Remember the British RIP bill (Regulation of Investigatory Powers) requires British citizens and companies to hand over crypto keys to the gov on demand; a two year jail term awaits those who refuse or have lost their keys. ( See http://www.fipr.org/rip/ )
Don't think that the US could do the same ?
Then consider that certain URI linking is on its way to becoming illegal in the US; enforcing this is a good reason for the FBI to tap ISP lines, no? (see http://www.eff.org/br/br1.html#1 )
The FBI and other LEAs are worried because of the potential of moving the actual servers outside of the US. If the email resides on servers elsewhere, then the US laws don't have much effect.
A LEA could get "taps" on the dial-up or other connection points, but it makes it much tougher to snag that email as the monitored person could dial in from anywhere to any connection point to get their mail. The FBI much rather be able to have the server capture all mail traffic, so they have only one place to go.
This general concern holds for other telecommunications providers. CALEA is the requirements for providing access to telephone, paging, two-way radio, and cell phone systems for "tapping" by law enforcement agencies.
With fines of up to 10K $US per day to service providers who can not provide a CALEA port when served with a tap request, the government is serious about being able to monitor all communications of someone they are investigating. Moving the servers of a US provider outside of the US makes it harder to use that hammer.
But are the yreally new ? A lot of "new interfaces" were either devised back in the `70s when displays were low res and low pixel count, or poked at as a possible direction by people with large research budgets that could pay for what was then high end displays.
Shoot, I'd like to see an implementation of Ted Nelson' Xanadu-style interface. The more/less style is a different method than the common current tree-orientated schemes; outside of outline modes in word processors I don't think it is in use in common human interfaces.
But I think you're right - the interia of "the standard way" in GUIs is likely to greatly slow the spread of any new format.
As others have said, a lot of educational software is pretty poor; a lot of the `Net is trash. And so was the previous generations of "leanring aids" - bad video tapes, movies, text books, and so on.
IMO using a computer as an electronic set of flash cards isn't bad, so long as that is not all that is done to teach the subject. Differing people learn in differing ways, sometimes the rote drillwork helps, sometimes it turns students off.
The problem is devising and allowing teaching schemes that encourge thinking - using the `Net and/or books as a reference, knowing how to find useful information and sort it from the chaff, and how to assemble it. That makes computers as useful in philosophy or history as in math and CS.
Make all the tools and concepts available, train the teachers and provide enough of them with enough time to help students determine which tools are the most useful for that student.
Using computers to replace teachers or teacher aides, is a bad idea driven more by accounting than the desire to educate.
Except that LN2 boiling doesn't take as much energy as most room temp liquids do. Plus once the gas layer is in place the liquid is somewhat isolated from the heat source, which then gets a lot hotter. If you can have boiling while keeping the hot surface mostly covered with liquid, boiling is a great way to remove heat from things. Some of the liquid cooling schemes used a region with lots of sharp points to help induce boiling there, and keep the hot stuff down under a few cm of fluid for the slight rise in boiling point incured by a bit of increased pressure.
I believe that you could patent the concept of rotary blades within the context of a reaper (at least at this time), provided that there was no existing/known usage of rotory blades for reaping. Your patent for such would not prevent me for patenting rotary blades for cutting hair...
The LN2 would boil around the parts generating heat. The resulting gas film is a resonalbe good insulator, frustrating the entire idea of cooling the board. The fluoriner is much higher boiling, keeping a (very cold) liquid coating over the board.
And there is a problem in that LN2 tends to condense oxygen from any air that it comes into contact with. Thus tanks of LN slowly become liquid air, then LOX. And liquid air or oxygen tends to make things burn well, things such as you motherboard. It's better to have some isolation.
Note that DEC did do LN cooling of their LSI11J processors, years ago. That let them run at 3-5 X the rated processor clock. The `J was in a ceramic package.
And, no, LN2 isn't cold enough to make very many things superconductive. Most materials take liquid helium ranges. A NBS document lists Si as going superconductive at 6.7-7.1 degrees K at 120-130 kbars pressure ( 118-128 atmospheres) I don't know what effect teh doping has on it.
The LN2 would boil around the parts generating heat. The resulting gas film is a resonalbe good insulator, frustrating the entire idea of cooling the board. The fluoriner is much higher boiling, keeping a (very cold) liquid coating over the board.
