Some people in networking research believe that the problem with Multicast (and even QoS) has nothing to do with scalability, but more with economics. Although in this case, ISPs would reduce traffic going through their network by enabling multicast, there is no popular method of accounting for internal traffic when multicast is enabled on all routers. For most ISPs this is unacceptable, since large customers are billed based on the amount of traffic sent. Since there's no economic model developed for multicast-traffic, ISPs would rather throttle back BitTorrent than enable multicast. Someone please correct me if I'm mistaken on any of these points.
Most networking researchers seem to believe multicast is technologically feasible and helpful, which is why a lot of Internet architecture research seems to provide methods for multicast, even though hardly anybody uses it today.
The patent infringement I believe is that clients register their IP addresses with a database so other clients can learn of their 'online status' and associated IP address in order to communicate.
This seems quite similar to the 'tracker' functionality of BitTorrent, which is essentially the same thing. I imagine there exists some sort of prior art from the 80s, but I can't think of any at the moment...
I have 'OneLink' DSL with Speakeasy, and the advantage is that you get pretty much a 'guaranteed' service.
I was about 6000 feet from my CO previously, and despite that, I could not maintain 1.5/768 DSL due to a crappy line installed by PG&E. Unfortunately, because there was no onelink service, I couldn't do anything if PG&E didn't want to help me out, which they didn't.
However, once I got OneLink, PG&E was now forced to lay a new line down or fix the line so that it was up to spec. Now I'm able to get 6.0/768 stable. Has not gone down for more than 10 minutes in the past 1.5 years (other than maintenance).
Sure, open algorithms such as RSA are typically better because of their scrutiny by the the world's best cryptographers. However, a bunch of the NSA's algorithms are good BECAUSE they are secret (to their best of their ability). Yes, it probably won't hold the test of time as well, but that's not to say they don't work.
I guess it's kind of analagous to 'security through obscurity.' Yeah, it's not the best way to do things, but it sometimes works well enough for some applications.
I wonder how hard it would be to just put water/your own mix in it instead.
Assuming they are 'smart,' they would follow the lead of the inkjet companies like you said, and use a proprietary solution cartridge system that forces you to use their solution.
If they do the latter, then I won't even consider this product. I really hope they don't screw this up:)
I just learned about Distributed Hash Tables this past semester and thought they were really cool. On the bittorrent page linked in the blurb, it mentions the use of a DHT in order to do the join/lookup required for locating peers.
If you are interested in how it works, you can check out http://en.wikipedia.org/wiki/Distributed_hash_tabl e for more info on them and links to example DHT implementations (such as CAN, Chord, and Kademlia).
I just learned about Distributed Hash Tables this past semester and thought they were really cool. On the page it mentions the use of aDHT in order to do the join/lookup required for locating peers.
The pre-load most likely does not include core game files like the executable, just the static libraries of models, maps, textures, etc that will not change until the game's release. When the game is 'released,' Steam will probably just download the rest of core game files (which should be a relatively small download), and then will launch the game.
It seems like the only danger of breaking the encryption is getting access to files that may spoil the plot, but I've heard there's already a file out there that does that.
The article you linked to mentioned that you could get out of Episodic SP by turning it into a lucid dream. You mentioned that you tried to move a finger or similar to break the paralysis.
Ironically, in all my episodes of SP, the way I was able to break out was by imagining someone taking an axe and cutting off my head. It's successfully worked more than once, although I really wish I could think of a better, less gruesome way to wake up.
And clicking a stopwatch, and measuring how long launching a program takes, or how long a reboot lasts isn't that much of a "benchmark".
According to Patterson and Hennessy's Computer Organization and Design textbook, "Time is the measure of computer performance: the computer that performs the same amount of work in the least time is the fastest." (Page 58, Second Edition).
Of course, just because it's the fastest doesn't mean it is the one YOU should buy, but I think the standard of 'performance' is execution time.
Financial aid helps those who can't normally afford to go to these schools, sometimes covering up to the entire amount of tuition.
Although living in these areas (Berkeley, Palo Alto/Stanford), is hard enough when rent prices are generally higher than other places (I can't speak for Boston).
Isn't it possible that these graphics card companies can be doing similar operations to other games and their popular benchmarking utilities? For example, I could make my drivers optimized for demo001.dm3 for Quake3, and then some of the reviews that don't use their own demos for testing will be VERY generous to me.
This is probably unrealistic (if you ran the demo with a wider POV, you could probably detect errors easily, at least the way I've heard Nvidia did it with 3dmark), but I'm still going to remain paranoid:]
There are magnetic locks that exist already that let you walk by the sensor with the magnetic 'key' somewhere in proximity, and it will unlock automatically. Granted, it's not as secure as having something embedded into your body (albeit a little less scary), but it's as secure as having a traditional key-lock system, without the nuisance of having to pull out the key and insert it into the door lock.
