Right... like big or little endian makes the slightest bit of difference unless you have one of each.
It does. Consider the following code:
int i = [something]; short * s = (short *)
On little-endian processors, this achieves something useful (a pointer to a short initialised with the value of i, assuming it fits within the range of a short). On big-endian processors it gives you the most significant bits of i, which is unlikely to be useful.
My guess is that they will release new DVD-R's (general) discs that do not have the CDZ pre-written and only allow the drive manufacturers to release firmware to write the CDZ on those discs. This way they can charge extra money for the discs and not just be able to use old DVD+R(W)'s. The drive manufacturers could make some extra money too by not releasing firmware that allows this on older drives and marketing "new" drives that have this ability.
I don't think so - wouldn't this allow DVD-DVD duplication to be performed without needing a decss step? Preventing that was largely the point of CSS in the first place. I suspect what'll happen is the writer will read existing CSS key info from the media; the data downloaded will then have some other DRM info stripped and replaced with CSS encryption for the key from the media it will be recorded to before burning.
Me too..sort of. I mean....this being the death of Netflix?? Hardly!!
Netflix doesn't have any such stipulations...no late fees, no drm....you can do anything you want with your "backups"....
Well, the same's true with this, really. You burn to an image of a CSS DVD, then use DeCSS (or DVDShrink or some similar program that packages DeCSS) to strip the CSS so you have a DRM-free disc image. Burn as many times as you want. Rip to whatever format takes your fancy. Etc.
I thought most people considered themselves lucky if there documents could open in successive versions of Office.
Not that I've come across. I've used every version of Winword from 1.0 up to 2002, and each one has been able to open documents from the previous versions with a substantially higher degree of success than I get today opening them with OO.o.
And we won't even talk about how many networks I ran into that looked like
Server ----- hub ----- hub ----- hub ----- hub
and they wondered why the people on the 4th hub would lose server connections random
The problem there is the hub manufacturers who stick a switchable crossover into one of the ports of the hub and put in the instructions that you can daisychain it to another hub by plugging it in to one of those ports and flicking the switch. Reading that, you'd have no clue that there's any potential issue.
It crashed once every 31 days (there was some bug where it would crash after xxxx hours which was between 30-31 days).
You sure it wasn't every 24.86 days (or 2^31 milliseconds)?
I once ran a network for a couple of businesses in our office block who had pooled together for a shared DSL line. But the only box we had available that could support the hardware provided by the ISP was an old Win98SE box. Windows 98's dial up networking has this bug: after 2^31 milliseconds it stops transmitting packets, and you have to tear down the connection and reconnect. Guess nobody thought you'd ever have a dialup connected for that long.
To you left-ponders, "glorifying terrorism" is now illegal in the UK.
No, what is illegal is making "a statement that is likely to be understood by some or all of the members of the public to whom it is published as a direct or indirect encouragement or other inducement to them to the commission, preparation or instigation of acts of terrorism or Convention offences" (Terrorism Act 2006). Note that encouraging or inducing others to commit a criminal offence is also illegal in the US, and was already illegal in the UK.
The Prevention of Terrorism Act 2005 allows the secretary of state to make an order that will place its subject under "a prohibition or restriction on his movements to, from or within the United Kingdom, a specified part of the United Kingdom or a specified place or area within the United Kingdom" (i.e., house arrest or confinement in a detention centre).
Then:
11 Jurisdiction and appeals in relation to control order decisions etc.
(1) Control order decisions and derogation matters are not to be questioned in any legal proceedings other than-
(a) proceedings in the [High] court; or
(b) proceedings on appeal from such proceedings.
(2) The court is the appropriate tribunal for the purposes of section 7 of the Human Rights Act 1998 (c. 42) in relation to proceedings all or any part of which call a control order decision or derogation matter into question.
(3) No appeal shall lie from any determination of the court in control order proceedings, except on a question of law.
There is no right to demand a trial, all you can do is question the legality (but not factual accuracy) of proceedings that have already occurred. An option open to a judge in a habeas corpus proceeding is to require a trial be held (see the Habeas Corpus Act 1679); that option is specifically excluded for a judge considering the case of somebody held under a terrorism control order.
Theoretically, yes. If you can find some way of delivering the sequence into the correct cells and inserting it into their DNA without affecting non-cancerous cells. But the same is true of many other solutions which are already known: killing cells isn't hard. Finding the ones to target is.
