It's interesting to note that the Microsoft Security Bulletin does not disclose the component of the "Server Service" that is subject to the vulnerability. In particular, one cannot simply disable the relevant service. Actually, I don't even know whether their software is built to make such things possible. The reason I'm suspicious is because they recommend blocking certaing ports with a firewall rather than disabling the relevant component.
I'm completely unfamiliar with MS server software, but there seems to be a sharp contrast this bulletin with standard Unix practice where one can either edit inetd.conf and restrat the daemon (kill -HUP) or use rc.d start/stop scripts depending on the setup.
It's true that the onboard ATI & nVidia solutions aren't much better than Intel's. I suspect, however, that they share significant API with the high-end non-integreated cards from the same companies which are the real cash cows and therefore the technology they are trying to keep secret.
Well, Intel's integrated graphics chipset is a far cry from the nVidia / ATI high-end accellerators. Cloning it will be next to useless (who'll buy a separate graphics card to replaace an on-board solution?) since most other chipset manufacturers already have on-board solutions of their own. I doubt this will change the high-end makers rationale for keeping their drivers secret.
Don't confuse local/long distance (having to do with the location of the customers) and in/out of network (having to do with the providers they use. If you subscribe to Verizon then calling Verizon is cheaper than calling SBC etc.
If there is a compelling public interest, the government can legitimately restrict that choice. That's the case in many domains, and arguably, it should be the case here
At another thread someone has claimed that the situation is a near-monopoly (only one or two broadband providers in each area). In that case there's certainly a need for government regulation to prevent a monopoly extracting rents for the use of their infrastructure. Perhaps because I'm a city person I thought there was more competition. Probably rural areas will be harder-hit.
Here's the roll call for this bill. You can also get the full record of the bill (H.R. 5252), in particular see what happened to particular amendments.
What I want to know is, how can I get around their speed throttling for sites that do not pay up? I am not that savvy when it comes to coding my own scripts, but are there any tools that will help make things stay the same usage wise (if not price wise)?
If you ask me, the solution is simple: get a different broadband provider. By the way: do you mind that Verizon charges less when you call in-network as opposed to out-of-network?
If they can up-throttle comapnies that pay them, what's to stop them from down-throttling traffic from everyone who doesn't?
First of all, we're talking mostly about down-throttling: letting some packets go first and holding the others back. Secondly, there's nothing to stop them except that their paying customers will be pissed if they get slow service. Now the ISPs will try to say it's the website's fault: "MSN is our global partner, while Google declined our king offer to join the family, so of course MSN is faster", but I think people will quickly wise up and switch ISPs if that happens.
Unfortunately, this won't work that easily -- perhaps that's why the ISPs want to charge the service providers and not the end-users: it's easy to lie about the protocol/content of the packet, but it's very hard to lie about the source and destination address.
Indeed, people are going to be pissed off -- which is why I expect some ISPs to stay away from packet discrimination. People who care about it will simply flock there. The market is a better solution than hackers.
I assume we're talking here about ISPs discriminating in favour of their own paid subscription services, as opposed to the backbone operators doing the same. Now the ISP's infrastructure is private, and there seems to be a competition among ISPs. Will they all practice packet discrimination? I doubt it.
You can say that this breaks the "spirit of the internet", but some packet discrimination is essential when routers have to choose which packets to forward first, especially when some traffic should be low-latency, other high-bandwidth, other low-priority. I agree that the best solution is for the end-users to pay for their traffic, not the solution provider, but again -- it's the ISP's infrastructure and they can choose their own business model.
I think commercial space exploration essential -- we've realized most of the governmental benefits of space travel (GPS, weather satellites, went to the moon...), and except for science (which isn't enough to drive a space program) and the military (which won't pay for other stuff), nothing will happen unless we all start paying for it directly.
However, the government always like to meddle in private people's affairs. Sometimes it's necessary (someone should probably make sure the launch vehicles don't crash-land on people's houses), but most of the time it isn't, especially when we're talking about testing experimental technology.
Probably not everyone agrees, so: can you make a good case for the Space Precautionary Act?
According to the first page (site seems to slashdotted -- where's the CC ?), the best new features are the enhanced security and the new "user experience". From my Unix-based vantage point it hard to be excited when the cool new window manager is supposed to be an important reason to buy a new version of an operating system.
Actually, Microsoft was promising a genuine fundamental innovation (WinFS), but could deliver this in Vista. That would be something worth seeing.
