I agree. He should have used the *full* name of another one of the testers.
Microsoft would be happy cos they'd have somebody to blame. He'd be happy because he doesn't get blamed. The poor sap that got blamed would be very unhappy.
- but that's just pure selfishness, since, as they say in Star Trek- 'the needs of the many outweigh the needs of the few or the one.'
On the arocket email group the consensus seems to be that it looks like fuel slosh being driven by the control system moving the nozzle in a circular mode. Eventually the magnitude of the control inputs seems to have created a roll angle, and that's what killed the telemetry and the engine apparently is likely to have shut down shortly afterwards.
Did anyone else notice the bump the Kestrel engine took during stage separation? On the 40MB video from SpaceX, it happend at 3:28 in or at T+00:02:52 on the screen clock. Maybe this is normal for the engine, but it was rather odd looking to me.
So far as is known, it didn't materially affect anything. The nozzle is made of Niobium which is quite malleable, and small dents only mildly modify the efficiency of the engine, and that's one of the known advantages of Niobium over other high temperature metals, and partly why it was used. So it probably only got dinged because they knew they could safely reduce the gap without worrying about the nozzle shattering or something.
But you said free speech has no limits? Suddenly it has limits? You can't electronically amplify speech? Does being able to pretend to be several different people online count as an amplification? How about the fact that you are often without any way to check their true identity? Doesn't that cause a disturbance to normal commerce?
I agree, and people should be allowed to shout their free speech as loud as they like, even with electronic amplification; anything up to 200 dBa is fine. If it kills people they should have worn headphones. The right to free speech is more important.
'many times' huh? What exactly does that mean?
The English Wikipedia has over 1.5 million articles. What does many times mean? 10 articles? A hundred articles? Were these major articles read by thousands of people a day, or were they tiny stub articles?
The average life of a vandalisation is typically measured in *minutes* on any major article. People are checking the wikipedia that frequently. You can go a very long time on the wikipedia without ever seeing a single vandalised article.
s a general matter, under the current "first to invent" system, you are right. But, the parent post mentioned a "first to file" system (as used in many other countries). Under "first to file" prior art doesn't matter. All that matters is who wins the race to the patent office.
I'm sorry, but you're quite wrong on this. In the UK we have a first to file system and prior art does matter to an ridiculous degree- you are not allowed to publish your invention prior to filing. The logic is that patents are intended to encourage people to reveal 'trade secrets', and if you've told everyone anyway, then it's not a secret, so no patent!
The only time the race matters is if two people independently develop something, and then try to patent it, in that case the first one there wins. I think that that's fair enough, and the patent date is at least inarguable, whereas you could quibble for decades about when you thought of something first.
You're wrong about the wavelength in this case. If you don't believe me stick an ant in a small-microwave transparent container to stop it running around, add a small cup of water beside it, to avoid overloading the magnetron, and turn it on.
The ant dies, because the microwaves cause the water molecules to oscillate at 2.5Ghz, and they bounce off each other, generating heat, cooking the ant. But the ant is *much* smaller than the wavelength of the oven (~12cm)
You can't use 12cm microwaves to locate the ant precisely (radar-style), but they cook it just fine.
More accurately, it has a lot to do with reducing the cost of spaceflight, which has everything to do with making space exploration more cost effective. That means you can do more exploration for the same price.
Well, one of the ways to generate energy without using carbon is to launch solar panels into orbit and beam solar energy back to earth using diffuse radio waves (which are as safe or safer than cell phone and are not a death ray in any way shape or form). You get more power that way from the solar panels, since they see sun 24x7, about 3 or 4 times more than you get on the Earth and it can cost about the same to launch the panels as it takes to make them if you do both in large quantity.
The economics of doing this gives electricity about the same price as it is right now. Energy pay back over the energy to make the rocket happens in a few years and the life is about 15 years on orbit.
The costs of doing this keep going down, due to mass production solar panels, and reductions in the cost of launch. Meanwhile the price of oil is generally rising making the carbon generation route more expensive. Provided you go for it with a few gigawatt powerstation, it's cost effective and pays back the money borrowed to do it in a few years. That kind of size is enough for a million homes or so.