And there is a problem in that LN2 tends to condense oxygen from any air that it comes into contact with. Thus tanks of LN slowly become liquid air, then LOX. And liquid air or oxygen tends to make things burn well, things such as you motherboard. It's better to have some isolation.
Hmmm... well, two of the authors work for ATT, the third is at NYU. But even the two working at ATT doesn't mean that ATT as such has anything to do with this. People tend to work somewhere, most working people work for others or with groups.
Besides, what is important is Does the concept work?, not what is the source of the idea. The concept has been published, if it looks good implement it. The only attack is ATT patenting it, in which case it does us no good for a number of years.
Modulation methods are important, as well. Some spread spectum methods allow several users to share a band, although each users noise floor goes up. With good forward error correction this isn't too much of a problem. (note that FEC takes additional bits, but if you're delievering big hunks of data the percentage used by FEC can drop to a floor level)
Cell structures also multiply bandwidth. If you have a protocol that uses a slice of the spectrum to deliever X bytes/sec, then having N cells can increase the available bandwidth up to N*X - so long as all the active users happen to be in different cells.
And, as someone else sort of mentioned, partitioning can help. Fiber for big pipes to nodes, wideband in cells and micro cells from there to local distribution, short range micro and pico cells for within a neighborhood or building.
The real problem is how long to we have before the machine intelligences hear all this racket we're making, and come to wipe us out ?
We're already sending emails to pagers, one-way and two-way. Note that most two-way is either of the pager acknowledging that it got the message, or the human pushing a button to say "yup, I got it" or "no, I'm busy"
Note that some wireless providers, such as Nextel, want you to use Email to send an alphanumeric message. In the old days you'de use TAP, or TNPP if you were another provider, but now they want Email input from everyone. So, yes, a wardialer style program can spam Nextel users. There's also some TAP hubs that are great inputs, once you know their dial-in number. Once you're there you can just cycle through the set of 10 digits number, noting which get kicked back as "not in service" so you can build a database of how to hit the next time.
So? A good slice of the population doesn't know much more than where the nearest video rental store and pizza shop are. Try asking them how just about any semi-tech aspect of day-to-day life, or geopolitics, and see what answers you get.
A few years ago a local paper ran a page on "how everyday things work" or some such. In it they asked various people how common items worked. A medical technician described TV as "like, sending sound waves through the air, you know? and the TV picks them up and turns them into electricity, and like, shows them to you". This person supposedly has had 2 years of college level education, and is running those lab tests on you; tests that might bear on your health, or say if you've been ingesting substances the government doesn't want you to, or if you have undesireble genes. And yet they're at the level of "I push the round button and wait for the yellow light to go out" - sometimes it makes me think of Cornbluth's "Marching Morons".
A number of years ago, back when for most people "on-line" meant computer BBSes at 1200 bps, I was in a discussion on voter registration. One person wanted people to be able to register at the polls. She said "it's unfair to make people sign up so far in advance, some of my friends are so stoned that they don't even know about an election until election day". So - ignoring the vote fraud problems, you want the president picked by people who are so aware of the issues that they'll vote for the guy with the nicest hair or coolest name?
I think the larger issue is How do we wake more people up? . Wake them up so that they can make resonably informed decisions on important issues. Teaching logic and retoric might help, but it goes beyond schools and into everyday interactions. I don't believe in cutting people out of decision making, but I want them to be able to use reason to make those decisions. The fix for poverty isn't to kill the poor, but to increase the number of well off people; the solution for stupid voters isn't to let fewer vote but to have a greater number of informed voters.
- and I thought that getting older meant that I'd be able to let other people worry about things.
Isn't Jabber basically an "instant messaging" ((aaahhhiiii)) system? BXXP is a full general purpose protocol, IRC/messaging is just one thing it can do well.
Looks like your app could do the blocking, although it will take more work at the proxy level. The other side does have to request the new channel, after all, you should be able to fence off resources.
The possiblity of sneaking past router filtering, and the enhanced privacy support. Add the peer-to-peer aspect...
Typically, a BXXP peer acting in the server role is also acting in a listening role. However, because BXXP is peer-to-peer in nature, no such requirement exists.
which is a bit different than HTTP. This may become important as more IP-speaking devices ( "appliances" ) show up. Even more so as short range wireless connections such as Bluetooth increase, where there might not be a handy server around. Be nice if your left shoe could talk to your right shoe without a server.
