Re:Cute, but is it secure?
on
MIT Roofnet
·
· Score: 1
They can set up a VPN and run it on top, and forget about man in the middle attacks.
Re:So now you can DOS this network ...
on
MIT Roofnet
·
· Score: 1
Nah, the students just disconnect the 2.45Ghz WiFi, and connect up everyone's favourite other 2.45 Ghz appliance, the Microwave oven, and call it dinner.
Poor starving students.:-)
Re:The beginning of a true Mesh network?
on
MIT Roofnet
·
· Score: 2, Informative
Linksys WAP11's and similar point-to-point wireless devices have latency up into 30-100 ms I think...
Whenever I've measured latency in WiFi it has typically been under 0.5ms; latency can be much worse with poor reception due to retries. I can't comment on that particularly product since I haven't used it but I would be very surprised if it was that high.
John Carmack recently investigated this question. Actually if John's trajectory screws up, he's not going anywhere real fast- air drag really, really limits the maximum sideways speed- I forget the range off hand but it was really low like 10km or something. His rocket should only manage to get to 100km because he leaves the atmosphere ASAP. But by the time he's left it he's already going straight up but he has little fuel left to get any significant sideways distance; it's really finely balanced.
I'm definitely not buying a house in its flight-path.
I don't think that FAA will let John launch if there is a house in the flight-path. Besides, John will be launching pretty much vertically- he's not going for orbit (which means going sideways very fast); he's only going for 100km (which means going straight-up very fast).
Uh, no, Ethernet has not traditionally suffered from high latencies;-)
I typically find latencies between nodes below 1ms; for instance 0.5 ms. That's fairly typical; indeed I usually find that most of the time is taken with lightspeed delays; and in congestion. Now, as the bandwidth of the link goes up, congestion goes down (everything else being equal). Lightspeed delays we can't do anything about at present though.
In fact I was puzzled why you were attacking ethernet like this- in my experience ethernet has entirely reasonable latency (except when congested).
Of course perhaps you are talking about the 'slow start' and other congestion avoidance protocols in ethernet. These mean that if you ask for a lot of bandwidth suddenly, it can take a considerable time to build up. Of course, the ethernet protocols probably aren't well tuned for terabit rates, but they certainly can be.
It wasn't in the ISS orbit because it wasn't going there....
You've missed the point; Columbia wasn't going there partly because it couldn't. Columbia didn't have the delta-v to reach the ISS even from the ground with that version of the tank; but the fact that it didn't have the new tank meant that it didn't have the new foam formulation either.
and it appears that in a way that the switch to a more environment-friendly external tank foam material in 1997
One of the reasons that Columbia wasn't in the same orbit as the ISS was that Columbia was using one of the older, heavier tanks, and as I understand it, those use the older type of foam.
Everyone will be downloading different stuff. Bittorrent only works when everyone is trying to get the same content at the same time.
Probably not entirely true. Some things will be very popular, and in that case you may be able to download it from someone else who already got it from the BBC. Think a little more like Kazaa, only more legal; I'm sure BitTorrent could be rejiggered appropriately to make it work better for this.
a) the person that discovers it will become very wealthy for a while
b) everyone will be playing the game and not working/farming/eating/reproducing and the human race will die out!
So if anyone knows the answer- do your duty to mankind and keep your mouth shut! Being very rich doesn't help if there's no food to buy, and no babes/bros to impress!
The answer to how close you can put the channels is known as Shannons Theorem. This basically says that the radio channels can't be put closer together than a certain amount, since each channel needs a certain bandwidth.
The amount of bandwidth basically depends on you much noise there is around- you can pack much more data into a narrow channel if the channel has hardly any noise, whereas if the channel is very noise you'll need a wide channel to send the same data.
Also in practice you need a gap between neighbouring channels- the receivers need to filter out the other channels and they don't do this perfectly (although the better the receivers are, the closer you can pack them in.)
However Shannons law only deals with broadcast communications where everyone can 'hear' everyone else equally, if you use directional antennas then it doesn't necessarily apply- two communications could use the very same channels.
What about non-free material goods? Does that also create a "...predatory social system that keeps people in a state of domination and division."?
You bet it does, atleast sometimes.
Does RMS even understand physics?
I expect so, although I don't know off hand his education.
It takes "work" to change random states of bits into useful tools and information. Work doesn't come free.
