Well, as far as I remember, there were more problems than that.
The problems with very high bandwidth links, TCP and RTT estimation start from the fact that TCP can do that estimation just every ACK received and with very high bandwidth links it changes much faster and in greater degree. So, TCP can't efectively estimate the available capacity, since it cannot probe the channel frequently enough.
Caltech's Vegas looks great on pictures, however, there were papers pointing out that it's not exactly fair, especially with multiple bottlenecks of real-world topology. Then there were papers fixing that, and papers critisizing those solutions, and as the result I don't see Vegas anywhere around (except for some Cisco routers maybe) - the best I see is NewReno+SACK+FACK+ECN. I can imagine that more aggressive scheme will have an advantage over TCP, although NewReno is pretty aggressive if compared to most RT rate control schemes, so it's difficult to imagine anything more aggressive than that, that would yield in times of congestion.
The best description of what they really propose seems to be in their Infocom's paper from April this year. That looks pretty good, too. But again, as it was with original Vegas, it will probably come out that it has some flaws, they will be fixed, the fixes will have some flaws, and so on. And for the time being everybody will continue to use NewReno. *snicker*
Fact is that there is enormous (partly bad) experience with using TCP Reno, and with current abundance of capacity in the backbones, it doesn't seem that there is much on an interest in precise traffic control. I've got to have my first problem watching some movie trailer, yet.;-)
One thing worth mentioning - no reasonable application uses TCP for multimedia (why Disney then?). RTP/UDP with a reasonable model-based rate control can easily at least match Vegas, and often outperform it because of the kind and amount of feedback used to adapt to the network conditions for particular application. Caltech's scheme was constructed for ultra highspeed networking and tested for processing of vast data volumes produced by LHC to overcome deficiences of traditional TCP in that case. They have a real nice article on experiments with that with good results. But that's not quite the same as typical situation.
Assuming that it even existed, you can always make something that realizes a totally different task from the intended - that might be actually beneficial - I could use a teleporter for example. >:->
The warranty comes from the fact that VLSI chips have a tendency to have natural faults that come from overheating singular transistors. While they are extensively tested, those faults can naturally happen during normal operation with very low probability. Of course, overclocking greatly increases that probability, because with every clock tick the same current flows through the chip, but since the ticks are more frequent, the total energy dissipated within a unit of time is greater, increasing chance that certain points can be overheated and polysilicon at transistor gates will melt and short-circuit to source and/or drain of a transistor.
In any case, those things happen normally, too. It's a probabilistic problem and depends on temporary conditions and variation in chip manufacturing process.
If you check, a lot of people who write open source software are with.edu domain.
The reality of the US Universities can be seen here - my university (very bad - you get at most 30% of anything that you develop), or here - across the street from us (a bit better - 50%).
The bottom line is that it absolutely doesn't pay off to be innovative and creative for profit there.
That would be actually a very interesting service. Say you go to the bank and ask the bank to provide a number for your kid with some monthly allowance from your account. They don't actually have to give your kid a plastic - just the number. Of course that would have to be exactly debit, no other uses possible. That means that you'd have to be able to actually use some PIN number. One option would be that the bank could list such transactions as pending on the account and you'd have to access the online banking site of your bank to authorize them.
A quick look at HMV - one of the retailers mentioned in the article reveals that they are going to use Liquid Audio format. Player is free, but the format is as proprietary as it gets. Now, that smells like another software empire 5 years from now, doesn't it? The best M$ move now would be to simply buy LiquidAudio (if they haven't yet...). Of course, Liquid Audio player is only for Windows - I'm guessing why and I don't have to think hard. When will people learn?
CNN BreakingNews:
Theo De Raadt, a reknown Canadian programmer and system engineer openly criticized DARPA for providing him with over 2 million dollars in grant money. He expressed his deep concerns that such a war-mongering agency would give him money for his peaceful project.