So - MS is still expecting to have "Windows" running on everything, and to own +95% of the markets. Seems that makes "Windows only" and "Portability" be equivalents. At least they would be if I owned enough MS stock.
Portability hasn't been a Microsoft concern for a long time.
There needs to be ruling passed from government that traffic can not be censored at a non-authoritarian point: in other words, a company i am not affiliated with can not stop my traffic.
But it is likely that some governments would go along with such censorship, moving it from "some company" to the.gov realm. Those routers are owned by private companies ("Big Bad Business"), governments, and quasi-public/private entities. When not directly government owned they tend to be regulated to some degree; this gives a door to tighter government control if that is desired.
Laws to enforce open communications help, if nothing else they give you a hammer to use when the authorities attempt to obstruct and censor; this may be somewhat effective against other governments through public pressure.
But when it is the government itself that puts the controls in place, usually for "national security", "stopping crime", "enforcing common morality", or "fighting heresy", then the law will be on the side of the blockers.
The companies that own/run the backbone tend to be multinationals. This makes them a little less interersted in local government regs that makes the companies' operations more complicated. But it's not easy to hook directly to them to cut around local routing that may be more annoying to end users; the backbone sells access as tens or hundereds of MB/second. And those big companies are likely to increasing become more tied to media companies, perhaps finding that "revenue lost because of Napster" is important to them
IMHO I think that there needs to be ways to get around the existing `Net. Freenet, encryption in general, real echash, alternative networks for neighborhood level distribution (helps fight traffic analysis), alternative distribution methods that pay the creators of arts better and people who actually support such alternative methods and buy stuff. The user who's sig is "I bought the CD after hearing the MP3" is on the right track. If you can show that the alternatives are fitting into the economy, ie generating jobs and tax revenues, then the government may listed to your argument that your are just replace the buggy whip way of doing business and not stomp on you.
There is signs of a protracted battle in the future, if it happens then "Music should be free and you can't quote my messages in a book" isn't going to cut it.
If you can't code, then document. If you can't do that then learn and spread the ideas.
Given the speed of the micro, handling one user will be a bit of a drag. These things are meant to be the equivalent of a couple of switches on the Web - you can hit the page to see the status of a few bits, and turn them on or off. Low speed A/D/A, serial ports that do short strings of I/O, and so on. If you want much more bandwidth you'll need additional hardware to handle it.
The pico servers I've seen aren't all that robust, dumping on much loading or anything weird coming their way. Some aren't even doing true http, which causes a few browsers to complain.
Actually, it was more due to IBM creating the IBM-PC, which made personal computers legitimate in the eyes of many mainstream business people, and some of the non-nerd general population.
A lot more computers meant a much larger market for software. A lot of new companies, targeting the new market for low cost software to match the low cost machines. And some of the mainframe software vendors trying to sell ports of their products for 24X the cost of the PC.
And yes, there was a fair amount of commerical software before 1980. But low-cost was Apple-II, most people weren't shopping for packages for their CDC or DEC machine.
A lot of the `60s and early `70s vintage systems were good for that sort of thing. Whilst in collage we got the old GE/Honeywell 200 to play music on the disk IO channels using a radio (the most range, the hardest to control), the line printer (mostly brush-style percussion), the card reader (gives a "slap" for each card), and the card punch (sort of a snare roll).
Ah - back when computers were BIG and real programmers could enter programs via the consol switches.
Happen sometimes as test code. I've worked on several embedded systems projects that had easter eggs in them. One time the folks working on the display driver and keyboard/joystick input board wrote a marching line of "pacMan" (tm) munching the screen clear, then took that to an actual pacman game.
They were testing the hardware drivers, and they were ready ahead before anyone else needed their section.
3) The system spec is designed, with clear interfaces, I/O limits, and the like. The product is developed opensource style to that spec, with some folks builting the test suite and tools.
Source + Spec + Test I think that might be even better.
Ah, but the real advantage of RAMBUS will not be apparent until version-II, which will auto-detech open source and free software, adn refuse to store it.
Slashdot Mirror
Don't think that the US could do the same ?
