You need to have a standard, and you need to have all the card manufacturers supply an upgrade to their firmware - if the crypto isn't in hardware to keep power consumption down.
But anyway, WEP was broken as designed. There are a dozen ways it sucks. Go to the IEEE 802.11, Task Group I web page (http://grouper.ieee.org/groups/802/11/Reports/tgi _update.htm) and read some of the reports there. Look at "Papers given" under each meeting report.
Task Group I has been working on a replacement for WEP that's "really secure". It's built around using 802.1X as an authentication and key distribution protocl, and AES-OCB as a data encryption/authentication algorithm. So far as I have read, it looks very good..1X allows for session keys and key aging. AES-OCB allows for efficient, one-step encryption and authentication (no need to encrypt and then compute a MAC on the result).
Cracking RC4 should accelerate ratification of this standard, I would think. If you wanted to write a software upgrade, I'd recommend following TGi's progress. And for now don't do anything that requires passwords on your wireless laptop (like log in to a NT domain).
I called into their support number (AT&T Broadband) to ask about this. The guy was a little defensive, but helpful. Perhaps he was surprised to have a clueful caller - I told him I ran Apache under Linux and was not affected by this worm.
He said they couldn't turn on 80 for just me. He claimed that they'd turn on port 80 again for everyone in a few days, but wouldn't commit to a time. They've got to get rid of the worm and virus first.
I asked if they were going to block other ports, for instance, 10000, if I started using that one for a server. Again, somewhat surpised reaction, but he said "no" they wouldn't. Makes sense, the worm uses 80.
So I guess we can all curse MS for this one, and us non-MS users can be just a little smug and annoyed for the moment. And let's hope MS's screwup doesn't ruin the whole game for the rest of us.
Noam Chomsky observed that the right to own property was unique among all others: If I have the right to a piece of property, *you don't*. Contrast this with the right to freedom of speech - my having the right does not diminish someone else's right. Funny how copyright and intellectual property rights land smack in the middle of this.
Personally I don't have a problem with people owning copyrights. The image of a strugging artist expecting to get paid for their efforts seems just to me. The problem for me comes with the current system which so strongly favors the "distribution channel" over the artist.
Yeah, you'd do a lot better with lasers, and it certainly is not the first time someone thought of it. The problem with lasers is (a) rain, and (b) thermals in the atmosphere, which make the beam wander around. This probably happens to RF, too, but since the beamwidth with a laser is so small, it really messes things up. Just ask youself why stars twinkle...
I started with TOPS-20 on the DEC-20 in 1981. Many a night I spent in the WPI WACCC, playing with VT100 escape codes, learning EMACS, and writing programs - in Pascal and PCL. My biggest program was cwplot, which was basically an object file you linked with your own Pascal object containing functions you wanted plotted on the Benson-Varian drum plotter. I plotted quantum wave functions with that thing. Other people used cwplot, too. It used the COMND JSYS, which made it extremely easy to write an interactive command interpreter in Pascal.
God, the effort I went through just to get graphical output. That was in the days when CPU time was accounted for. My advisor dropped me a note in my department mailbox, telling me I had used something like $2300 of CPU time under account "misc" - "I hope there is a good reason for this". Well, there was, I was writing cwplot. The thought of CPU time accounting seems silly now.
But the COMND JSYS was the greatest. It made the command interpreter pretty much self-documenting. When I first encountered Unix, I was extremely annoyed that the "man" command didn't even do that really. It took a year of solid use to finally "get" Unix (and C programming, too), over 5 years after my experience with TOPS-20. tcsh tries to emulate COMND, but essentially fails. How can one go adding documentation for every command on the system? The very thought is ridiculous.
BTW, anyone else remember PCL? You could write clever hacks in that - kind of a scripting language for "exec," the command processor. But you had to run the "exec" in the directory. And PCL docs were something you printed out on the wide line printer and then cut and bound yourself. Had a lot of fun with PCL...
And I still remember a little TECO: "y" would irrevocably erase a buffer in EMACS. Of course, EMACS is permanently wired to my fingers now. I've only had to stop using it when I was stuck with Macs, DOS or Windoze...
My first abortive experience with C was on TOPS-20. A CS guy at WPI had written a compiler for it, and because of the 36 bit word, all the "bytes" were 9 bits! I stuck to Pascal anyway - thought it was the greatest. Didn't learn C until 4 years later.
Thanks for the great article. It explained a lot of little things I never knew, most probably because WPI was cut off from the ARPANET, supposedly because of some transgressions committed before I got there...
I have to say I laughed out loud when I got to the part about the Flourinert turning to gel! In fact, that was the main reason I kept reading anyway. I have never heard of flourinert before, but I do know that there aren't too many things still liquid at LN2 temperatures. Jeeze, guys! Nothing to do, not too smart, too much money and too many toys. An accident waiting to happen.