What happens when big funding starts to demand what can and can't be placed into articles? "We're sorry, Wikipedia, but I'm going to need you to remove this, that and the other fact from the article because it might turn away our potential customers."
There's two possibilities: 1. The Wikimedia Foundation tells them that they don't control the content, and that wikipedia isn't censored, and either the advertiser relents or goes somewhere else. 2. They bend over backwards and implement content restrictions. At this point, because the content is free, we're perfectly able (and motivated enough) to fork the site and create an uncensored version. It would be incovenient, but I think it would be an unlikely scenario to start with.
If the robot is a finite state machine where of all of it's output and processes are strictly defined, there's no chance that it's somehow self-aware or anywhere close to that.
Are you sure? It is worth considering in answering this question that many (most?) experts believe that a real computer (which is, effectively, a finite state machine with a large number of possible states) will be able to realistically simulate the operation of a human brain at some point in the future. Experts don't agree, however, on whether such a simulation would actually be self aware.
Trying to blur the lines via some semantics argument doesn't hide the fact that the only behaviours machines have are the behaviours we instruct them to have.
Except that this isn't true. We have today a variety of machine-learning algorithms that allow the creation of machines that behave in ways that we have not instructed them to. The two most promising are neural networks and genetic algorithms, both of which effectively specify a basic machine that is capable of performing many tasks and a method of training that allows it to learn to perform the task without us specifying the method by which it should be performed.
1. What context does the js execute in? Browser or Acrobat? If Acrobat, does it have access to your cookies? (I'd guess not) 2. What versions/browsers are affected? I'm using FF2 with Acrobat 5, and nothing seems to happen, but this could be because I've got an odd setup.
Why doesn't everyone make such a big stink about preventing the number of falls in the United States?
Because AFAIK the US doesn't have a significantly higher number of such deaths when compared to other similar countries. The fact that the murder rate is above average implies that there should be some way of reducing it to average.
Then you have homicides "committed during a rape, robbery, burglary, theft, motor vehicle theft, arson, and violations of prostitution and commercial vice laws, other sex offenses, narcotic drug laws, and gambling laws." Again, I don't see that cameras would decrease homicides in those circumstances
Actually, cameras do decrease these instances because they make armed robbery and burglary less attractive crimes (as these are typically committed at premises that are likely to have such cameras), pushing criminals towards crimes that are less likely to result in a death.
Where's your evidence for this? The stats you've presented so far suggest that the US (with few cameras) has a substantially higher murder rate than the UK (with many cameras). Granted, correlation !=> causation, but that's as good as the argument you're using.
Perhaps the US police are just better at catching murderers than the UK police, having had more experience...?
I don't see how phone call monitoring is any different to the records your ISP is required legally to keep about every url you visit. Please explain the difference.
1. In the UK, your ISP is not legally required to keep URL histories for their customers. There is a voluntary code of practice that suggests that they should track the hostnames of sites you access through their proxy servers. As long as you aren't using a proxy (and they aren't transparently proxying), you should be fine. 2. While they do track every e-mail sent via their servers (and are, I believe, legally required to), this can easily be circumvented by not using their e-mail service (which I don't). The only way of avoiding phone call monitoring (other than not using phones!) is using a voice scrambler. I understand that this technology is expensive, difficult to use, and (in the UK) illegal.
Goggle already tracks all the emails I send through their system, and uses it for their commercial advantage - were it not google doing this it'd be yahoo, or NTL
Try a different system, if you're concerned. Hushmail PGP-encrypts all email on arrival without any other processing, and never has an unencrypted copy of your private key on their servers.
even if I used my own mail server, every ADSL contract I have ever seen has a clause in there that allows them to monitor the traffic from servers that you run
Yes. Their acceptable use policy would be unenforceable otherwise, because they have to be able to collect evidence that you've breached it. There's nothing sinister here; they don't routinely packet-sniff and log the details of communications sent to servers you run. Only if there's a complaint.
Companies, employers, parents etc already have access to your credit history
In the UK, nobody has (legal) access to your credit history unless you give them permission. Employers aren't even allowed to ask.
Now were the govenrment to bring in a law that would allow them to hold people for as long as they want wothout evidence or reason, then that would be something to worry about
You mean the fact that they already have this power over anyone who isn't British (Anti-terrorism, Crime and Security Act 2001) isn't worrying enough?
Police have always been able to search your home without telling you - with a court order - this doesn't ever seem likly to change.