Well, those that are in the article need a tank of liquid hellium, rather than nitrogen. But, why would you want to use spin currents and try to inject them? That's what you have magnetic field for.
as a company's employee - maybe as an expert. AFAIK he was a genius at using tools, but I don't remember him creating any of them. Maybe I'm mistaken? That brings another question: if somebody creates a tool and somebody else uses it, who is the bad guy? Recent stories (like the one of DeCSS and the one about RIAA suing students) show that people start to go after those that make tools. Shouldn't we start prosecuting gun, hammer, ax, and car manucaturers?
OK. I've managed to get hold on the original article in PRL.
I'm not an expert in semiconductor physics, but from general look, it seems that:
the article (and discovery) needs to be evaluated by other people - we had hoaxes like that before,
the technology is kind of useless right now.
The latter comes from the fact that it requires ferromagnetic semiconductors - as yet we don't have too many of those that work at reasonable temperatures - room temperatures at least and +200oF to make it useful for any computer chips. Those that were used in article have Curie point at 45oK. (Curie point is the temperature below which a material becomes ferromagnetic and above which it loses those properties. This is due to crystalization of the material - iron normally consists of many crystals that serve as little magnets. Each crystal normally points in a random direction which makes it non-magnetic. After magnetizing, they point in the same direction and create a magnet. At certain temperature - different for different materials, the material loses internal crystalline structure and ceases to have ferromagnetic properties - crystals melt away.) 45oK is a bit too low for any practical purposes with wide application. I'm not sure how high can they go at the moment. If that is comparable to superconductors, there aren't any great chances to see wide applications of that. Even if we obtain high-temperature ferromagnetic semiconductors, there is no guarantee that the effect will hold in higher temperatures. Also, the temperatures are so low, that there might be some superconducting effect involved.
Well, there are some geniuses that can visualize complex, 2000 files, 1mln lines of code software design and remember all components. There are some who can possibly write this whole thing in edlin and not miss a character, compile and voila - there you have a perfect operating system, word processing application, VLSI design package. The only problem with them is that they are stubbornly hiding in some caves. I have never seen any.
I appreciate IDEs for:
functions' and variables' definitions at the mouse pointer,
API descriptions - ditto,
built-in extensive debugging support with good data visualization,
possiblity to automate obvious actions - like CVS updates, compilations - likelihood of typing rm -rf src at 4am decreases dramatically
in many cases - really, really good support for collaborative program development,
possible dozen other things I don't remember right now but that come in handy now and then.
I've written large programs in vi, I've used IDEs starting from Turbo C 2.0 (even fixed a bug in that one). There is absolutely no advantage of vi over any IDE except for relatively small size, there are numerous advantages of good IDEs over vi. NetBeans is surprisingly good as a total freeware. While Eclipse looks simpler and will be definitely more responsive looking just at the GUI screenshots, I do appreciate some differences from Eclipse.
I've been using NetBeans 3.4. As yet it forced me to upgrade from 256MB RAM to 640 - with 256 it was simply impossible to use. I'd gladly welcome a good update, so I open my NetBeans, it asks about autoupdate and there it is not...
Are they gonna actually post it on autoupdate? I don't really feel like downloading the stuff from the web and reinstalling everything manually...
Now, I can't really say a bad word about NetBeans responsiveness - it just requires unbelievable amounts of memory - with 640MB and 1.1GHz CPU it works just as it should. I hope they will fix RMI debugging support sometime though.
You forgot one thing: you must ensure that the constructor of the circuit to be disconnected must not be around the device after the blade has been installed.
There has been some start to the list. Let's continue:
cell phones - arguably the conceal where you are calling from
prepaid cell phones - no comment
wireless phones - you can call from any place in your house and within quite a decent range outside of it
how about phone booths?
all NATs as it has been mentioned zillion times
walkie-talkie thingies
baby monitors
I think that the list is essentially infinite. The real question is how the law would be used. Law in itself is just a piece of paper. GB still has laws dating many centuries back that give some people rights to do weird things (e.g. take a herd of domestic animals across the main London bridge...)
