There are probably no boot loaders that would only load Linux. How would you distinguish that the executable being run was Linux and not yet another boot loader which would load Windows?
Umm... isn't one of the "strengths" of P2P that this would only be effective if everybody refused to peer with these addresses? Even if it were effective, wouldn't the parties involved just call up the phone company and order a DSL line--- with an address from the phone company's IP address block?
The same anonymity which P2P promises cuts both ways. Installing filters like this is a big waste of time. Now, accepting the connections but keeping them occupied via a fake "honeypot" network might at least be interesting...
No, I'm scared spitless. (I was just responding to the claim that this was a good test of the system's reliability.) We can just hope that anybody smart enough to destroy the Internet is busy using or running it.
The attackers were idiots. They used ICMP echo requests (easily filterable, since the DNS servers don't _have_ to answer those) and quit after an hour. More publicity stunt than actual attempt to damage, IMNSHO.
I've been trying to publish a paper about exactly this (and how to redesign DNS to avoid the vulnerability) and I'm just pissed that they didn't tell me in advance so that I could do some measurements.:)
I think this argument is complete B.S. Yes, most of the time your DSL line or cable modem connection is unused. Your ISP depends on that fact to provision and allocate its network resources. If enough people use P2P networks, then the usage changes and ISPs have to reprovision--- and guess what, that costs them more money. Which will get passed on to you.
(Or, more likely, the P2P traffic gets clamped down to the point where it's no longer worthwhile for anybody.)
There is no reason to believe that bandwidth is cheaper when you buy it in small chunks from 1000s of Internet users than in a big chunk from one provider. If DSL lines really are economically more efficient, buy a whole lot of DSL lines directly--- but you can get 10Mbit colocation (almost 60x DSL uplink speed) for $200/month. P2P network resources are "free" only because the cost has not yet been captured and passed on to the people running P2P nodes.
We could imagine a world in which there are P2P-allowed and P2P-prohibited ISPs. Which one would cost more? What application is going to be compelling enough to make up the cost difference?
Currently the only thing (outside of academia) which has proven successful is massive copyright violation.
In some cases the studio also has existing agreements with distributors, which prevent them from releasing the DVD while the distributor still has rights to the theatrical release in the region. No doubt they would have liked to have the same feature in videotapes. I can't see how selling a DVD in North America would get you sued for violating a contract in Europe, but that's movie industry logic for you...
And, unfortunately, pay for the privilege. Accepting a single arbitrator costs the domain name holder nothing; asking for a panel requires that he or she pay for half the cost--- win or lose.
This does even out the economic incentives somewhat. But even so, the UDRP provides choice of arbitrator to the complaintant, so market forces favor those arbitrators who tend to return pro-complaintant decisions. (I'm not suggesting any deliberate corruption, just magnification of any differences which naturally occur.)
Why? Who gave the South African government control over the.za domain? If there was a ".southafrica" domain, would they have automatic right to control that, too? If I invent a new namespace tomorrow, does that mean governments automatically get portions of it that they control? There's no "matter of course" about it. Many ccTLDs are not controlled directly or indirectly by the corresponding governments, but by universities or telephone companies.
Now, since the organization or person controlling a country's ccTLD usually resides in that country, it's not as if the government has no say...
But, the point is precisely that "this guy" is who the people running the root name servers chose as the administrator of the.za domain. There _is_ a process that ICANN has for transferring domain ownership. The South African government just doesn't want to play along; it wants to tell ICANN what to do, and it doesn't have that right. Nor is government appropriation of a previously private role to be taken lightly.
On the contrary, he welcomed the sinners of his day unconditionally, then instructed them to repent. Even the Ephesians passage you quote paints repentance and good behavior as a result, not a condition. See, for example,
Romans 5:1-10, or Luke 5:27ff.
A transport layer (TCP) checksum cannot detect all application-layer errors. And if you're running, say, a banking system, the 1 in 2^16 errors (or more) errors that the TCP checksum doesn't catch might be worth worrying about. (Fortunately, most of the Internet runs on top of link technologies which use CRCs, so the TCP checksum doesn't get exercised all that much.)
So why bash people for adding another layer of checking?
While "real-time" distributed systems are challenging, they're not impossible. (Distributed processing doesn't necessarily imply batching large data sets.) There has been a lot of Dept. of Defense-funded work on "Distributed Interactive Simulation" which splits the simulation processing over several different nodes. But this work operates in a trusted environment, not one where people will hack on the clients.
Dell still does offer Linux preinstalled, just not on all boxes. You can check out their Linux page and see the PowerEdge servers and Precision workstations. (No laptops.) My company has one of these in the wiring closet.