UDP is designed to expand it's use of bandwidth to the amount available, and dropped packets are just dropped.
No, dude, get a clue. It's designed to do no such thing. A moronically written application sitting on top of UDP might try to pull crap like that, and there's nothing in UDP to stop it, but equally there's nothing in UDP to encourage it either. And if your Isp finds you running an application like that, with no sensible congestion strategy you'd be history and rightly so.
So, riddle me this. What happens when you have TCP and UDP running on the same line? UDP takes up all the bandwidth NOT in use by TCP. TCP, seeing that the bandwidth is saturated (and dropping packets) throttles down. UDP, seeing that there's suddenly more room for UDP, throttles up. In response, TCP throttles down. In response UDP throttles up.
Most QOS apps are roughly constant bitrate. So the UDP throttles up until it gets enough bandwidth, and then holds it at that, and yeah TCP gets out of the way enough until it gets what it needs. This is not bad- this is what you want to happen right? You want P2P and web stuff to not hold up VOIP traffic. And if UDP is unable to get enough bandwidth it kills the connection; which is also what you want to happen. In other words UDP is a way of getting QOS on a conventional IP network; which is why it is used. This is not bad or evil unless they take unreasonable amounts of traffic (on average); but that's why ISPs traffic shape, so that on average their customers don't abuse their connections. It all works reasonably well provided the customers don't exceed their contention ratio on average.
Well, the network I'm on right now has QOS. So far as I can tell they mainly use it to degrade P2P traffic during peak times. They still need to deliver the contention ratio that we buy though, so I don't think it saves them that much money.
The point is that it isn't empty, and the other point is that bandwidth is dirt cheap.
Actually QOS schemes ultimately rely on having plenty of bandwidth. As soon as you run out, all QOS does for you is degrade your service slightly more 'gracefully', but the network becomes rapidly less neutral.
No, they don't have any fancy Quality of Service software, they've just got plenty of bandwidth in their core network. You don't need Quality of Service if the network is practically empty all the time; there's no congestion.
Someone else who comes up with the idea, dilligently works to develop it, and then gives the knowledge up to the public IS entitled to a patent.
No, because somebody can patent the idea before that, even if they haven't managed to get it to work. Then when some poor sod does manage to do that, they can't produce the product, and in some cases all working copies get transfered to the patent owner. The patent owner owns the idea lock stock and barrel until the patent runs out.
get rid of corrupt American politicians that took huge backhanders during the CAN-SPAM fiasco
get the politicians to write legislation with real bite. It can take up to 15 seconds to delete an email e.g. so 15 seconds of prison time for every sent spam email sounds about right; i.e. 8 months in prison for a million emails. On second thoughts 60 seconds in prison, because they knew what they was doing was wrong, so 30 months in prison. A few spam runs, and it's essentially life imprisonment. Yay! (My heart bleeds, but essentially they kill person lifetimes every time they do a spam run).
work out how the spammers get paid, and freeze it out; no dosh, no dodgy email.
No, it's somewhat a grey area, it's going to depend on what they did and how far outside the terms of the contract they go; but there's legislation pushing it towards copyright infringement in most cases.
In practice the court has decide whether it is a small breach of contract or a large breach of contract. In the Microsoft vs Sun case the court ruled that it was a bit arguable- different parts of the license said different things, so they bounced it back to the original court and called it contract infringement rather than copyright. In a different case it would have been a major breach of contract, and then for all intents and purposes the contract is irrelevant, they would have been nowhere near meeting it, so they're acting as if the contract doesn't exist, and copyright law would come down on them like the wrath of God.
In this case they've just completely ignored major terms of the contract; they'll try and make it look like it's the same as the Microsoft/Sun, but I doubt they'll manage it.
Slashdot Mirror
Only in your bizarro world where Bill Gates is on the test team.
I agree. He should have used the *full* name of another one of the testers.