It does though. If I distribute source code and somebody has a problem and fixes it; it costs vanishingly little on average (divided by the users), but everyone gains, including the person that fixed it. It's a positive sum game; rather than zero sum game.
Working a material good out of rock, wood, sand, etc, and working bits out of random noise, turns out to be equivalent.
No, because the duplication costs of software is almost zero.
People who do "work" probably are more deserving of the prizes. The betterment of one's self should always be our higher goal.
Really? How about betterment of family, society, technology. In the long run we are all dead, but some of the improvements we make can live on.
Be contructive, not destructive. Lend a helping hand to those who are trying, but don't offer any favors to those who are not.
Everyone helps in different ways. The trick is to dump those that don't help out.
In the end, everyone gets their just rewards.
Only if there is policing. Stallmans genius is to come up with a license that legally allows policing.
Interesting bits on the neurons and depression... if you happen to know of any texts geared toward the layperson on this, post them if you would.
I think at this stage it is just a theory; I'm not sure exactly how much experimental support it has; but there is some, and it explains things that otherwise aren't understood- for example why it takes 5-6 weeks for depression to lift (it probably takes that long for the neurons to install themselves); also why exercise seems to help (exercise seems to increase neuronal production too.) But I don't think it is the dominant theory.
Even if you can solve the problem of physical decay, how long do the neurons in the brain last? New neurons cannot be created, only new connections can be made...
That's actually been shown to be false, although it used to be believed. Neurons are born throughout life, particularly in certain parts of the brain- there are stem cells in the human brain.
Indeed, indications are that depression is caused by insufficient neurons being produced; antidepressants seems to increase survival of the new nerve cells, as well as raising serotonin levels.
What are you going to do with all those years? Can you seriously imagine what it would be like to work for 200 years, as opposed to 65? That's more tha 3 times the current retirement age!
Well, if you can save up enough money you can live off the interest indefinitely. About a million bucks is in the ballpark.
Wireless networks have greater latencies than wired networks. Its just a fact.
No; or not significantly. Last time I measured it on my 802.11b network it was well under 2ms. Sure, if you have interference, then you'll be hitting the retries, in which case the average latency will go up; but under good conditions, it's got negligable latency. (Some newbies have suggested that WiFi takes longer due to propogation through the air- actually the speed of radio waves in air is almost twice that of ethernet signals copper.)
Slashdot Mirror
They can set up a VPN and run it on top, and forget about man in the middle attacks.
Poor starving students. :-)
Whenever I've measured latency in WiFi it has typically been under 0.5ms; latency can be much worse with poor reception due to retries. I can't comment on that particularly product since I haven't used it but I would be very surprised if it was that high.
I don't have experience with this
So where did you get the numbers from?
John Carmack recently investigated this question. Actually if John's trajectory screws up, he's not going anywhere real fast- air drag really, really limits the maximum sideways speed- I forget the range off hand but it was really low like 10km or something. His rocket should only manage to get to 100km because he leaves the atmosphere ASAP. But by the time he's left it he's already going straight up but he has little fuel left to get any significant sideways distance; it's really finely balanced.
I don't think that FAA will let John launch if there is a house in the flight-path. Besides, John will be launching pretty much vertically- he's not going for orbit (which means going sideways very fast); he's only going for 100km (which means going straight-up very fast).
I typically find latencies between nodes below 1ms; for instance 0.5 ms. That's fairly typical; indeed I usually find that most of the time is taken with lightspeed delays; and in congestion. Now, as the bandwidth of the link goes up, congestion goes down (everything else being equal). Lightspeed delays we can't do anything about at present though.
In fact I was puzzled why you were attacking ethernet like this- in my experience ethernet has entirely reasonable latency (except when congested).
Of course perhaps you are talking about the 'slow start' and other congestion avoidance protocols in ethernet. These mean that if you ask for a lot of bandwidth suddenly, it can take a considerable time to build up. Of course, the ethernet protocols probably aren't well tuned for terabit rates, but they certainly can be.
What happens when the spammers start using worms and viruses to create open relays on people you trust?
You see, with Linux we just rip off other people's work (hi SCO).
You've missed the point; Columbia wasn't going there partly because it couldn't. Columbia didn't have the delta-v to reach the ISS even from the ground with that version of the tank; but the fact that it didn't have the new tank meant that it didn't have the new foam formulation either.