This story is another step in the direction the courts have been taking lately. It seems that making something available on the Internet exposes you to litigation anywhere in the world: France for Nazi memorabilia; Podunk, IA for porn; California for DVD trade secrets or E-book tools, etc. It's not a particularly encouraging trend.
The judges understood that there was an issue, they just said it was the publisher's problem to deal with it. This is in some ways understandable, though--- the system doesn't like being told "the technology doesn't support it", because it sounds like "we didn't think it was important enough to do", which isn't a defense. Peer-to-peer systems are likely to get trashed by the courts for the same reason; for example, saying that "there's no way we can comply with the DMCA and remove this copyrighted content" isn't going to convince any judge that you qualify for the DMCA's safe-harbour provisions (or that you should be able to run the software.)
Hmph! Just because us distributed computation people have to count messages in addition to everything else doesn't mean it's not an algorithm!
I admit the line between algorithm and protocol is fuzzy at best. It's say that Diffie-Hellman is an algorithm, but a specific use of it is a protocol. Just like BGP (the Internet routing protocol) is a (distributed) implementation of the Bellman-Ford algorithm.
Good point. I've never been bothered by the food cards--- since I usually pay by credit card, they _already_ have a unique identifier to track my purchasing habits if they want to.
However, I'm not sure that the grocery store people have actually been able to put that data to good use. The canonical "beer and diapers" story is urban legend at this point (unless somebody can provide me a good reference?) and only involved analysis of register receipts. What sorts of inferences could really be made by having the time-series for individual customers? Am I really supposed to believe I'd get a different ad in my mailbox if I spent more at the grocery store?
The only thing for which the cards make any sense to me is the ability to run long-term promotions--- like Safeway's "spend $500, get a coupon" or airline mileage tie-ins.
I agree. This sort of thing seems to me like all the "modern" sculptors who got too much in love with the new materials (fiberglass, latex, etc.) and forgot that they should be in the business of producing pleasing (or challenging, or instructive) works. As my wife--- an author and physics geek--- might say, "It's OK to do experimental art, but eventually you need to admit that the experiment didn't work and move on." I admit I might be proven wrong, but I haven't seen any "hyperlinked art" that I enjoy as much as traditional fiction.
Re:I can't find any reference...
on
SSSCA Hearing
·
· Score: 1
Sen. Hollings hasn't yet introduced the legislation, just drafted it. See the Wired article for a copy they obtained.
Let's turn it around--- how is it "logical, financially viable, and potentially useful" that you continue to have a monopoly right in perpetuity? Are you really going to publish more content because you have a 120-year right than a 20-year right? How much would you be willing to pay to extend this right if there were a "property tax" on this IP?
How many works will not be produced by other people because of your monopoly right? Would the world at large be noticably richer if the "copyright" on "Le Morte de Arthur" was still in force, or are we all better off with a rich public domain to draw upon?
There are the sort of questions Dr. Lessig raises in his book, "The Future of Ideas", which I highly recommend. (I agree that 20 years may be too short, but I don't think it automatically makes sense to include the entire life of the author.)
Re:17 USC 602 (a) (simplified )
on
Sony vs Modchips
·
· Score: 1
However, isn't the DVD region coding an "access control" mechanism, which is what the DMCA prohibits circumventing? The text of the law doesn't refer to encyrption, only
"circumvent[ing] a technological measure that effectively controls access to a work..."
It doesn't seem to have hurt them _too_ badly. The $320 million combined that the two other cable companies are paying is about what ATT was offering for a buyout. But I agree, it was a stupid move.
Excite@Home got started as @Home, an ISP specializing in cable modem service. (They provided the Internet connectivity, DHCP, proxies, web caches, servers, etc. for the cable companies.) Later they merged with Excite to add a "portal" (and other content) to their business, hoping to turn into the next AOL.
Perhaps the clearest way of putting the business arrangment (as it looked from the customer end, anyway) was that @Home's Internet service was distributed by AT&T. The combination of cable modem + Internet services is what gets labeled "ATT@Home", just like the same arrangement with Cox cable is "Cox@Home". AT&T was basically just in the business of hooking up subscribers to @Home's network; I paid ATT a subscription fee, then they turned around and gave part of it to @Home.
ATT Broadband is the name they've been using all along for their cable business, although now it seems to mean digital phones, too. But the intricacies of their internal arrangements escape me...
Slashdot Mirror
There are probably no boot loaders that would only load Linux. How would you distinguish that the executable being run was Linux and not yet another boot loader which would load Windows?