Microsoft would be happy cos they'd have somebody to blame. He'd be happy because he doesn't get blamed. The poor sap that got blamed would be very unhappy.
- but that's just pure selfishness, since, as they say in Star Trek- 'the needs of the many outweigh the needs of the few or the one.'
Yeah, and you can trust a company that says that when they've just been fined for deliberately misleading consumers why exactly?
On the arocket email group the consensus seems to be that it looks like fuel slosh being driven by the control system moving the nozzle in a circular mode. Eventually the magnitude of the control inputs seems to have created a roll angle, and that's what killed the telemetry and the engine apparently is likely to have shut down shortly afterwards.
Did anyone else notice the bump the Kestrel engine took during stage separation? On the 40MB video from SpaceX, it happend at 3:28 in or at T+00:02:52 on the screen clock. Maybe this is normal for the engine, but it was rather odd looking to me. So far as is known, it didn't materially affect anything. The nozzle is made of Niobium which is quite malleable, and small dents only mildly modify the efficiency of the engine, and that's one of the known advantages of Niobium over other high temperature metals, and partly why it was used. So it probably only got dinged because they knew they could safely reduce the gap without worrying about the nozzle shattering or something.
Hibernate would be OK, since it writes everything to the hard-drive and then does a shutdown. It's standby you have to worry about.
But you said free speech has no limits? Suddenly it has limits? You can't electronically amplify speech? Does being able to pretend to be several different people online count as an amplification? How about the fact that you are often without any way to check their true identity? Doesn't that cause a disturbance to normal commerce?
I agree, and people should be allowed to shout their free speech as loud as they like, even with electronic amplification; anything up to 200 dBa is fine. If it kills people they should have worn headphones. The right to free speech is more important.
'many times' huh? What exactly does that mean? The English Wikipedia has over 1.5 million articles. What does many times mean? 10 articles? A hundred articles? Were these major articles read by thousands of people a day, or were they tiny stub articles? The average life of a vandalisation is typically measured in *minutes* on any major article. People are checking the wikipedia that frequently. You can go a very long time on the wikipedia without ever seeing a single vandalised article.
On that basis predicting the weather should be easy, since molecules in the atmosphere are dumb as rocks, even dumber that dumb people.
And yet... weather forecasting requires supercomputers.
You're confusing dumbness with predictability. They're not the same thing, although dumb people can be predictable sometimes.
s a general matter, under the current "first to invent" system, you are right. But, the parent post mentioned a "first to file" system (as used in many other countries). Under "first to file" prior art doesn't matter. All that matters is who wins the race to the patent office.
I'm sorry, but you're quite wrong on this. In the UK we have a first to file system and prior art does matter to an ridiculous degree- you are not allowed to publish your invention prior to filing. The logic is that patents are intended to encourage people to reveal 'trade secrets', and if you've told everyone anyway, then it's not a secret, so no patent!
The only time the race matters is if two people independently develop something, and then try to patent it, in that case the first one there wins. I think that that's fair enough, and the patent date is at least inarguable, whereas you could quibble for decades about when you thought of something first.
You're wrong about the wavelength in this case. If you don't believe me stick an ant in a small-microwave transparent container to stop it running around, add a small cup of water beside it, to avoid overloading the magnetron, and turn it on.
The ant dies, because the microwaves cause the water molecules to oscillate at 2.5Ghz, and they bounce off each other, generating heat, cooking the ant. But the ant is *much* smaller than the wavelength of the oven (~12cm)
You can't use 12cm microwaves to locate the ant precisely (radar-style), but they cook it just fine.
That's not entirely true, they're quite complex; but I won't explain the benefits here. They have a real component.
Really? Got a patent number for us, Mr Anon?
Sharks can sense electromagnetic fields and they often chew on stuff that generates fields. There's power running down the cable; to power repeaters.
Ironically the cables are armoured against the sharks (with or without the frickin lasers); so they have to use grapples instead.
It's been like that for ages. They kept finding shark teeth embedded in the cables.
More accurately, it has a lot to do with reducing the cost of spaceflight, which has everything to do with making space exploration more cost effective. That means you can do more exploration for the same price.