Would also make it easier to unlock the front door when you try to find the keyhole in pitch black situations:]
Quick note: It has been shown that the 'peak values' of FPS for you to achieve the greatest trajectory are those that, when divided by 1000, will have a whole number.
The most ideal value would be 1000 FPS, since it divides into 1000 perfectly, and once. Other values are 500, 333, 250, 200, 166, 142, 125...
You'll notice that if you do com_maxfps of 350, drawfps in quake3 will 'cap it' at 333, and so on for any of these numbers. Thus, the closer you get to these actual values, the better trajectory you will have. 125 is just a popular one that is used since most graphics cards during this discovery could get 125fps consistently. I believe 333fps provides a better trajectory than say, 125fps, but there are webpages out there showing the actual math behind all of this which will prove this.
It seems to me there is a large disparity in the kind of development between the two different kind of AI investigations. Game AI, although more about the 'result' as stated in the article, has to be based upon the research done in the academia. While it says academia could learn a thing or two by understanding what GAMES are using from AI, they can better focus and optimize and even research better platforms for the games to use (This is just paraphrasing some of what the article might have said, including my own interpretion, if at all accurate).
What I've noticed is, since the human brain knowledge IS 85% speculation, we often use AI strategies to fake knowledge. I mean for FPS bots, they have used paths and nodes to simulate familiarity and some order for the bot, but still that gets too much into a pattern which is not necessarily very human.
I guess my main concern is knowing exactly how far Game AI trails the progress of Academia AI, and when, if ever, the two will progress together.
Slashdot Mirror
http://www.sigcomm.org/HotNets-IV/papers/ballani.p df
Math may be slightly wrong on my part but...
2.5in is about 6.35 cm, so it's actually closer to 32cm perimeter, which roughly is about 4.8 km/m or 80m/s.
But also keep in mind that the outer edge of a 3.5in plate spinning at 7200rpm is about 4.4km/m or 73m/s, so it's not that surprising.
In fact, the 3.5in 10000RPM raptors edge spins at about 110m/s.
A lecture given by Von Ahn on Human Computation is available here: http://video.google.com/videoplay?docid=-824646398 0976635143
Some people in networking research believe that the problem with Multicast (and even QoS) has nothing to do with scalability, but more with economics. Although in this case, ISPs would reduce traffic going through their network by enabling multicast, there is no popular method of accounting for internal traffic when multicast is enabled on all routers. For most ISPs this is unacceptable, since large customers are billed based on the amount of traffic sent. Since there's no economic model developed for multicast-traffic, ISPs would rather throttle back BitTorrent than enable multicast. Someone please correct me if I'm mistaken on any of these points.
Most networking researchers seem to believe multicast is technologically feasible and helpful, which is why a lot of Internet architecture research seems to provide methods for multicast, even though hardly anybody uses it today.
ICQ was created in 1996, and this patent was *filed* in 1995 and granted in 2000. Does this mean ICQ is still prior art?
Basically, does prior art have to be prior to the filing or the granting?
The patent infringement I believe is that clients register their IP addresses with a database so other clients can learn of their 'online status' and associated IP address in order to communicate.
This seems quite similar to the 'tracker' functionality of BitTorrent, which is essentially the same thing. I imagine there exists some sort of prior art from the 80s, but I can't think of any at the moment...
You're only a happy engineer since your UID is neat, anyway.
Yikes, I meant SBC, not PG&E.
:P
I used to be pissed off at my SBC bill, but I no longer have one. Now I'm pissed off at my PG&E bill, so that's probably why I made the mistake
I have 'OneLink' DSL with Speakeasy, and the advantage is that you get pretty much a 'guaranteed' service.
I was about 6000 feet from my CO previously, and despite that, I could not maintain 1.5/768 DSL due to a crappy line installed by PG&E. Unfortunately, because there was no onelink service, I couldn't do anything if PG&E didn't want to help me out, which they didn't.
However, once I got OneLink, PG&E was now forced to lay a new line down or fix the line so that it was up to spec. Now I'm able to get 6.0/768 stable. Has not gone down for more than 10 minutes in the past 1.5 years (other than maintenance).
It's in the comments of the blog post:
n ing-milestones/#comment-28122
http://www.problogger.net/archives/2005/07/12/ear
Isn't this simply branch prediction?
http://en.wikipedia.org/wiki/Branch_predictor
Hasn't this been around for a while?
Sure, open algorithms such as RSA are typically better because of their scrutiny by the the world's best cryptographers. However, a bunch of the NSA's algorithms are good BECAUSE they are secret (to their best of their ability). Yes, it probably won't hold the test of time as well, but that's not to say they don't work.
I guess it's kind of analagous to 'security through obscurity.' Yeah, it's not the best way to do things, but it sometimes works well enough for some applications.
I wonder how hard it would be to just put water/your own mix in it instead.