Umm... isn't one of the "strengths" of P2P that this would only be effective if everybody refused to peer with these addresses? Even if it were effective, wouldn't the parties involved just call up the phone company and order a DSL line--- with an address from the phone company's IP address block?
The same anonymity which P2P promises cuts both ways. Installing filters like this is a big waste of time. Now, accepting the connections but keeping them occupied via a fake "honeypot" network might at least be interesting...
They used "Klingons" in the synopsis on the front page of their web site; I have no idea about the actual show.
How about "Klingon"? In any case, little things like lack of grounds for suit never stopped an IP lawyer from sending threatening letters...
No, I'm scared spitless. (I was just responding to the claim that this was a good test of the system's reliability.) We can just hope that anybody smart enough to destroy the Internet is busy using or running it.
The attackers were idiots. They used ICMP echo requests (easily filterable, since the DNS servers don't _have_ to answer those) and quit after an hour. More publicity stunt than actual attempt to damage, IMNSHO.
I've been trying to publish a paper about exactly this (and how to redesign DNS to avoid the vulnerability) and I'm just pissed that they didn't tell me in advance so that I could do some measurements. :)
I think this argument is complete B.S. Yes, most of the time your DSL line or cable modem connection is unused. Your ISP depends on that fact to provision and allocate its network resources. If enough people use P2P networks, then the usage changes and ISPs have to reprovision--- and guess what, that costs them more money. Which will get passed on to you. (Or, more likely, the P2P traffic gets clamped down to the point where it's no longer worthwhile for anybody.)
There is no reason to believe that bandwidth is cheaper when you buy it in small chunks from 1000s of Internet users than in a big chunk from one provider. If DSL lines really are economically more efficient, buy a whole lot of DSL lines directly--- but you can get 10Mbit colocation (almost 60x DSL uplink speed) for $200/month. P2P network resources are "free" only because the cost has not yet been captured and passed on to the people running P2P nodes.
We could imagine a world in which there are P2P-allowed and P2P-prohibited ISPs. Which one would cost more? What application is going to be compelling enough to make up the cost difference? Currently the only thing (outside of academia) which has proven successful is massive copyright violation.
In some cases the studio also has existing agreements with distributors, which prevent them from releasing the DVD while the distributor still has rights to the theatrical release in the region. No doubt they would have liked to have the same feature in videotapes. I can't see how selling a DVD in North America would get you sued for violating a contract in Europe, but that's movie industry logic for you...
And, unfortunately, pay for the privilege. Accepting a single arbitrator costs the domain name holder nothing; asking for a panel requires that he or she pay for half the cost--- win or lose.
This does even out the economic incentives somewhat. But even so, the UDRP provides choice of arbitrator to the complaintant, so market forces favor those arbitrators who tend to return pro-complaintant decisions. (I'm not suggesting any deliberate corruption, just magnification of any differences which naturally occur.)
Interesting--- do you have a source for this we can check out? Who makes the rules?
Why? Who gave the South African government control over the .za domain? If there was a ".southafrica" domain, would they have automatic right to control that, too? If I invent a new namespace tomorrow, does that mean governments automatically get portions of it that they control? There's no "matter of course" about it. Many ccTLDs are not controlled directly or indirectly by the corresponding governments, but by universities or telephone companies.
Now, since the organization or person controlling a country's ccTLD usually resides in that country, it's not as if the government has no say...
But, the point is precisely that "this guy" is who the people running the root name servers chose as the administrator of the .za domain. There _is_ a process that ICANN has for transferring domain ownership. The South African government just doesn't want to play along; it wants to tell ICANN what to do, and it doesn't have that right. Nor is government appropriation of a previously private role to be taken lightly.
On the contrary, he welcomed the sinners of his day unconditionally, then instructed them to repent. Even the Ephesians passage you quote paints repentance and good behavior as a result, not a condition. See, for example, Romans 5:1-10, or Luke 5:27ff.
Ever heard of the end-to-end principle?
A transport layer (TCP) checksum cannot detect all application-layer errors. And if you're running, say, a banking system, the 1 in 2^16 errors (or more) errors that the TCP checksum doesn't catch might be worth worrying about. (Fortunately, most of the Internet runs on top of link technologies which use CRCs, so the TCP checksum doesn't get exercised all that much.)
So why bash people for adding another layer of checking?
While "real-time" distributed systems are challenging, they're not impossible. (Distributed processing doesn't necessarily imply batching large data sets.) There has been a lot of Dept. of Defense-funded work on "Distributed Interactive Simulation" which splits the simulation processing over several different nodes. But this work operates in a trusted environment, not one where people will hack on the clients.