The article seems to claim that one chip could be overclocked to 4.1GHz. That's a far cry from saying that all such chips will work at that clock speed. A sample size of one isn't very informative about a population.
This ain't a Microsoft problem. When Linus decided to change the driver model in the kernel, many hardware vendors had to rewrite their drivers. When Solaris 2.5 came out, all those SunOS 4.3 drivers became obsolete. Of course, if documentation of the upcoming Vista security model was hard to come by then these vendors would have a real beef, but no-one is saying that this is the case.
Note that NASA would be spending the $100M to ensure that things will work, preiod. The private developers are willing to assume a greater level of risk -- which is the main reason for the cost saving.
Note that even if they wanted to, LEA don't have the computing power available to monitor every call. On the other hand, analysing the call graph is quite tractable and completely orthogonal to the content. Enryption won't protect you from the government knowing who you're taking to!
Nope -- my point was that if Yahoo! is to operate in China, they should follow Chinese law. I'd say that IBM should not have operated in Germany. Similarly I don't think Yahoo! should operate in China under the current conditions (one reason why I don't have a Yahoo! account). I still think Yahoo! pulling out won't help the people of China any, but it's a small step.
The idea is a bit more than simply using codes in license actibation: the idea is to give a code with the software, that works for a short while. For longer-term use you have to register it and get the "permanent" password. In another variation the password changes periodically as a function of the previous password.
Of course this also existed in the 80s in the case of shareware. I still think it's clever, but my main complaint is that this isn't original or non-obviuos.
Indeed, and customers have a right to choose which companies they do business with. When Yahoo! operates in China, it should comply with Chinese law, no matter how evil it is. It is clear to us that Yahoo!'s Chinese customers prefer their service to no service at all (no-one made them sign up, probably, even if it's China).
If Yahoo!'s pulling out of China would put pressure on the government to allow more dissidence, it would be good, but I doubt it: if Yahoo! pulled out the Chinese government would form its own company. Thus, I suspect the people who are signing up for this service do so because it's better than nothing.
We feel different about it: we don't like a share in this kind of business, and probably do have a hope of sending a message to China. I therefore say unto you: cancel your Yahoo! accounts and find a different provider you like.
An interesting aside: the USSC has made it clear that the US Constitution only protects the rights of US citizens, even though the text simply forbids Congress from abridging various freedoms. Others who reside in the US apparently don't deserve protection. So why do the people of the US care so much about lack of rights for the people of China -- what about the Freedom of the Press and Due Process rights of Chinese people in the US?
Reading the patents (6,044,471 and 6,785,825) one is struck by a few things:
The patents (especially the second one) show a clever idea: force the user to register the software in order for it to operate. I don't know if this was an original idea in 2004, but it is clever. Of course, just because it's clever doesn't mean it's patentable.
This idea doesn't seem to be clever enough to be non-obvious as required by section 103 of US Code Title 35.
More seriuosly, the patents claim to provide an "apparatus and method", but fail to describe any actual appartus beyond the computer running the software under question. This is really a patent on an idea which contravenes section 102 of Title 35.
Beyond all this, the real question is of economics: did it cost Mr. Colvin $118M to develop this "invention"? Society has no incentive to allow people to monopolize ideas which have a zero development cost: people would invent them anyway since there's a profit motive even if other people can employ the invention. It should therefore be clear that the Patent Clause and US Code Title 35 were not intended to cover this invention. The fact that it was accepted anyway tell us a lot (that we already knew) about the US patent system. For example "non-obvious" has devloved to mean "not already known", a situation which is beyond words.
You realize, of course, that medical insurance that covers very expensive treatments has to be very expensive (it's the premiums that pay for the treatments!). My point was exactly that "complete coverage you can't afford" is not better than "incomplete coverage" even though this requires the understanding that not everyone will get the best coverage because not everyone is equally wealthy. Of course we can simply all share everything (ever been to a Kibbutz?) -- but that solution excepted there have to be inequalities.
Car safety works the same way. Let's say spending $1000 more on your car will make it somewhat safer. Perhaps investing the same $1000 on your car insurance instead might be better for you? Perhaps spending the same money on food for your kids might even be better, even though you'd be risking your life do drive a more dangerous car?
The "Libertarian" point is that it's usually not the government's place to make the decision for you. Sometime it is: when you make decisions that affect other people (e.g. about installing features that make your car less likely to kill a pedestrian). In general, however, it's likely that you're in the best place to decide what's best for you.