:)
Assuming they are 'smart,' they would follow the lead of the inkjet companies like you said, and use a proprietary solution cartridge system that forces you to use their solution.
If they do the latter, then I won't even consider this product. I really hope they don't screw this up
I just learned about Distributed Hash Tables this past semester and thought they were really cool. On the bittorrent page linked in the blurb, it mentions the use of a DHT in order to do the join/lookup required for locating peers.
l e for more info on them and links to example DHT implementations (such as CAN, Chord, and Kademlia).
If you are interested in how it works, you can check out http://en.wikipedia.org/wiki/Distributed_hash_tab
I just learned about Distributed Hash Tables this past semester and thought they were really cool. On the page it mentions the use of aDHT in order to do the join/lookup required for locating peers.
l e/ for more info on them and links to example DHT implementations (such as CAN, Chord, and Kademlia).
If you are interested in how it works, you can check out http://en.wikipedia.org/wiki/Distributed_hash_tab
"The trick here is that no one is in better position to do such a thing than the company that owns most of the network infrastructure."
Couldn't you replace 'network infrastructure' with 'OS Market Share' and say the same thing about Microsoft?
Isn't this what everybody complains about? If not, I'd love to hear the difference
Apple to Guy: "Your domain, left testicle, and right testicle. Choose any two."
The pre-load most likely does not include core game files like the executable, just the static libraries of models, maps, textures, etc that will not change until the game's release. When the game is 'released,' Steam will probably just download the rest of core game files (which should be a relatively small download), and then will launch the game.
It seems like the only danger of breaking the encryption is getting access to files that may spoil the plot, but I've heard there's already a file out there that does that.
The article you linked to mentioned that you could get out of Episodic SP by turning it into a lucid dream. You mentioned that you tried to move a finger or similar to break the paralysis.
Ironically, in all my episodes of SP, the way I was able to break out was by imagining someone taking an axe and cutting off my head. It's successfully worked more than once, although I really wish I could think of a better, less gruesome way to wake up.
And clicking a stopwatch, and measuring how long launching a program takes, or how long a reboot lasts isn't that much of a "benchmark".
According to Patterson and Hennessy's Computer Organization and Design textbook, "Time is the measure of computer performance: the computer that performs the same amount of work in the least time is the fastest." (Page 58, Second Edition).
Of course, just because it's the fastest doesn't mean it is the one YOU should buy, but I think the standard of 'performance' is execution time.
Financial aid helps those who can't normally afford to go to these schools, sometimes covering up to the entire amount of tuition.
Although living in these areas (Berkeley, Palo Alto/Stanford), is hard enough when rent prices are generally higher than other places (I can't speak for Boston).
Isn't it possible that these graphics card companies can be doing similar operations to other games and their popular benchmarking utilities? For example, I could make my drivers optimized for demo001.dm3 for Quake3, and then some of the reviews that don't use their own demos for testing will be VERY generous to me.
:]
This is probably unrealistic (if you ran the demo with a wider POV, you could probably detect errors easily, at least the way I've heard Nvidia did it with 3dmark), but I'm still going to remain paranoid
There are magnetic locks that exist already that let you walk by the sensor with the magnetic 'key' somewhere in proximity, and it will unlock automatically. Granted, it's not as secure as having something embedded into your body (albeit a little less scary), but it's as secure as having a traditional key-lock system, without the nuisance of having to pull out the key and insert it into the door lock.
:]
Would also make it easier to unlock the front door when you try to find the keyhole in pitch black situations
Quick note: It has been shown that the 'peak values' of FPS for you to achieve the greatest trajectory are those that, when divided by 1000, will have a whole number.
...
The most ideal value would be 1000 FPS, since it divides into 1000 perfectly, and once. Other values are 500, 333, 250, 200, 166, 142, 125
You'll notice that if you do com_maxfps of 350, drawfps in quake3 will 'cap it' at 333, and so on for any of these numbers. Thus, the closer you get to these actual values, the better trajectory you will have. 125 is just a popular one that is used since most graphics cards during this discovery could get 125fps consistently. I believe 333fps provides a better trajectory than say, 125fps, but there are webpages out there showing the actual math behind all of this which will prove this.
It seems to me there is a large disparity in the kind of development between the two different kind of AI investigations. Game AI, although more about the 'result' as stated in the article, has to be based upon the research done in the academia. While it says academia could learn a thing or two by understanding what GAMES are using from AI, they can better focus and optimize and even research better platforms for the games to use (This is just paraphrasing some of what the article might have said, including my own interpretion, if at all accurate).
What I've noticed is, since the human brain knowledge IS 85% speculation, we often use AI strategies to fake knowledge. I mean for FPS bots, they have used paths and nodes to simulate familiarity and some order for the bot, but still that gets too much into a pattern which is not necessarily very human.
I guess my main concern is knowing exactly how far Game AI trails the progress of Academia AI, and when, if ever, the two will progress together.