Sure, these.tv guys are turkeys, but I think it's not unreasonable to charge premiums for domain names that are more "in demand." Let's let the market decide how much they're worth.
Seems like this proposal is relying on the web site being able to change its address and routing info faster than the attackers are able to react. This is a losing proposition.
It's apparent that a distributed attack requires a distributed solution. But as many posters have pointed out, that usually requires a lot of cooperation from a lot of ISPs. But what about using the caching infrastructure? Assume the web site under attack actually allowed its content to be cached (pretty big assumption due to cookies and banner ads). Now if the site is attacked, all the content that's currently cached could still be served from caches. This doesn't cover dynamically generated content or content that hasn't been cached. So this is a partial solution, but it's deployable right now, today, this instant.
Sorry if this has already been brought up - I haven't been reading every comment in the moderation discussions.
The question: does MM create a feedback loop? The people that moderation rewards get to reward the moderators, right? It certainly seems like it's got the potential to be unstable.
What if, instead of MM, we just defined a "bad" moderator to be someone whose use of points is contrary to other moderators'? For example, if one moderator loved an article and many others hated it, then the moderator who loved the article is doing a bad job. I'm not sure how well this would fit into the/. moderation system though.
It's not light (6.7 pounds) It's not fast (300 MHz G3) Its display isn't great (800x600 max) Its price is sorta ok ($1600?)
But Apple put it in a brightly colored case, so people will want to buy it. I'm starting to think that Apple's strongest feature is its slick marketing.
To make a distributed solution work, the system would need to know the relative popularity of each movie: that way, the highly demanded movies could be placed on multiple servers since a single $2k PC won't be able to serve the movie to many clients at once (no, "multicast" won't solve this problem). That's why a central server makes so much sense: if the server is up to the challenge, then you don't have the headache of trying to do good load balancing.
On the other hand, there is a bunch of research about how to build video servers, and some of it does use clusters of relatively cheap boxes. Forgive me for not having a URL to it.
The web page describing this NIC is really unimpressive. What I'd really like to know is:
How many simultaneous connections can this thing support, and is it slower when multiple connections are used? How's the performance when one of these cards is hooked up to a machine with a standard software TCP/IP?
Putting TCP/IP in hardware is nice, I guess, but then nasty real-world issues crop up. What happens if there's a bug in the implementation? There's also the nasty challenge of writing a driver for a card like this, but I won't claim that's a defect of this NIC design.
I read the book. I found it entertaining. Beyond that, I wouldn't say it was worthwhile. After all, it goes on for 800 masturbatory pages to reach the completely generic conclusion that AI is possible.
A lot of people seem to be so taken with this book. I can only conclude that the enormous satisfaction of solving the little puzzles and games sprinkled through the text outweighs the enormous letdown of the text's message.
Since we've already got moderators, why is it beneficial to do auto-scoring of posts based on past behavior? It's not going to save the moderators any trouble since they still have to check posts to make sure that the auto-scoring is correct.
However, there is one thing that auto-scoring does do: it labels people as useless or intelligent. It won't take long for an upper class of people with high default scores to form.
With Katz, it's always geek-this and geek-that. I don't think Futurama is a parody of computers and technology. It's much more Sci-Fi-oriented. The distinction is slim, but it's real: in the first episode, the world was destroyed by aliens, not by the Y2K bug.
Slashdot Mirror
Sure, these .tv guys are turkeys, but I think it's not unreasonable to charge premiums for domain names that are more "in demand." Let's let the market decide how much they're worth.
Seems like this proposal is relying on the web site being able to change its address and routing info faster than the attackers are able to react. This is a losing proposition.
It's apparent that a distributed attack requires a distributed solution. But as many posters have pointed out, that usually requires a lot of cooperation from a lot of ISPs. But what about using the caching infrastructure? Assume the web site under attack actually allowed its content to be cached (pretty big assumption due to cookies and banner ads). Now if the site is attacked, all the content that's currently cached could still be served from caches. This doesn't cover dynamically generated content or content that hasn't been cached. So this is a partial solution, but it's deployable right now, today, this instant.
Wow, 10,000 stories. I wonder how many unique stories were posted? Much fewer than 10K I bet.
Sorry if this has already been brought up - I haven't been reading every comment in the moderation discussions.
/. moderation system though.
The question: does MM create a feedback loop? The people that moderation rewards get to reward the moderators, right? It certainly seems like it's got the potential to be unstable.
What if, instead of MM, we just defined a "bad" moderator to be someone whose use of points is contrary to other moderators'? For example, if one moderator loved an article and many others hated it, then the moderator who loved the article is doing a bad job. I'm not sure how well this would fit into the
It's not light (6.7 pounds)
It's not fast (300 MHz G3)
Its display isn't great (800x600 max)
Its price is sorta ok ($1600?)
But Apple put it in a brightly colored case, so people will want to buy it. I'm starting to think that Apple's strongest feature is its slick marketing.
Though I have to admit, I like the handle.
To make a distributed solution work, the system would need to know the relative popularity of each movie: that way, the highly demanded movies could be placed on multiple servers since a single $2k PC won't be able to serve the movie to many clients at once (no, "multicast" won't solve this problem). That's why a central server makes so much sense: if the server is up to the challenge, then you don't have the headache of trying to do good load balancing.
On the other hand, there is a bunch of research about how to build video servers, and some of it does use clusters of relatively cheap boxes. Forgive me for not having a URL to it.
So let me get this straight: the term "GNU/Linux" is offensive to people, but everyone's happy when Redhat calls their distribution "Redhat Linux"?
OK.
The web page describing this NIC is really unimpressive. What I'd really like to know is:
How many simultaneous connections can this thing support, and is it slower when multiple connections are used? How's the performance when one of these cards is hooked up to a machine with a standard software TCP/IP?
Putting TCP/IP in hardware is nice, I guess, but then nasty real-world issues crop up. What happens if there's a bug in the implementation? There's also the nasty challenge of writing a driver for a card like this, but I won't claim that's a defect of this NIC design.
I read the book. I found it entertaining. Beyond that, I wouldn't say it was worthwhile. After all, it goes on for 800 masturbatory pages to reach the completely generic conclusion that AI is possible.
A lot of people seem to be so taken with this book. I can only conclude that the enormous satisfaction of solving the little puzzles and games sprinkled through the text outweighs the enormous letdown of the text's message.
Since we've already got moderators, why is it beneficial to do auto-scoring of posts based on past behavior? It's not going to save the moderators any trouble since they still have to check posts to make sure that the auto-scoring is correct.
However, there is one thing that auto-scoring does do: it labels people as useless or intelligent. It won't take long for an upper class of people with high default scores to form.
With Katz, it's always geek-this and geek-that. I don't think Futurama is a parody of computers and technology. It's much more Sci-Fi-oriented. The distinction is slim, but it's real: in the first episode, the world was destroyed by aliens, not by the Y2K bug.