. Think ~$10,000 if you ever need to do it.
Of course, it's hard to see why you ever would. A hybrid is not like a battery electric; it uses the same sort of battery technology that's specced to last (typically) 10,000 deep charge-discharge cycles. But you're unlikely to ever see more than 50% discharge on a hybrid's battery pack (probably not even that), so you'll get a lot more cycles. 20,000 or more, I would guess. That equates to more than 200,000 miles with anything like a normal usage pattern. If you're doing long journeys it's probably more like 400,000 miles.
I think 65-100A delivering 21.5kW-33kW at least would be required to push the car around at a decent speed
Probably not, no. At 21.5kW a good electric motor typically delivers something like 100Nm (70 lbs.ft) at 4000rpm (or nearly-proportionally higher torque at lower speeds), which is more than enough to accelerate a reasonable weight vehicle quite quickly.
I agree that 6.5A@330V wouldn't be enough, but I don't think as much as 65 would be necessary. Looking at the specs for the Prius, its motor produces 450Nm at 400RPM, which probably equates to approximately 10kW. The _motor_ (not the battery pack) is described as max 6.5A, which suggests there's a voltage step-up somewhere in the control system.
3ms is still quite a bit slower than the NYSE, who (I think) claim 1ms for their linux-based system. But I imagine there are other factors here, like physical distance, the precise definition of 'transaction' and whether that's the guaranteed or typical speed.
I don't know about NYSE, but this system was supposed to have redundant implementations at two different locations with a 1-second failover between them. Obviously, in the event of a failure, that didn't work, but I presume they did at least implement it. This would likely have a substantial effect on transaction latency. I dunno if NYSE has a similar system or not?
I see no difference in Oracle performance on Windows vs Linux. Performance is usually more about architecture than the platform. You can make highly available networks with Windows, people do it all the time, much more often than a lot of Slashdotters seem to think.
Sure you can. The problem is that when you see a high profile Windows deployment like this (and, in fact, a lot of smaller deployments as well), you'll frequently find that MS has paid the consultants and told them what technology to use. So you often end up with a compromised architecture that's designed to show off a fancy new feature of the latest MS server software, rather than a properly engineered architecture that's as simple as possible (but no simpler), which is the way architecture should be done.
An example I know about in more detail: my company was hired about 6 or 7 years back to fix a catastrophic web site fuckup... the existing consultants hadn't finished the site in about a year and didn't seem to be capable of doing so, so the business owner came to us with 4 weeks before it was due to launch. It was a fairly simple web system, nothing fancy, a reverse auction system that allowed suppliers to bid using a back-end system on requests for supply that customers entered in the front end. The problem? Microsoft had paid the consultants to implement the back end with ASP and Access (!) while Macromedia had paid them to implement the front end entirely in flash version 4 (that's before flash supported actionscript). Every minor change to the user interface (e.g., "no I don't want to show 10 results on that list, let's go for 15") required hours of work including negotiations backwards and forwards between the ASP programmers and the Flash designers concerning the data structures (flat by necessity, so filled with items like 'result13title' and 'result13description' and stuff like that) that were passed between the client and the server.
Needless to say we ripped it all out, reimplemented in HTML & Javascript, and they launched with a much better site. But this is a cautionary tale: if you aren't careful, you can end up with a solution paid for by somebody who's trying to sell something, rather than one that's actually designed to fit your needs. And that solution may do something particularly well that you don't need, while it utterly fails to supply something you do need. In this case, it seems reliability (I would guess LSE's top requirement) has been sacrificed for performance (probably on the list, but I doubt it was top of it).
And there have been, what, 6 or so cases worldwide of death from Mad Cow? The threat is of the same order of magnitude.
Since 1990 there have been 1334 in the UK alone. Baseline figures for the US (in absence of any contaminated food sources, etc.) are 1 per 9,000 adults aged over 55, which based on a back-of-envelope calculation means somewhere in the region of 200 per annum.
More generally, unless you're a worker in the field where a risk occurs, a good rule of thumb is that it's a waste of time to worry about any danger that you hear about on the news. If it happens rarely enough to be newsworthy, it's not a danger to you
That I can subscribe to. :)
No different then what can happen on a unix box I suppose.
Note that the current system is built around a large cluster of 2.2GHz servers, while the unix-based system it replaced (which coped perfectly happily with a substantial portion of the same traffic) ran from a smaller cluster of much slower servers.
The primary purpose for the new system, introduced less than a year ago, was to expand capacity. For it to have failed within a year due to lack of capacity basically means that it has failed in that objective.
5 nines does not mean what you think it means.
No, you're right. By my calculation, the actual figure is more like 360 years.
(Remember, this is a system that only operates 7.5 hours per day, 250 days per year)
You've seen the first scene of "four weddings and a funeral", surely?
So prions can make proteins into prions in other species, e.a. cross the species barrier? Big deal.
The summary's wrong. Yes, we've known this for ages. What's new is we have a nice, fast, reliable way of studying without actually infecting animals, which could help us answer the important questions, like 'can this particular prion infect humans?'
So the public is donating a lot of computing time and electrical energy. What does the public get back?
If Folding@Home goes towards lining the pockets of a university endowment or a drug company's coffers, count me out. If the research product is required to be free from patents, and available for public good...full speed ahead. Somehow, I seriously doubt that any successful results will be freely available.
The project is run by a non-profit company. Results are routinely made public. I don't see any reason to be concerned here.
Uhuh. The summary is just wrong, is what's going on here. If you read the article, what it's actually about is a new way to study prions in vitro and test them to determine whether there's a possibility of them crossing a particular species barrier.
Which is, when you think about it, a very useful test to have available.
The good news is that prion-based diseases are vanishingly rare. Avoid cannibalism and you're safe: so far meteor strikes are about as dangerous a threat, along with heart-attack induced by the stress of winning the lottery.
Don't overestimate the threat of meteor strikes. While there have been a lot of near misses and property damage, there are apparently no recorded instances of somebody being killed by a meteorite since 1929.
Just so you know.