Say, that, hypothetically, Verizon starts to execute its right to control who connected what to their service. That would be the nail to the end of their broadband service. Nobody would buy any service from them and all those who have it would drop it immediately (or after waiting 'til their 8 months expire to avoid paying some $170 of penalty). IMHO, that's gonna be one more thing we are going to have a good laugh at.
If guns are outlawed, only outlaws will have guns.
Isn't that a simple consequence of the fact that the guns would be outlawed? Simply, everybody who has a gun will be an outlaw. Same applies to hackers, students, water drinkers and air breathers.
My school has card system. It's used by everybody as a photo ID for on-campus checks (nobody else really accepts that - the picture is pretty obscure and they are valid for 5 years). Students pay with it for meals in one (1) eatery. I'm not sure about dorms. I haven't ever seen any soda dispensers operated with cards - everything around uses coins. Simple - I don't that the system was advertised as highly secure (it might be now - with all new "security bubble" that seems to be growing fast) - it's just a card ID system. Whoever wants to use it as a highly secure thing is responsible for any problems that come out of that.
Well, as I said there is that 0.1%. Talking about kids - you can go out now and see people 10, 12, or 14 that try their hands on that - mostly in MMORPs.
I also agree that this percentage doesn't exactly depend on the age. Personally, I think that a lot of coding tasks are just brain-damaging. Interesting that everybody worries about RSS in their wrists, but nobody cares about their brains. That's why I think that people shouldn't start too early in that business.
Right, only current technology is somewhat flawed. Let's see - I can buy and download copyrighted content, then I have to download the key. If I lose the key, sometimes I can recover it within a short period of time after the content was sold to me, sometimes I can't. Who wants to buy into stuff like that?
So, to protect myself I'd rip that content right away after getting, so if I happen to reinstall Windows because of some stupid bug in it, and happen to have the copy of the content, I can still play it. Also, I can play it then anywhere I want (like I can now with a CD that I own) for my pleasure.
So, they happen to kill that with DMCA. Great...
Before we get a copyrighted content distribution system that doesn't require all that hassle, forget it. But there is no existing solution that would be scalable to millions of users and achieve availability of CD. In short words, the days of copyrighted protection by the way that originated from printed books is over. Either we come up with a different system of supporting artists (that doesn't necessarily rely on the very good will of the customer), or the industry can say bye bye to profits from CDs. They will still have DVDs (because that barely pays off to rip) and royalties from broadcasting. Is that really so bad?
Has anybody done any research how the rate of CD/DVD sales increase/decrease changed after RIAA started its actions?
I understand that the purpose of RIAA is to increase the multimedia industry (the big ones) profits. Now, in my opinion they didn't start very well:
They started off with lawsuits against students - are they really counting on those students paying off any possible sentences? Com'on - they will file bankruptcy (if they lose that is) right after walking out of the courtroom.
I believe those lawyers at RIAA charge quite a bit for that stuff - does it really increase the profits?
Who is actually gonna be encouraged to buy anything from those guys (that is CD/DVD business) if everybody has a hangover after their actions?
Somehow I don't see those bilions of dollars flowing into the industry crooks' pockets...
Won't work. Those that are about to retire have already earned rights to their benefits. Theoretically, those benefits should be there anyway - they've been paying for them the whole life. Another place where the theory doesn't meet with reality.
Well, I happen to have seen stuff produced by teenage kid programmers and 99.9% is not pretty... Certainly there is that 0.1%, but that's like with anything else. If you want to write something, you can hire anybody. If you want some quality, that takes a little experience. On the other hand, quality is probably not on The Most Wanted list, recently.
802.16-based networks are supposed to have encryption, too. It's not included in 802.16 standard itself, but the networks are supposed to be a physical layer for DOCSIS (Data Over Cable Standard Interface Specification - privacy, authentication and management layer for Cable Modems). Over that we will have 802.3 (Ethernet) protocols and standard IP stack. That's at least as far as I could get in that.
Anyway DOCSIS specified the use of RSA and DES (other symetrical ciphers are optional), message signatures, senders authentication. In 802.16 it will be actually easier than in Cable, because the network is supposed to be connection-oriented.