Yes, but the court is independent of the police, (theoretically) non-political, and must be convinced that there is a good-enough reason to do so. Given that the power must exist (I think this is reasonable), then this is the best way of having it.
They can tap your phone, gain access to your Internet communications and require anybody who they can show grounds to believe has access to present your encryption keys to them without ever involving a court (Regulation of Investigatory Powers Act 2000). This is the real problem.
I think the mechanism is that people will often make the decision to go through lights on yellow. They know they shouldn't, but figure if they're quick they'll get across fine. There's not really anything wrong with this: it is, in fact, the purpose of the yellow phase existing. But if somebody who's just decided to do this spots a camera in the instant between deciding to do it and crossing the line, they'll often slam the breaks on. Quite frequently, this results in somebody behind them (who was also planning on doing the same thing) hitting them in the rear end.
I don't know how well having good warning signs about the cameras mitigates this issue. It might or might not work effectively. I don't even know if the strategy was in use in the area the study was conducted in.
Not even sure about that. Sure, we have a lot of cameras here. City councils have installed them in their most problematic commercial districts. Business and private property owners have installed them in the hope of deterring crime. The police have mobile cameras they drive around potential trouble-spots. And these decisions are largely justified.
I can't think of any way of improving the situation. We already have laws (e.g. the Data Protection Act) that govern the way video recorded from such cameras can be used, and places where cameras can legally be installed (i.e., not in a place where you would normally expect to have privacy). As far as I'm aware, these laws aren't widely violated.
Personally, I find the prospect of having my car's number plate recorded when scanned by automatic number-plate recognition systems substantially more worrying. It's a lot easier to track somebody that way than via security cameras. The result described in the article probably took hundreds of hours of work. Looking up somebody's number plate on a database of plate/time/location records would be the work of minutes.
Well, that's the idea of paravirtualization: make an easier target to emulate and then port the OS to it. And yes, people are doing OS development with paravirtualized targets as their initial architecture. Xen is quite popular among the hobbyist-OS community.
Slashdot Mirror
Right... like big or little endian makes the slightest bit of difference unless you have one of each.
It does. Consider the following code:
int i = [something];
short * s = (short *)
On little-endian processors, this achieves something useful (a pointer to a short initialised with the value of i, assuming it fits within the range of a short). On big-endian processors it gives you the most significant bits of i, which is unlikely to be useful.
My guess is that they will release new DVD-R's (general) discs that do not have the CDZ pre-written and only allow the drive manufacturers to release firmware to write the CDZ on those discs. This way they can charge extra money for the discs and not just be able to use old DVD+R(W)'s. The drive manufacturers could make some extra money too by not releasing firmware that allows this on older drives and marketing "new" drives that have this ability.
I don't think so - wouldn't this allow DVD-DVD duplication to be performed without needing a decss step? Preventing that was largely the point of CSS in the first place. I suspect what'll happen is the writer will read existing CSS key info from the media; the data downloaded will then have some other DRM info stripped and replaced with CSS encryption for the key from the media it will be recorded to before burning.
Me too..sort of. I mean....this being the death of Netflix?? Hardly!!
Netflix doesn't have any such stipulations...no late fees, no drm....you can do anything you want with your "backups"....
Well, the same's true with this, really. You burn to an image of a CSS DVD, then use DeCSS (or DVDShrink or some similar program that packages DeCSS) to strip the CSS so you have a DRM-free disc image. Burn as many times as you want. Rip to whatever format takes your fancy. Etc.
I thought most people considered themselves lucky if there documents could open in successive versions of Office.
Not that I've come across. I've used every version of Winword from 1.0 up to 2002, and each one has been able to open documents from the previous versions with a substantially higher degree of success than I get today opening them with OO.o.
And we won't even talk about how many networks I ran into that looked like
Server ----- hub ----- hub ----- hub ----- hub
and they wondered why the people on the 4th hub would lose server connections random
The problem there is the hub manufacturers who stick a switchable crossover into one of the ports of the hub and put in the instructions that you can daisychain it to another hub by plugging it in to one of those ports and flicking the switch. Reading that, you'd have no clue that there's any potential issue.
It crashed once every 31 days (there was some bug where it would crash after xxxx hours which was between 30-31 days).
You sure it wasn't every 24.86 days (or 2^31 milliseconds)?
