At last someone gets it. The guy might be a rich irresponsible asshole, but he still had the balls to do it. To gazillions of car nuts he's now immortal. That's more fame than most of us will ever have.
Yes, there was exactly the same problem when the walkman was first popular in the early 80s (I'm showing my age here): lots of sensationalist hype about hearing loss, which didn't pan out into much reality.
Last I checked, in the US your credit card liability is limited by law to $50. Most banks, in the interest of increasing customer confidence, have a $0 policy. (which is really rather surprising considering banks never allow themselves to be on the losing end)
They're not really on the losing end. Their credit card fraud losses are covered many times over by the extortionate interest they charge the tens of millions of poor fools who run thousands of dollars of debt month after month.
A good source of referrals is someone who has to deal with the mistakes. My opthamologist, for example, loathes what the lasik guys have to done to the reputation of his profession - he calls them "the used car salesmen of opthalmology" - and he is the first person I would ask for a referral to a good lasik surgeon.
The American Academy of Opthamology has a searchable database covering numerous countries. I think a cataract specialist will be the best person to ask, as they deal specifically with lens disease.
What happens when it gets hacked? Its inevitable that some "haxor" is gonna try to DoS the whole thing
That will be difficult. As another poster pointed out, this will be a private IP network disconnected from the Internet.
or maby even make the "phone virus". Gotta take the good with the bad.
What would the virus infect? Aside from some network management workstations (typically running some flavour of *NIX), almost all of the equipment on the network will be either dedicated VOIP hardware or routers, both of which are designed not to have users' processes running on them, and so are very robust to viruses.
How can a protocol be rated faster than DSL? Shouldn't the rating be against another protocol?
DSL is a layer 1 & 2 protocol. TCP and BIC-TCP are layer 4 protocols. So, you're correct that they're not comparing like with like.
Also, TCP will also reach similar speeds under the right conditions. BIC-TCP will just reach those speeds with less ramp-up time, and over a wider range of conditions. One of those conditions is that it is not running over a DSL line, or a T3 or an OC3 or OC48 or anything that the average internet user will see in the next decade or two. So, the article is wildly over-hyping a minor protocol tweak that is irrelevant to almost all internet users.
NAT can be seen as increasing the IP address space by 16 bits (the size of the TCP or UDP port space: 17 bits if both are used). So, it effectively extends the IPv4 address space to 48 bits, which is enough for a long while. (It's also a hack, but so was the telephone, IMHO.)
If that space runs out, there's always the possibility of making an extension to the TCP port space to allow 32 bits (or more): deploying a nice, backwards-compatible TCP extension on the Internet would probably be easier than rolling out all of IPv6. It's not clear, howevever, whether that would actually stop the IPv4 space running out: I think that's more an issue of the politics of handing out IPv4 address space.
But does anyone know why they chose macadamia nuts? Seems a very strange choice.
Macadamia trees are a native plant in that part of Australia, and grow very well. There's a large macadamia nut industry there, which was throwing out huge amounts of nutshell.
The really bizarre part that has not been explained is that the frequency deviated.
I'll try for an even simpler examplation of this one.
If you suddenly increase the amount of power drawn from a generator, it will try to slow down. Usually, the other generators in the system will stop this, by generating more power, and nothing much will happen. On the other hand, if the generator does slow down (by pulling the whole grid with it), then the grid frequency will drop.
The opposite will happen if you suddenly decrease the amount of power drawn from a generator.
That sort of thing was happening as the grid shook itself to bits, and the frequency would have been wavering all over the place.
The other providers obviously use packet switching as evidenced by the intermittent delays as much as a couple of seconds.
Packet switching allows this sort of thing, but fortunately doesn't require it. A well-engineered packet network, with a known traffic pattern, will rarely drop or excessively delay a packet, for much the same reasons that a correctly engineered Frame Relay network will rarely do either. AT&T have a pretty good record at network engineering.
I would seriously hope to see some more advanced algorithms or better yet, more bandwidth installed, before these systems become more heavily adopted.
More efficient algortithms are not needed. It comes down to economics, more than anything else. Yes, VoIP algorithims are inefficient. They use more bandwidth than PCM voice. The extra bandwidth used by VoIP will cost money (and other resources). Call this the "excess bandwidth cost" of VoIP.
You are also correct that more bandwidth is needed, but that is not driven by voice requirements: it's driven mostly by broadband data.
As the proportion of voice traffic in a network decreases (it's already less than 12.5% of AT&T's traffic according to the article), then the excess bandwith cost of VoIP decreases as a proportion of the network cost.
AT&T will have calculated that sometime shortly after 2004, it will be cheaper to pay the excess bandwidth cost of VoIP, than to maintain a separate voice network. Cisco sum up this economic situation as "the voice rides for free" on the IP network. That's a marketing over-simplification, yet captures an essential truth.
One of the most sad aspects of such behavior is POSSIBLE in this modern world. Imagine 1000 years ago, before modern weapons began to surface... A blabbering idiot, desperate for social interaction. Can you imagine such a warrior behaving in that fashion? Would such a people survive if their only defense were chatterboxes? I think not.
I think this falls into the (very common) trap of glorifying life in the past. 1000 years ago, most people in western culture weren't warriors. They were poor peasants scratching fairly short and boring lives. They would be probably be illiterate, and probably would never go more than 20 miles or so from the place where they were born. They would have been happy for the distraction of the occasional conversation full of small talk and simple gossip, in order to escape the dullness of their everyday lives.
In that sense, things certainly haven't become worse.
Slashdot Mirror
Yes, there was exactly the same problem when the walkman was first popular in the early 80s (I'm showing my age here): lots of sensationalist hype about hearing loss, which didn't pan out into much reality.
A good source of referrals is someone who has to deal with the mistakes. My opthamologist, for example, loathes what the lasik guys have to done to the reputation of his profession - he calls them "the used car salesmen of opthalmology" - and he is the first person I would ask for a referral to a good lasik surgeon. The American Academy of Opthamology has a searchable database covering numerous countries. I think a cataract specialist will be the best person to ask, as they deal specifically with lens disease.
Ants? I'm more worried about the humans.
Also, TCP will also reach similar speeds under the right conditions. BIC-TCP will just reach those speeds with less ramp-up time, and over a wider range of conditions. One of those conditions is that it is not running over a DSL line, or a T3 or an OC3 or OC48 or anything that the average internet user will see in the next decade or two. So, the article is wildly over-hyping a minor protocol tweak that is irrelevant to almost all internet users.
NAT can be seen as increasing the IP address space by 16 bits (the size of the TCP or UDP port space: 17 bits if both are used). So, it effectively extends the IPv4 address space to 48 bits, which is enough for a long while. (It's also a hack, but so was the telephone, IMHO.) If that space runs out, there's always the possibility of making an extension to the TCP port space to allow 32 bits (or more): deploying a nice, backwards-compatible TCP extension on the Internet would probably be easier than rolling out all of IPv6. It's not clear, howevever, whether that would actually stop the IPv4 space running out: I think that's more an issue of the politics of handing out IPv4 address space.
I'll try for an even simpler examplation of this one.
If you suddenly increase the amount of power drawn from a generator, it will try to slow down. Usually, the other generators in the system will stop this, by generating more power, and nothing much will happen. On the other hand, if the generator does slow down (by pulling the whole grid with it), then the grid frequency will drop. The opposite will happen if you suddenly decrease the amount of power drawn from a generator.
That sort of thing was happening as the grid shook itself to bits, and the frequency would have been wavering all over the place.