The ruling considers the differences between this case and the NES lockout case. In the NES case, people were copying the program that output the necessary unlock code. The court ruled that the lockout defense was insufficient, because it was possible to write a different program to produce that same unlock code.
In this case, the program itself was the unlock code (a checksum was done of it).
Re:Update TCP, don't add new protocol
on
Replacing TCP?
·
· Score: 1
New transport protocol (like the one in the article) are building from what we've learned from TCP, and in general try to provide the same API (a socket is a socket). All that's needed to make use of them is to change the type of sockets you are using. (This could be done at the system library level if the new protocol is deemed a replacement)
Adding flags to TCP that alter the entire behavior of the protocol is asking for trouble. It will only make TCP implementations more complicated.
Keep in mind that the new protocols aren't the best for all applications. The protocol described in the article is great for situations with some packet loss, but isn't the best when there isn't packet loss.
Re:Would this "steal bandwith"?
on
Replacing TCP?
·
· Score: 1
This protocol, like most TCP replaces are TCP-friendly. That is there are designed so that send the same amount of traffic on average as TCP would in a perfect situation. So in theory, they shouldn't make existing TCP connections slower.
That's one of the arguments for new transport protocols (TCP replacements), TCP has been tweaked for decades, so if researcher can in a few months using a handful of graduate students notably beat it performance-wise, TCP is pretty bad.
(disclaimer: currenting doing adapative transport protocol research)
As he's legally obligated to do what's best of his business interests, and the state enforces its laws with guns, he does have a gun to his head in the matter.
Don't know about Unreal, but starcraft can't be run at a non-native resolution. It runs at 640x480 I believe always. Powerbooks at just smart enough to scale it properly to fit the screen.
Checked exceptions are vital to writing reliable code.
Those "catch (SomeSillyException) {/* Won't happen */ }" lines you're running are places in your code where something unexpected is happening and you are ignoring it. 9 times out of 10 is this a bug or security problem.
Physical memory is always better than swap, nobody is claiming otherwise. A system with 1GB of RAM and a 1GB of swap should be faster than the same system withou swap.
While I admit that I'll be surprised if they win as well, this case is really distinct from Eldred v. Ashcroft, and actually relies heavily on the decisions from it.
Yeah, forget the law offhand. In short, barring written permission of the user or a court order, what they did was illegal. (This ignores terrorism junk laws, which I'm not up on)
You need to hold your ISPs up standards. It's not that hard.
I've got a 640/256 DSL line. I've got no terms of service, beyond that it costs $20/mo. Period, end story. Got a static ip by just asking (under most ISP setups, if you're using have constant on connections, this doesn't really cost anything).
Maybe, I've gotten lucky, but these are reasonable terms.
Another useful bit, pay for your connection for a period ahead of time (I'm on a quarterly billing cycle). What's this mean, yeah, they can threaten to disconnect me, but they've offered the service, I've paid, they've excepted payment, contract made, end story.
Don't accept companies that want to be able to terminate you at any moment. That gives them a great deal of power. My connection is vital to my life/profession/academics, I need lead time to switch ISPs. I've got 3 months notice.
BTW, DSL providers are generally much more reasonable. The fact they give you solid bandwidth numbers take a lot of wiggle room out of the deal.
I second Timbuk2. Very solid, waterproof, and if powerbooks are your thing, you can get them custom fit. Particularly nice is they have a padded inner sleave for the laptop. Providing full protected, while leaving room for other stuff.
They can only use that to keep from distributing APIs that compromise security not merely related. Personally if an API is such that disclosing the API (not algorithms, not implementation details, just the API) is sufficient to compromise the security of the product, then the API contains a backdoor.
Comit yourself beforehand to going to every class every single time and stick to it. Even if you are just going to design your latest programming project in your notebook or daze-out, all the academic stuff is easiest if you are there everytime.
If you don't think you can do that, minimally keep track of how often you are missing your classes. It's easy to lose track and realize you haven't been to calculus for 3 weeks. (That's a bad thing by the way)
Slashdot Mirror
The ruling considers the differences between this case and the NES lockout case. In the NES case, people were copying the program that output the necessary unlock code. The court ruled that the lockout defense was insufficient, because it was possible to write a different program to produce that same unlock code.