Well, one of the ways to generate energy without using carbon is to launch solar panels into orbit and beam solar energy back to earth using diffuse radio waves (which are as safe or safer than cell phone and are not a death ray in any way shape or form). You get more power that way from the solar panels, since they see sun 24x7, about 3 or 4 times more than you get on the Earth and it can cost about the same to launch the panels as it takes to make them if you do both in large quantity.
The economics of doing this gives electricity about the same price as it is right now. Energy pay back over the energy to make the rocket happens in a few years and the life is about 15 years on orbit.
The costs of doing this keep going down, due to mass production solar panels, and reductions in the cost of launch. Meanwhile the price of oil is generally rising making the carbon generation route more expensive. Provided you go for it with a few gigawatt powerstation, it's cost effective and pays back the money borrowed to do it in a few years. That kind of size is enough for a million homes or so.
UDP is designed to expand it's use of bandwidth to the amount available, and dropped packets are just dropped.
No, dude, get a clue. It's designed to do no such thing. A moronically written application sitting on top of UDP might try to pull crap like that, and there's nothing in UDP to stop it, but equally there's nothing in UDP to encourage it either. And if your Isp finds you running an application like that, with no sensible congestion strategy you'd be history and rightly so.
So, riddle me this. What happens when you have TCP and UDP running on the same line? UDP takes up all the bandwidth NOT in use by TCP. TCP, seeing that the bandwidth is saturated (and dropping packets) throttles down. UDP, seeing that there's suddenly more room for UDP, throttles up. In response, TCP throttles down. In response UDP throttles up.Most QOS apps are roughly constant bitrate. So the UDP throttles up until it gets enough bandwidth, and then holds it at that, and yeah TCP gets out of the way enough until it gets what it needs. This is not bad- this is what you want to happen right? You want P2P and web stuff to not hold up VOIP traffic. And if UDP is unable to get enough bandwidth it kills the connection; which is also what you want to happen. In other words UDP is a way of getting QOS on a conventional IP network; which is why it is used. This is not bad or evil unless they take unreasonable amounts of traffic (on average); but that's why ISPs traffic shape, so that on average their customers don't abuse their connections. It all works reasonably well provided the customers don't exceed their contention ratio on average.
Well, the network I'm on right now has QOS. So far as I can tell they mainly use it to degrade P2P traffic during peak times. They still need to deliver the contention ratio that we buy though, so I don't think it saves them that much money.
The point is that it isn't empty, and the other point is that bandwidth is dirt cheap.
Actually QOS schemes ultimately rely on having plenty of bandwidth. As soon as you run out, all QOS does for you is degrade your service slightly more 'gracefully', but the network becomes rapidly less neutral.
No, they don't have any fancy Quality of Service software, they've just got plenty of bandwidth in their core network. You don't need Quality of Service if the network is practically empty all the time; there's no congestion.
Someone else who comes up with the idea, dilligently works to develop it, and then gives the knowledge up to the public IS entitled to a patent.
No, because somebody can patent the idea before that, even if they haven't managed to get it to work. Then when some poor sod does manage to do that, they can't produce the product, and in some cases all working copies get transfered to the patent owner. The patent owner owns the idea lock stock and barrel until the patent runs out.
A number of things:
No, it's somewhat a grey area, it's going to depend on what they did and how far outside the terms of the contract they go; but there's legislation pushing it towards copyright infringement in most cases.
In practice the court has decide whether it is a small breach of contract or a large breach of contract. In the Microsoft vs Sun case the court ruled that it was a bit arguable- different parts of the license said different things, so they bounced it back to the original court and called it contract infringement rather than copyright. In a different case it would have been a major breach of contract, and then for all intents and purposes the contract is irrelevant, they would have been nowhere near meeting it, so they're acting as if the contract doesn't exist, and copyright law would come down on them like the wrath of God.
In this case they've just completely ignored major terms of the contract; they'll try and make it look like it's the same as the Microsoft/Sun, but I doubt they'll manage it.