This doesn't come without costs, of course: some people will make decisions that are worse than the ones the government could have made for them (for example because they had worse education in decision-making). Because of this, not all "Libertarian" ideas have equal merit (at least as far as I'm concerned).
Slashdot Mirror
It's interesting to note that the Microsoft Security Bulletin does not disclose the component of the "Server Service" that is subject to the vulnerability. In particular, one cannot simply disable the relevant service. Actually, I don't even know whether their software is built to make such things possible. The reason I'm suspicious is because they recommend blocking certaing ports with a firewall rather than disabling the relevant component.
I'm completely unfamiliar with MS server software, but there seems to be a sharp contrast this bulletin with standard Unix practice where one can either edit inetd.conf and restrat the daemon (kill -HUP) or use rc.d start/stop scripts depending on the setup.
It's true that the onboard ATI & nVidia solutions aren't much better than Intel's. I suspect, however, that they share significant API with the high-end non-integreated cards from the same companies which are the real cash cows and therefore the technology they are trying to keep secret.
Well, Intel's integrated graphics chipset is a far cry from the nVidia / ATI high-end accellerators. Cloning it will be next to useless (who'll buy a separate graphics card to replaace an on-board solution?) since most other chipset manufacturers already have on-board solutions of their own. I doubt this will change the high-end makers rationale for keeping their drivers secret.
Don't confuse local/long distance (having to do with the location of the customers) and in/out of network (having to do with the providers they use. If you subscribe to Verizon then calling Verizon is cheaper than calling SBC etc.
Roll no. 239 is the one where they rejected 269-152 the net neutrality amendment.
At another thread someone has claimed that the situation is a near-monopoly (only one or two broadband providers in each area). In that case there's certainly a need for government regulation to prevent a monopoly extracting rents for the use of their infrastructure. Perhaps because I'm a city person I thought there was more competition. Probably rural areas will be harder-hit.
Here's the roll call for this bill. You can also get the full record of the bill (H.R. 5252), in particular see what happened to particular amendments.
If you ask me, the solution is simple: get a different broadband provider. By the way: do you mind that Verizon charges less when you call in-network as opposed to out-of-network?
First of all, we're talking mostly about down-throttling: letting some packets go first and holding the others back. Secondly, there's nothing to stop them except that their paying customers will be pissed if they get slow service. Now the ISPs will try to say it's the website's fault: "MSN is our global partner, while Google declined our king offer to join the family, so of course MSN is faster", but I think people will quickly wise up and switch ISPs if that happens.
Unfortunately, this won't work that easily -- perhaps that's why the ISPs want to charge the service providers and not the end-users: it's easy to lie about the protocol/content of the packet, but it's very hard to lie about the source and destination address.
Indeed, people are going to be pissed off -- which is why I expect some ISPs to stay away from packet discrimination. People who care about it will simply flock there. The market is a better solution than hackers.
I assume we're talking here about ISPs discriminating in favour of their own paid subscription services, as opposed to the backbone operators doing the same. Now the ISP's infrastructure is private, and there seems to be a competition among ISPs. Will they all practice packet discrimination? I doubt it.
You can say that this breaks the "spirit of the internet", but some packet discrimination is essential when routers have to choose which packets to forward first, especially when some traffic should be low-latency, other high-bandwidth, other low-priority. I agree that the best solution is for the end-users to pay for their traffic, not the solution provider, but again -- it's the ISP's infrastructure and they can choose their own business model.
The real question is: will there be a rule against taking sharp objects on board space planes?
I think commercial space exploration essential -- we've realized most of the governmental benefits of space travel (GPS, weather satellites, went to the moon ...), and except for science (which isn't enough to drive a space program) and the military (which won't pay for other stuff), nothing will happen unless we all start paying for it directly.
However, the government always like to meddle in private people's affairs. Sometimes it's necessary (someone should probably make sure the launch vehicles don't crash-land on people's houses), but most of the time it isn't, especially when we're talking about testing experimental technology.
Probably not everyone agrees, so: can you make a good case for the Space Precautionary Act?
According to the first page (site seems to slashdotted -- where's the CC ?), the best new features are the enhanced security and the new "user experience". From my Unix-based vantage point it hard to be excited when the cool new window manager is supposed to be an important reason to buy a new version of an operating system.
Actually, Microsoft was promising a genuine fundamental innovation (WinFS), but could deliver this in Vista. That would be something worth seeing.