Then consider that certain URI linking is on its way to becoming illegal in the US; enforcing this is a good reason for the FBI to tap ISP lines, no? (see http://www.eff.org/br/br1.html#1 )
A LEA could get "taps" on the dial-up or other connection points, but it makes it much tougher to snag that email as the monitored person could dial in from anywhere to any connection point to get their mail. The FBI much rather be able to have the server capture all mail traffic, so they have only one place to go.
This general concern holds for other telecommunications providers. CALEA is the requirements for providing access to telephone, paging, two-way radio, and cell phone systems for "tapping" by law enforcement agencies.
With fines of up to 10K $US per day to service providers who can not provide a CALEA port when served with a tap request, the government is serious about being able to monitor all communications of someone they are investigating. Moving the servers of a US provider outside of the US makes it harder to use that hammer.
Shoot, I'd like to see an implementation of Ted Nelson' Xanadu-style interface. The more/less style is a different method than the common current tree-orientated schemes; outside of outline modes in word processors I don't think it is in use in common human interfaces.
But I think you're right - the interia of "the standard way" in GUIs is likely to greatly slow the spread of any new format.
IMO using a computer as an electronic set of flash cards isn't bad, so long as that is not all that is done to teach the subject. Differing people learn in differing ways, sometimes the rote drillwork helps, sometimes it turns students off.
The problem is devising and allowing teaching schemes that encourge thinking - using the `Net and/or books as a reference, knowing how to find useful information and sort it from the chaff, and how to assemble it. That makes computers as useful in philosophy or history as in math and CS.
Make all the tools and concepts available, train the teachers and provide enough of them with enough time to help students determine which tools are the most useful for that student.
Using computers to replace teachers or teacher aides, is a bad idea driven more by accounting than the desire to educate.
Except that LN2 boiling doesn't take as much energy as most room temp liquids do. Plus once the gas layer is in place the liquid is somewhat isolated from the heat source, which then gets a lot hotter. If you can have boiling while keeping the hot surface mostly covered with liquid, boiling is a great way to remove heat from things. Some of the liquid cooling schemes used a region with lots of sharp points to help induce boiling there, and keep the hot stuff down under a few cm of fluid for the slight rise in boiling point incured by a bit of increased pressure.
I believe that you could patent the concept of rotary blades within the context of a reaper (at least at this time), provided that there was no existing/known usage of rotory blades for reaping. Your patent for such would not prevent me for patenting rotary blades for cutting hair...
And there is a problem in that LN2 tends to condense oxygen from any air that it comes into contact with. Thus tanks of LN slowly become liquid air, then LOX. And liquid air or oxygen tends to make things burn well, things such as you motherboard. It's better to have some isolation.
Note that DEC did do LN cooling of their LSI11J processors, years ago. That let them run at 3-5 X the rated processor clock. The `J was in a ceramic package.
And, no, LN2 isn't cold enough to make very many things superconductive. Most materials take liquid helium ranges. A NBS document lists Si as going superconductive at 6.7-7.1 degrees K at 120-130 kbars pressure ( 118-128 atmospheres) I don't know what effect teh doping has on it.
And there is a problem in that LN2 tends to condense oxygen from any air that it comes into contact with. Thus tanks of LN slowly become liquid air, then LOX. And liquid air or oxygen tends to make things burn well, things such as you motherboard. It's better to have some isolation.
Besides, what is important is Does the concept work?, not what is the source of the idea. The concept has been published, if it looks good implement it. The only attack is ATT patenting it, in which case it does us no good for a number of years.
Cell structures also multiply bandwidth. If you have a protocol that uses a slice of the spectrum to deliever X bytes/sec, then having N cells can increase the available bandwidth up to N*X - so long as all the active users happen to be in different cells.
And, as someone else sort of mentioned, partitioning can help. Fiber for big pipes to nodes, wideband in cells and micro cells from there to local distribution, short range micro and pico cells for within a neighborhood or building.
The real problem is how long to we have before the machine intelligences hear all this racket we're making, and come to wipe us out ?
Note that some wireless providers, such as Nextel, want you to use Email to send an alphanumeric message. In the old days you'de use TAP, or TNPP if you were another provider, but now they want Email input from everyone. So, yes, a wardialer style program can spam Nextel users. There's also some TAP hubs that are great inputs, once you know their dial-in number. Once you're there you can just cycle through the set of 10 digits number, noting which get kicked back as "not in service" so you can build a database of how to hit the next time.