One of the reasons that Columbia wasn't in the same orbit as the ISS was that Columbia was using one of the older, heavier tanks, and as I understand it, those use the older type of foam.
The BBC have the resources and ability to make the use of Bittorrent mandatory...
Probably not entirely true. Some things will be very popular, and in that case you may be able to download it from someone else who already got it from the BBC. Think a little more like Kazaa, only more legal; I'm sure BitTorrent could be rejiggered appropriately to make it work better for this.
a) the person that discovers it will become very wealthy for a while
b) everyone will be playing the game and not working/farming/eating/reproducing and the human race will die out!
So if anyone knows the answer- do your duty to mankind and keep your mouth shut! Being very rich doesn't help if there's no food to buy, and no babes/bros to impress!
In other words: they've deleted your Linux root partition ;-)
The amount of bandwidth basically depends on you much noise there is around- you can pack much more data into a narrow channel if the channel has hardly any noise, whereas if the channel is very noise you'll need a wide channel to send the same data.
Also in practice you need a gap between neighbouring channels- the receivers need to filter out the other channels and they don't do this perfectly (although the better the receivers are, the closer you can pack them in.)
However Shannons law only deals with broadcast communications where everyone can 'hear' everyone else equally, if you use directional antennas then it doesn't necessarily apply- two communications could use the very same channels.
Yes, the fact that they would be making more money must be the reason that the companies ran to number portability so quickly ;-)
How many years were they blocking it for again- and it turns out they make more money from it?? Ten? Five?
You bet it does, atleast sometimes.
Does RMS even understand physics?
I expect so, although I don't know off hand his education.
It takes "work" to change random states of bits into useful tools and information. Work doesn't come free.
It does though. If I distribute source code and somebody has a problem and fixes it; it costs vanishingly little on average (divided by the users), but everyone gains, including the person that fixed it. It's a positive sum game; rather than zero sum game.
Working a material good out of rock, wood, sand, etc, and working bits out of random noise, turns out to be equivalent.
No, because the duplication costs of software is almost zero.
People who do "work" probably are more deserving of the prizes. The betterment of one's self should always be our higher goal.
Really? How about betterment of family, society, technology. In the long run we are all dead, but some of the improvements we make can live on.
Be contructive, not destructive. Lend a helping hand to those who are trying, but don't offer any favors to those who are not.
Everyone helps in different ways. The trick is to dump those that don't help out.
In the end, everyone gets their just rewards.
Only if there is policing. Stallmans genius is to come up with a license that legally allows policing.
Just my 2 cents.
Whoa. I'm rich! ;-)
I'll stick my neck out a bit further- definite explosions. The whole point is that you make the fuel detonate. That's how it works ;-)
If you're dumb enough to lend your money to the bank for 0.25% APR then you probably aren't smart enough to make a million in the first place.
I think at this stage it is just a theory; I'm not sure exactly how much experimental support it has; but there is some, and it explains things that otherwise aren't understood- for example why it takes 5-6 weeks for depression to lift (it probably takes that long for the neurons to install themselves); also why exercise seems to help (exercise seems to increase neuronal production too.) But I don't think it is the dominant theory.
Everyone who has enough money; by definition. Note that if too many people try to do this you get inflation which self regulates.
That's actually been shown to be false, although it used to be believed. Neurons are born throughout life, particularly in certain parts of the brain- there are stem cells in the human brain.
Indeed, indications are that depression is caused by insufficient neurons being produced; antidepressants seems to increase survival of the new nerve cells, as well as raising serotonin levels.
What are you going to do with all those years? Can you seriously imagine what it would be like to work for 200 years, as opposed to 65? That's more tha 3 times the current retirement age!
Well, if you can save up enough money you can live off the interest indefinitely. About a million bucks is in the ballpark.
Anyone who's ever gone to a bar and not had more than one drink knows this.
You're obviously a rotten drunk then ;-)
No; or not significantly. Last time I measured it on my 802.11b network it was well under 2ms. Sure, if you have interference, then you'll be hitting the retries, in which case the average latency will go up; but under good conditions, it's got negligable latency. (Some newbies have suggested that WiFi takes longer due to propogation through the air- actually the speed of radio waves in air is almost twice that of ethernet signals copper.)
Mixing your metric with your imperial? You should work for NASA ;-)