Slashdot Mirror
Well, as far as I remember, there were more problems than that.
;-)
The problems with very high bandwidth links, TCP and RTT estimation start from the fact that TCP can do that estimation just every ACK received and with very high bandwidth links it changes much faster and in greater degree. So, TCP can't efectively estimate the available capacity, since it cannot probe the channel frequently enough.
Caltech's Vegas looks great on pictures, however, there were papers pointing out that it's not exactly fair, especially with multiple bottlenecks of real-world topology. Then there were papers fixing that, and papers critisizing those solutions, and as the result I don't see Vegas anywhere around (except for some Cisco routers maybe) - the best I see is NewReno+SACK+FACK+ECN. I can imagine that more aggressive scheme will have an advantage over TCP, although NewReno is pretty aggressive if compared to most RT rate control schemes, so it's difficult to imagine anything more aggressive than that, that would yield in times of congestion.
The best description of what they really propose seems to be in their Infocom's paper from April this year. That looks pretty good, too. But again, as it was with original Vegas, it will probably come out that it has some flaws, they will be fixed, the fixes will have some flaws, and so on. And for the time being everybody will continue to use NewReno. *snicker*
Fact is that there is enormous (partly bad) experience with using TCP Reno, and with current abundance of capacity in the backbones, it doesn't seem that there is much on an interest in precise traffic control. I've got to have my first problem watching some movie trailer, yet.
One thing worth mentioning - no reasonable application uses TCP for multimedia (why Disney then?). RTP/UDP with a reasonable model-based rate control can easily at least match Vegas, and often outperform it because of the kind and amount of feedback used to adapt to the network conditions for particular application. Caltech's scheme was constructed for ultra highspeed networking and tested for processing of vast data volumes produced by LHC to overcome deficiences of traditional TCP in that case. They have a real nice article on experiments with that with good results. But that's not quite the same as typical situation.
Assuming that it even existed, you can always make something that realizes a totally different task from the intended - that might be actually beneficial - I could use a teleporter for example. >:->
The warranty comes from the fact that VLSI chips have a tendency to have natural faults that come from overheating singular transistors. While they are extensively tested, those faults can naturally happen during normal operation with very low probability. Of course, overclocking greatly increases that probability, because with every clock tick the same current flows through the chip, but since the ticks are more frequent, the total energy dissipated within a unit of time is greater, increasing chance that certain points can be overheated and polysilicon at transistor gates will melt and short-circuit to source and/or drain of a transistor.
In any case, those things happen normally, too. It's a probabilistic problem and depends on temporary conditions and variation in chip manufacturing process.
If you check, a lot of people who write open source software are with .edu domain.
The reality of the US Universities can be seen here - my university (very bad - you get at most 30% of anything that you develop), or here - across the street from us (a bit better - 50%).
The bottom line is that it absolutely doesn't pay off to be innovative and creative for profit there.
That would be actually a very interesting service. Say you go to the bank and ask the bank to provide a number for your kid with some monthly allowance from your account. They don't actually have to give your kid a plastic - just the number. Of course that would have to be exactly debit, no other uses possible. That means that you'd have to be able to actually use some PIN number. One option would be that the bank could list such transactions as pending on the account and you'd have to access the online banking site of your bank to authorize them.
A quick look at HMV - one of the retailers mentioned in the article reveals that they are going to use Liquid Audio format. Player is free, but the format is as proprietary as it gets. Now, that smells like another software empire 5 years from now, doesn't it? The best M$ move now would be to simply buy LiquidAudio (if they haven't yet...). Of course, Liquid Audio player is only for Windows - I'm guessing why and I don't have to think hard. When will people learn?
CNN BreakingNews: Theo De Raadt, a reknown Canadian programmer and system engineer openly criticized DARPA for providing him with over 2 million dollars in grant money. He expressed his deep concerns that such a war-mongering agency would give him money for his peaceful project.
I think it's the shortest definition and the most accurate. And actually means that cracker and hacker are mutually exclusive.