A few years ago a local paper ran a page on "how everyday things work" or some such. In it they asked various people how common items worked. A medical technician described TV as "like, sending sound waves through the air, you know? and the TV picks them up and turns them into electricity, and like, shows them to you". This person supposedly has had 2 years of college level education, and is running those lab tests on you; tests that might bear on your health, or say if you've been ingesting substances the government doesn't want you to, or if you have undesireble genes. And yet they're at the level of "I push the round button and wait for the yellow light to go out" - sometimes it makes me think of Cornbluth's "Marching Morons".
A number of years ago, back when for most people "on-line" meant computer BBSes at 1200 bps, I was in a discussion on voter registration. One person wanted people to be able to register at the polls. She said "it's unfair to make people sign up so far in advance, some of my friends are so stoned that they don't even know about an election until election day". So - ignoring the vote fraud problems, you want the president picked by people who are so aware of the issues that they'll vote for the guy with the nicest hair or coolest name?
I think the larger issue is How do we wake more people up? . Wake them up so that they can make resonably informed decisions on important issues. Teaching logic and retoric might help, but it goes beyond schools and into everyday interactions. I don't believe in cutting people out of decision making, but I want them to be able to use reason to make those decisions. The fix for poverty isn't to kill the poor, but to increase the number of well off people; the solution for stupid voters isn't to let fewer vote but to have a greater number of informed voters.
- and I thought that getting older meant that I'd be able to let other people worry about things.
Isn't Jabber basically an "instant messaging" ((aaahhhiiii)) system? BXXP is a full general purpose protocol, IRC/messaging is just one thing it can do well.
The possiblity of sneaking past router filtering, and the enhanced privacy support. Add the peer-to-peer aspect ...
Portability hasn't been a Microsoft concern for a long time.
http://www.cs.mdx.ac.uk/harold/papers/Javaspae.htm l
(and it's intentional that's not a direct hyperlink)
Laws to enforce open communications help, if nothing else they give you a hammer to use when the authorities attempt to obstruct and censor; this may be somewhat effective against other governments through public pressure.
But when it is the government itself that puts the controls in place, usually for "national security", "stopping crime", "enforcing common morality", or "fighting heresy", then the law will be on the side of the blockers.
The companies that own/run the backbone tend to be multinationals. This makes them a little less interersted in local government regs that makes the companies' operations more complicated. But it's not easy to hook directly to them to cut around local routing that may be more annoying to end users; the backbone sells access as tens or hundereds of MB/second. And those big companies are likely to increasing become more tied to media companies, perhaps finding that "revenue lost because of Napster" is important to them
IMHO I think that there needs to be ways to get around the existing `Net. Freenet, encryption in general, real echash, alternative networks for neighborhood level distribution (helps fight traffic analysis), alternative distribution methods that pay the creators of arts better and people who actually support such alternative methods and buy stuff. The user who's sig is "I bought the CD after hearing the MP3" is on the right track. If you can show that the alternatives are fitting into the economy, ie generating jobs and tax revenues, then the government may listed to your argument that your are just replace the buggy whip way of doing business and not stomp on you.
There is signs of a protracted battle in the future, if it happens then "Music should be free and you can't quote my messages in a book" isn't going to cut it.
If you can't code, then document. If you can't do that then learn and spread the ideas.
The pico servers I've seen aren't all that robust, dumping on much loading or anything weird coming their way. Some aren't even doing true http, which causes a few browsers to complain.
A lot more computers meant a much larger market for software. A lot of new companies, targeting the new market for low cost software to match the low cost machines. And some of the mainframe software vendors trying to sell ports of their products for 24X the cost of the PC.
And yes, there was a fair amount of commerical software before 1980. But low-cost was Apple-II, most people weren't shopping for packages for their CDC or DEC machine.
Ah - back when computers were BIG and real programmers could enter programs via the consol switches.
They were testing the hardware drivers, and they were ready ahead before anyone else needed their section.
Source + Spec + Test
I think that might be even better.
Ah, but the government is already using alternative sources of income called lobbiests.
Ah, but the real advantage of RAMBUS will not be apparent until version-II, which will auto-detech open source and free software, adn refuse to store it.