In this case, the program itself was the unlock code (a checksum was done of it).
New transport protocol (like the one in the article) are building from what we've learned from TCP, and in general try to provide the same API (a socket is a socket). All that's needed to make use of them is to change the type of sockets you are using. (This could be done at the system library level if the new protocol is deemed a replacement)
Adding flags to TCP that alter the entire behavior of the protocol is asking for trouble. It will only make TCP implementations more complicated.
Keep in mind that the new protocols aren't the best for all applications. The protocol described in the article is great for situations with some packet loss, but isn't the best when there isn't packet loss.
This protocol, like most TCP replaces are TCP-friendly. That is there are designed so that send the same amount of traffic on average as TCP would in a perfect situation. So in theory, they shouldn't make existing TCP connections slower.
That's one of the arguments for new transport protocols (TCP replacements), TCP has been tweaked for decades, so if researcher can in a few months using a handful of graduate students notably beat it performance-wise, TCP is pretty bad.
(disclaimer: currenting doing adapative transport protocol research)
Yeah, they say to the police, and they'll get no worse than the charges they'd initially get, while you'll get charged with grand theft. Great scheme.
The difference between this rule and a rule about changing your oil is that there's a federal law saying an landlord can enforce the former.
The point of the case is that the gov't won't tell us whether such a law exists, and if so the details.
As he's legally obligated to do what's best of his business interests, and the state enforces its laws with guns, he does have a gun to his head in the matter.
Don't know about Unreal, but starcraft can't be run at a non-native resolution. It runs at 640x480 I believe always. Powerbooks at just smart enough to scale it properly to fit the screen.
So you're the guy that made me spend an hour trying to explain to my father what a stack trace was!
Man, you need to get your priorities straight.
Runtime code generation is very fun, arguable useful, and unquestionable too big a security issue to consider ever using.
Checked exceptions are vital to writing reliable code.
/* Won't happen */ }" lines you're running are places in your code where something unexpected is happening and you are ignoring it. 9 times out of 10 is this a bug or security problem.
Those "catch (SomeSillyException) {
Runs out to patent: A process by which ferrets to rob banks
A quick check of the page shows no screenshots. OSS without screenshots is just not competitive.
That's exactly what many US undergrad OS courses are.
Physical memory is always better than swap, nobody is claiming otherwise. A system with 1GB of RAM and a 1GB of swap should be faster than the same system withou swap.
Should is the tricky part.
While I admit that I'll be surprised if they win as well, this case is really distinct from Eldred v. Ashcroft, and actually relies heavily on the decisions from it.
Yeah, forget the law offhand. In short, barring written permission of the user or a court order, what they did was illegal. (This ignores terrorism junk laws, which I'm not up on)
You need to hold your ISPs up standards. It's not that hard.
I've got a 640/256 DSL line. I've got no terms of service, beyond that it costs $20/mo. Period, end story. Got a static ip by just asking (under most ISP setups, if you're using have constant on connections, this doesn't really cost anything).
Maybe, I've gotten lucky, but these are reasonable terms.
Another useful bit, pay for your connection for a period ahead of time (I'm on a quarterly billing cycle). What's this mean, yeah, they can threaten to disconnect me, but they've offered the service, I've paid, they've excepted payment, contract made, end story.
Don't accept companies that want to be able to terminate you at any moment. That gives them a great deal of power. My connection is vital to my life/profession/academics, I need lead time to switch ISPs. I've got 3 months notice.
BTW, DSL providers are generally much more reasonable. The fact they give you solid bandwidth numbers take a lot of wiggle room out of the deal.
I second Timbuk2. Very solid, waterproof, and if powerbooks are your thing, you can get them custom fit. Particularly nice is they have a padded inner sleave for the laptop. Providing full protected, while leaving room for other stuff.
How many times do we have to go over this?
They can only use that to keep from distributing APIs that compromise security not merely related. Personally if an API is such that disclosing the API (not algorithms, not implementation details, just the API) is sufficient to compromise the security of the product, then the API contains a backdoor.
the US is civilized because they don't do that morally wrong and uncivilized thing very often?
If you don't think you can do that, minimally keep track of how often you are missing your classes. It's easy to lose track and realize you haven't been to calculus for 3 weeks. (That's a bad thing by the way)