Well, those that are in the article need a tank of liquid hellium, rather than nitrogen. But, why would you want to use spin currents and try to inject them? That's what you have magnetic field for.
Yeah, and Painter doesn't :-)
as a company's employee - maybe as an expert. AFAIK he was a genius at using tools, but I don't remember him creating any of them. Maybe I'm mistaken? That brings another question: if somebody creates a tool and somebody else uses it, who is the bad guy? Recent stories (like the one of DeCSS and the one about RIAA suing students) show that people start to go after those that make tools. Shouldn't we start prosecuting gun, hammer, ax, and car manucaturers?
I'm not an expert in semiconductor physics, but from general look, it seems that:
the article (and discovery) needs to be evaluated by other people - we had hoaxes like that before,
the technology is kind of useless right now.
The latter comes from the fact that it requires ferromagnetic semiconductors - as yet we don't have too many of those that work at reasonable temperatures - room temperatures at least and +200oF to make it useful for any computer chips. Those that were used in article have Curie point at 45oK. (Curie point is the temperature below which a material becomes ferromagnetic and above which it loses those properties. This is due to crystalization of the material - iron normally consists of many crystals that serve as little magnets. Each crystal normally points in a random direction which makes it non-magnetic. After magnetizing, they point in the same direction and create a magnet. At certain temperature - different for different materials, the material loses internal crystalline structure and ceases to have ferromagnetic properties - crystals melt away.) 45oK is a bit too low for any practical purposes with wide application. I'm not sure how high can they go at the moment. If that is comparable to superconductors, there aren't any great chances to see wide applications of that. Even if we obtain high-temperature ferromagnetic semiconductors, there is no guarantee that the effect will hold in higher temperatures. Also, the temperatures are so low, that there might be some superconducting effect involved.
I appreciate IDEs for:
functions' and variables' definitions at the mouse pointer,
API descriptions - ditto,
built-in extensive debugging support with good data visualization,
possiblity to automate obvious actions - like CVS updates, compilations - likelihood of typing rm -rf src at 4am decreases dramatically
in many cases - really, really good support for collaborative program development,
possible dozen other things I don't remember right now but that come in handy now and then.
;)
I've written large programs in vi, I've used IDEs starting from Turbo C 2.0 (even fixed a bug in that one). There is absolutely no advantage of vi over any IDE except for relatively small size, there are numerous advantages of good IDEs over vi. NetBeans is surprisingly good as a total freeware. While Eclipse looks simpler and will be definitely more responsive looking just at the GUI screenshots, I do appreciate some differences from Eclipse.
yeah, bog me down for trolling
I've been using NetBeans 3.4. As yet it forced me to upgrade from 256MB RAM to 640 - with 256 it was simply impossible to use. I'd gladly welcome a good update, so I open my NetBeans, it asks about autoupdate and there it is not...
Are they gonna actually post it on autoupdate? I don't really feel like downloading the stuff from the web and reinstalling everything manually...
Now, I can't really say a bad word about NetBeans responsiveness - it just requires unbelievable amounts of memory - with 640MB and 1.1GHz CPU it works just as it should. I hope they will fix RMI debugging support sometime though.
You forgot one thing: you must ensure that the constructor of the circuit to be disconnected must not be around the device after the blade has been installed.
cell phones - arguably the conceal where you are calling from
prepaid cell phones - no comment
wireless phones - you can call from any place in your house and within quite a decent range outside of it
how about phone booths?
all NATs as it has been mentioned zillion times
walkie-talkie thingies
baby monitors
I think that the list is essentially infinite. The real question is how the law would be used. Law in itself is just a piece of paper. GB still has laws dating many centuries back that give some people rights to do weird things (e.g. take a herd of domestic animals across the main London bridge...)
Say, that, hypothetically, Verizon starts to execute its right to control who connected what to their service. That would be the nail to the end of their broadband service. Nobody would buy any service from them and all those who have it would drop it immediately (or after waiting 'til their 8 months expire to avoid paying some $170 of penalty). IMHO, that's gonna be one more thing we are going to have a good laugh at.