The article seems to claim that one chip could be overclocked to 4.1GHz. That's a far cry from saying that all such chips will work at that clock speed. A sample size of one isn't very informative about a population.
This ain't a Microsoft problem. When Linus decided to change the driver model in the kernel, many hardware vendors had to rewrite their drivers. When Solaris 2.5 came out, all those SunOS 4.3 drivers became obsolete. Of course, if documentation of the upcoming Vista security model was hard to come by then these vendors would have a real beef, but no-one is saying that this is the case.
It has recently been determined that new versions of operating systems are not always 100% backward-compatible.
Note that NASA would be spending the $100M to ensure that things will work, preiod. The private developers are willing to assume a greater level of risk -- which is the main reason for the cost saving.
The real question for many of us is: will this protocol enforce anti-user controls? Perhaps someone knows more about these standards.
Note that even if they wanted to, LEA don't have the computing power available to monitor every call. On the other hand, analysing the call graph is quite tractable and completely orthogonal to the content. Enryption won't protect you from the government knowing who you're taking to!
Nope -- my point was that if Yahoo! is to operate in China, they should follow Chinese law. I'd say that IBM should not have operated in Germany. Similarly I don't think Yahoo! should operate in China under the current conditions (one reason why I don't have a Yahoo! account). I still think Yahoo! pulling out won't help the people of China any, but it's a small step.
The idea is a bit more than simply using codes in license actibation: the idea is to give a code with the software, that works for a short while. For longer-term use you have to register it and get the "permanent" password. In another variation the password changes periodically as a function of the previous password.
Of course this also existed in the 80s in the case of shareware. I still think it's clever, but my main complaint is that this isn't original or non-obviuos.
Indeed, and customers have a right to choose which companies they do business with. When Yahoo! operates in China, it should comply with Chinese law, no matter how evil it is. It is clear to us that Yahoo!'s Chinese customers prefer their service to no service at all (no-one made them sign up, probably, even if it's China).
If Yahoo!'s pulling out of China would put pressure on the government to allow more dissidence, it would be good, but I doubt it: if Yahoo! pulled out the Chinese government would form its own company. Thus, I suspect the people who are signing up for this service do so because it's better than nothing.
We feel different about it: we don't like a share in this kind of business, and probably do have a hope of sending a message to China. I therefore say unto you: cancel your Yahoo! accounts and find a different provider you like.
An interesting aside: the USSC has made it clear that the US Constitution only protects the rights of US citizens, even though the text simply forbids Congress from abridging various freedoms. Others who reside in the US apparently don't deserve protection. So why do the people of the US care so much about lack of rights for the people of China -- what about the Freedom of the Press and Due Process rights of Chinese people in the US?
Reading the patents (6,044,471 and 6,785,825) one is struck by a few things:
Beyond all this, the real question is of economics: did it cost Mr. Colvin $118M to develop this "invention"? Society has no incentive to allow people to monopolize ideas which have a zero development cost: people would invent them anyway since there's a profit motive even if other people can employ the invention. It should therefore be clear that the Patent Clause and US Code Title 35 were not intended to cover this invention. The fact that it was accepted anyway tell us a lot (that we already knew) about the US patent system. For example "non-obvious" has devloved to mean "not already known", a situation which is beyond words.
You realize, of course, that medical insurance that covers very expensive treatments has to be very expensive (it's the premiums that pay for the treatments!). My point was exactly that "complete coverage you can't afford" is not better than "incomplete coverage" even though this requires the understanding that not everyone will get the best coverage because not everyone is equally wealthy. Of course we can simply all share everything (ever been to a Kibbutz?) -- but that solution excepted there have to be inequalities.
Car safety works the same way. Let's say spending $1000 more on your car will make it somewhat safer. Perhaps investing the same $1000 on your car insurance instead might be better for you? Perhaps spending the same money on food for your kids might even be better, even though you'd be risking your life do drive a more dangerous car?
The "Libertarian" point is that it's usually not the government's place to make the decision for you. Sometime it is: when you make decisions that affect other people (e.g. about installing features that make your car less likely to kill a pedestrian). In general, however, it's likely that you're in the best place to decide what's best for you.
This doesn't come without costs, of course: some people will make decisions that are worse than the ones the government could have made for them (for example because they had worse education in decision-making). Because of this, not all "Libertarian" ideas have equal merit (at least as far as I'm concerned).