I once ran a network for a couple of businesses in our office block who had pooled together for a shared DSL line. But the only box we had available that could support the hardware provided by the ISP was an old Win98SE box. Windows 98's dial up networking has this bug: after 2^31 milliseconds it stops transmitting packets, and you have to tear down the connection and reconnect. Guess nobody thought you'd ever have a dialup connected for that long.
Funniest post on slashdot for weeks. Thanks. :)
To you left-ponders, "glorifying terrorism" is now illegal in
the UK.
No, what is illegal is making "a statement that is likely to be understood by some or all of the members of the public to whom it is published as a direct or indirect encouragement or other inducement to them to the commission, preparation or instigation of acts of terrorism or Convention offences" (Terrorism Act 2006). Note that encouraging or inducing others to commit a criminal offence is also illegal in the US, and was already illegal in the UK.
You sure?
The Prevention of Terrorism Act 2005 allows the secretary of state to make an order that will place its subject under "a prohibition or restriction on his movements to, from or within the United Kingdom, a specified part of the United Kingdom or a specified place or area within the United Kingdom" (i.e., house arrest or confinement in a detention centre).
Then:
There is no right to demand a trial, all you can do is question the legality (but not factual accuracy) of proceedings that have already occurred. An option open to a judge in a habeas corpus proceeding is to require a trial be held (see the Habeas Corpus Act 1679); that option is specifically excluded for a judge considering the case of somebody held under a terrorism control order.
Couldn't you simply watch the right MacGyver episode?
No, cause in MacGuyver, they always left out a key detail, so it wouldn't work if you tried to reproduce it.
Either that or they sucked at research.
Theoretically, yes. If you can find some way of delivering the sequence into the correct cells and inserting it into their DNA without affecting non-cancerous cells. But the same is true of many other solutions which are already known: killing cells isn't hard. Finding the ones to target is.
What happens when big funding starts to demand what can and can't be placed into articles? "We're sorry, Wikipedia, but I'm going to need you to remove this, that and the other fact from the article because it might turn away our potential customers."
There's two possibilities:
1. The Wikimedia Foundation tells them that they don't control the content, and that wikipedia isn't censored, and either the advertiser relents or goes somewhere else.
2. They bend over backwards and implement content restrictions. At this point, because the content is free, we're perfectly able (and motivated enough) to fork the site and create an uncensored version. It would be incovenient, but I think it would be an unlikely scenario to start with.
If the robot is a finite state machine where of all of it's output and processes are strictly defined, there's no chance that it's somehow self-aware or anywhere close to that.
Are you sure? It is worth considering in answering this question that many (most?) experts believe that a real computer (which is, effectively, a finite state machine with a large number of possible states) will be able to realistically simulate the operation of a human brain at some point in the future. Experts don't agree, however, on whether such a simulation would actually be self aware.
Trying to blur the lines via some semantics argument doesn't hide the fact that the only behaviours machines have are the behaviours we instruct them to have.
Except that this isn't true. We have today a variety of machine-learning algorithms that allow the creation of machines that behave in ways that we have not instructed them to. The two most promising are neural networks and genetic algorithms, both of which effectively specify a basic machine that is capable of performing many tasks and a method of training that allows it to learn to perform the task without us specifying the method by which it should be performed.
Yet, we may have a problem understanding its reasoning.
;)
'Cause, of course, traditional human philosophers' reasoning is simple to follow.
There's a lot of missing information here.
1. What context does the js execute in? Browser or Acrobat? If Acrobat, does it have access to your cookies? (I'd guess not)
2. What versions/browsers are affected? I'm using FF2 with Acrobat 5, and nothing seems to happen, but this could be because I've got an odd setup.
Anyone know?
Why doesn't everyone make such a big stink about preventing the number of falls in the United States?
Because AFAIK the US doesn't have a significantly higher number of such deaths when compared to other similar countries. The fact that the murder rate is above average implies that there should be some way of reducing it to average.
Then you have homicides "committed during a rape, robbery, burglary, theft, motor vehicle theft, arson, and violations of prostitution and commercial vice laws, other sex offenses, narcotic drug laws, and gambling laws." Again, I don't see that cameras would decrease homicides in those circumstances
Actually, cameras do decrease these instances because they make armed robbery and burglary less attractive crimes (as these are typically committed at premises that are likely to have such cameras), pushing criminals towards crimes that are less likely to result in a death.
More cameras don't prevent more murders
Where's your evidence for this? The stats you've presented so far suggest that the US (with few cameras) has a substantially higher murder rate than the UK (with many cameras). Granted, correlation !=> causation, but that's as good as the argument you're using.