If guns are outlawed, only outlaws will have guns.
Isn't that a simple consequence of the fact that the guns would be outlawed? Simply, everybody who has a gun will be an outlaw. Same applies to hackers, students, water drinkers and air breathers.
My school has card system. It's used by everybody as a photo ID for on-campus checks (nobody else really accepts that - the picture is pretty obscure and they are valid for 5 years). Students pay with it for meals in one (1) eatery. I'm not sure about dorms. I haven't ever seen any soda dispensers operated with cards - everything around uses coins. Simple - I don't that the system was advertised as highly secure (it might be now - with all new "security bubble" that seems to be growing fast) - it's just a card ID system. Whoever wants to use it as a highly secure thing is responsible for any problems that come out of that.
Well, as I said there is that 0.1%. Talking about kids - you can go out now and see people 10, 12, or 14 that try their hands on that - mostly in MMORPs. I also agree that this percentage doesn't exactly depend on the age. Personally, I think that a lot of coding tasks are just brain-damaging. Interesting that everybody worries about RSS in their wrists, but nobody cares about their brains. That's why I think that people shouldn't start too early in that business.
Right, only current technology is somewhat flawed. Let's see - I can buy and download copyrighted content, then I have to download the key. If I lose the key, sometimes I can recover it within a short period of time after the content was sold to me, sometimes I can't. Who wants to buy into stuff like that?
So, to protect myself I'd rip that content right away after getting, so if I happen to reinstall Windows because of some stupid bug in it, and happen to have the copy of the content, I can still play it. Also, I can play it then anywhere I want (like I can now with a CD that I own) for my pleasure.
So, they happen to kill that with DMCA. Great...
Before we get a copyrighted content distribution system that doesn't require all that hassle, forget it. But there is no existing solution that would be scalable to millions of users and achieve availability of CD. In short words, the days of copyrighted protection by the way that originated from printed books is over. Either we come up with a different system of supporting artists (that doesn't necessarily rely on the very good will of the customer), or the industry can say bye bye to profits from CDs. They will still have DVDs (because that barely pays off to rip) and royalties from broadcasting. Is that really so bad?
I understand that the purpose of RIAA is to increase the multimedia industry (the big ones) profits. Now, in my opinion they didn't start very well:
They started off with lawsuits against students - are they really counting on those students paying off any possible sentences? Com'on - they will file bankruptcy (if they lose that is) right after walking out of the courtroom.
I believe those lawyers at RIAA charge quite a bit for that stuff - does it really increase the profits?
Who is actually gonna be encouraged to buy anything from those guys (that is CD/DVD business) if everybody has a hangover after their actions?
Somehow I don't see those bilions of dollars flowing into the industry crooks' pockets...
Won't work. Those that are about to retire have already earned rights to their benefits. Theoretically, those benefits should be there anyway - they've been paying for them the whole life. Another place where the theory doesn't meet with reality.
Well, I happen to have seen stuff produced by teenage kid programmers and 99.9% is not pretty... Certainly there is that 0.1%, but that's like with anything else. If you want to write something, you can hire anybody. If you want some quality, that takes a little experience. On the other hand, quality is probably not on The Most Wanted list, recently.
802.16-based networks are supposed to have encryption, too. It's not included in 802.16 standard itself, but the networks are supposed to be a physical layer for DOCSIS (Data Over Cable Standard Interface Specification - privacy, authentication and management layer for Cable Modems). Over that we will have 802.3 (Ethernet) protocols and standard IP stack. That's at least as far as I could get in that.
Anyway DOCSIS specified the use of RSA and DES (other symetrical ciphers are optional), message signatures, senders authentication. In 802.16 it will be actually easier than in Cable, because the network is supposed to be connection-oriented.
Yeah:
You turn it on, it turns itself off, you charge the stuff for two ours.
Wash, rinse, repeat...
Check out this:
802.16 News
Broadband Wireless Exchange