Great quote from Allchin: "The fact that I even mentioned the Message Queuing thing bothers me."
Dell still does offer Linux preinstalled, just not on all boxes. You can check out their Linux page and see the PowerEdge servers and Precision workstations. (No laptops.) My company has one of these in the wiring closet.
This story is another step in the direction the courts have been taking lately. It seems that making something available on the Internet exposes you to litigation anywhere in the world: France for Nazi memorabilia; Podunk, IA for porn; California for DVD trade secrets or E-book tools, etc. It's not a particularly encouraging trend.
The judges understood that there was an issue, they just said it was the publisher's problem to deal with it. This is in some ways understandable, though--- the system doesn't like being told "the technology doesn't support it", because it sounds like "we didn't think it was important enough to do", which isn't a defense. Peer-to-peer systems are likely to get trashed by the courts for the same reason; for example, saying that "there's no way we can comply with the DMCA and remove this copyrighted content" isn't going to convince any judge that you qualify for the DMCA's safe-harbour provisions (or that you should be able to run the software.)
Hmph! Just because us distributed computation people have to count messages in addition to everything else doesn't mean it's not an algorithm!
I admit the line between algorithm and protocol is fuzzy at best. It's say that Diffie-Hellman is an algorithm, but a specific use of it is a protocol. Just like BGP (the Internet routing protocol) is a (distributed) implementation of the Bellman-Ford algorithm.
Good point. I've never been bothered by the food cards--- since I usually pay by credit card, they _already_ have a unique identifier to track my purchasing habits if they want to.
However, I'm not sure that the grocery store people have actually been able to put that data to good use. The canonical "beer and diapers" story is urban legend at this point (unless somebody can provide me a good reference?) and only involved analysis of register receipts. What sorts of inferences could really be made by having the time-series for individual customers? Am I really supposed to believe I'd get a different ad in my mailbox if I spent more at the grocery store?
The only thing for which the cards make any sense to me is the ability to run long-term promotions--- like Safeway's "spend $500, get a coupon" or airline mileage tie-ins.
I agree. This sort of thing seems to me like all the "modern" sculptors who got too much in love with the new materials (fiberglass, latex, etc.) and forgot that they should be in the business of producing pleasing (or challenging, or instructive) works. As my wife--- an author and physics geek--- might say, "It's OK to do experimental art, but eventually you need to admit that the experiment didn't work and move on." I admit I might be proven wrong, but I haven't seen any "hyperlinked art" that I enjoy as much as traditional fiction.
Sen. Hollings hasn't yet introduced the legislation, just drafted it. See the Wired article for a copy they obtained.
No longer have the exclusive rights.
Let's turn it around--- how is it "logical, financially viable, and potentially useful" that you continue to have a monopoly right in perpetuity? Are you really going to publish more content because you have a 120-year right than a 20-year right? How much would you be willing to pay to extend this right if there were a "property tax" on this IP?
How many works will not be produced by other people because of your monopoly right? Would the world at large be noticably richer if the "copyright" on "Le Morte de Arthur" was still in force, or are we all better off with a rich public domain to draw upon?
There are the sort of questions Dr. Lessig raises in his book, "The Future of Ideas", which I highly recommend. (I agree that 20 years may be too short, but I don't think it automatically makes sense to include the entire life of the author.)
However, isn't the DVD region coding an "access control" mechanism, which is what the DMCA prohibits circumventing? The text of the law doesn't refer to encyrption, only "circumvent[ing] a technological measure that effectively controls access to a work..."
It doesn't seem to have hurt them _too_ badly. The $320 million combined that the two other cable companies are paying is about what ATT was offering for a buyout. But I agree, it was a stupid move.
Excite@Home got started as @Home, an ISP specializing in cable modem service. (They provided the Internet connectivity, DHCP, proxies, web caches, servers, etc. for the cable companies.) Later they merged with Excite to add a "portal" (and other content) to their business, hoping to turn into the next AOL.
Perhaps the clearest way of putting the business arrangment (as it looked from the customer end, anyway) was that @Home's Internet service was distributed by AT&T. The combination of cable modem + Internet services is what gets labeled "ATT@Home", just like the same arrangement with Cox cable is "Cox@Home". AT&T was basically just in the business of hooking up subscribers to @Home's network; I paid ATT a subscription fee, then they turned around and gave part of it to @Home.
ATT Broadband is the name they've been using all along for their cable business, although now it seems to mean digital phones, too. But the intricacies of their internal arrangements escape me...