Perhaps the US police are just better at catching murderers than the UK police, having had more experience...?
I don't see how phone call monitoring is any different to the records your ISP is required legally to keep about every url you visit. Please explain the difference.
1. In the UK, your ISP is not legally required to keep URL histories for their customers. There is a voluntary code of practice that suggests that they should track the hostnames of sites you access through their proxy servers. As long as you aren't using a proxy (and they aren't transparently proxying), you should be fine.
2. While they do track every e-mail sent via their servers (and are, I believe, legally required to), this can easily be circumvented by not using their e-mail service (which I don't). The only way of avoiding phone call monitoring (other than not using phones!) is using a voice scrambler. I understand that this technology is expensive, difficult to use, and (in the UK) illegal.
Goggle already tracks all the emails I send through their system, and uses it for their commercial advantage - were it not google doing this it'd be yahoo, or NTL
Try a different system, if you're concerned. Hushmail PGP-encrypts all email on arrival without any other processing, and never has an unencrypted copy of your private key on their servers.
even if I used my own mail server, every ADSL contract I have ever seen has a clause in there that allows them to monitor the traffic from servers that you run
Yes. Their acceptable use policy would be unenforceable otherwise, because they have to be able to collect evidence that you've breached it. There's nothing sinister here; they don't routinely packet-sniff and log the details of communications sent to servers you run. Only if there's a complaint.
Companies, employers, parents etc already have access to your credit history
In the UK, nobody has (legal) access to your credit history unless you give them permission. Employers aren't even allowed to ask.
Now were the govenrment to bring in a law that would allow them to hold people for as long as they want wothout evidence or reason, then that would be something to worry about
You mean the fact that they already have this power over anyone who isn't British (Anti-terrorism, Crime and Security Act 2001) isn't worrying enough?
Police have always been able to search your home without telling you - with a court order - this doesn't ever seem likly to change.
Yes, but the court is independent of the police, (theoretically) non-political, and must be convinced that there is a good-enough reason to do so. Given that the power must exist (I think this is reasonable), then this is the best way of having it.
They can tap your phone, gain access to your Internet communications and require anybody who they can show grounds to believe has access to present your encryption keys to them without ever involving a court (Regulation of Investigatory Powers Act 2000). This is the real problem.
I think the mechanism is that people will often make the decision to go through lights on yellow. They know they shouldn't, but figure if they're quick they'll get across fine. There's not really anything wrong with this: it is, in fact, the purpose of the yellow phase existing. But if somebody who's just decided to do this spots a camera in the instant between deciding to do it and crossing the line, they'll often slam the breaks on. Quite frequently, this results in somebody behind them (who was also planning on doing the same thing) hitting them in the rear end.
I don't know how well having good warning signs about the cameras mitigates this issue. It might or might not work effectively. I don't even know if the strategy was in use in the area the study was conducted in.
With results like these, is there really a good basis for argument FOR these cameras?
The cameras that article is talking about are cameras to catch red-light jumping.
The cameras the original article is talking about are private security cameras in shops, etc.
These are different types of camera, so there is no reason to believe that results of studies into the effects of one should be valid for the other.
the UK has gone way overboard
Not even sure about that. Sure, we have a lot of cameras here. City councils have installed them in their most problematic commercial districts. Business and private property owners have installed them in the hope of deterring crime. The police have mobile cameras they drive around potential trouble-spots. And these decisions are largely justified.
I can't think of any way of improving the situation. We already have laws (e.g. the Data Protection Act) that govern the way video recorded from such cameras can be used, and places where cameras can legally be installed (i.e., not in a place where you would normally expect to have privacy). As far as I'm aware, these laws aren't widely violated.
Personally, I find the prospect of having my car's number plate recorded when scanned by automatic number-plate recognition systems substantially more worrying. It's a lot easier to track somebody that way than via security cameras. The result described in the article probably took hundreds of hours of work. Looking up somebody's number plate on a database of plate/time/location records would be the work of minutes.
The article does one better and specifically mentions KVM+Qemu, but don't take my word for it.
Which is entirely different to qemu with qemu-accel; the latter is *not* a kernel module.
Well, that's the idea of paravirtualization: make an easier target to emulate and then port the OS to it. And yes, people are doing OS development with paravirtualized targets as their initial architecture. Xen is quite popular among the hobbyist-OS community.