I agree Jon was talking with the wrong people if he expects WAVE to stop. But every time Pinkerton said "someone else will (fill the gap)", I thought, then Jon or some other spokesperson should bang on their door.
Talking to the customers is the right thing too, but there the argument needs to be taken to every potential customer, in a negative marketing effort.
I don't know - perhaps a billboard campaign that doesn't mention Pinkerton, but is lots more specific than "love your child" messages.
My four-year-old son has this to look forward to? He's already bigger and taller than everyone else his age, and he's smart, too. He also likes to "play rough." Jeeze, he fits the profile already.
It's really kind of amusing watching people champion these various protocols as if they were their favorite sports teams. Amusing at first, then kind of nauseating.
Next, Firewire is a bus protocol, Ethernet is a network protocol. Ethernet vs. Firewire is like saying Ferrari vs. Mac Truck. Even Firewire vs. USB is a rather silly, although I'm sure Intel sees it as a contest.
The next thing I have a problem with is this urge to turn all interconnect busses into a home networking solution. 150 foot USB! 10 meter wireless IDE! Yippee!
Then again, with Firewire, it does actually make sense. It would be so cool to have a house wired for Firewire. Then, I would need only one sat receiver or cable box downstairs in the den and simply hook up my bedroom (or outdoor or basement or kitchen) TV to the Firewire connection and we'd be in business. With the right accessories, you could control channels through Firewire too.
Obviously, if two people wanted to watch different things at the same time, there'd be a problem. But the solution now is to rent another cable or sat box (big bucks), or carry the damn thing to the bedroom and hook it up there.
Think also what it could do for home surveilliance. Put your little vidcams everywhere and hook them up to Firewire. Use the monitor to switch to different views. Heck, maybe even tell the cameras to deliver lots of little low-res views so the bandwidth doesn't saturate.
And since Firewire is designed for this, it would be far superior to Ethernet even if there were a 400 Mbps version. Firewire delivers the Quality of Service needed. There might even be a market for Ethernet to Firewire bridges, so you could use the Firewire cable to network the house for computers, or to hook Firewire up to your old cable modem.
It could be the beginning of a new (excuse me) paradigm. As a bus gets better at working long distances, maybe it will supplant the LAN in some settings. And Firewire is great because you don't need a computer in the middle of it. Making Firewire work over these long enough distances is the only thing that makes sense. Then someday there will be wireless...
Useful list, but you never mentioned SRP! Anyway, last I read about SRP it was only a means for secure authentication and had nothing with encrypting the link. SSH, of course, does a whole lot more, like providing secure host-to-host TCP tunnels, in addition to secure authentication. The key improvement of SRP was that compromise of the remote host's "password" database would be of no use to the thief.
The SETI people have explained their position as well as they need to, IMHO. The post you are replying to explains it again. And as a person who has a Ph.D. in spectral estimation and 20 years in programming (ie, ``smarter than the average bear''), I have contributed 1500 results and have no intention of leaving.
Re:Simon Singh was on the connection
on
The Code Book
·
· Score: 1
I heard that Connection show. So much of a big deal about Enigma. Everyone wants to know about Enigma. Well, it was a big deal, and much talked about. But no mention is ever made about the codes used on the Allied side. Having recently read Leo Marks' fascinating and entertaining memoir, "Between Silk and Cyanide: A Codemaker's War," I was sensitized to this. Check it out - it's a good read. Mr. Marks made his own contribution toward shortening the war by three months.
Over 5 years ago I was de-facto sysadmin for an IBM RS-6000. There was a program on it which would do just what you're talking about. I was very impressed, because this was BTW (Before The Web). You clicked a button and it would download the bug fix database or whatever, and you could select pertinant fixes for your machine. Very slick. Should be even easier in this web-enabled day and age.
Lots of people responding who don't want to see Linux cater to the masses seem to have forgotten something: If you don't want it, don't install it and don't use it!
There are plenty of distributions for experts, and since making a version of Linux that's luser-friendly will most likely mean an X overlay that hides the command shell, why worry that the core Linux will become dumbed-down? Things like WinLinux 2000 aren't there to replace e.g., Debian, so relax!
Better yet, go join one of those luser-friendly companies (or form your own) and make a friendly Linux, and make a lot of money off it! Except for the part about lots of money perhaps, that's what has been going on...
I've been with Mediaone for 2.5 years now - was probably one of the first 5 to have it in my town (Winchester, MA). They were Continental Cablevision then, and the service was Highway1. Booted up my box as Linux as soon as the installers left.
I'd just like to confirm how good the service seems to be around here. And, they've been pretty receptive about Linux, which is quite a difference from regular telephone ISP. Perhaps things are better now, but I could never figure out how to get all those "chap secrets" working for PPP... DHCP is so much easier.
I won't say they support Linux, because they say they don't. The good thing is that, instead of meaning Linux is prohibited, they just mean, don't ask us for help. They even set up a local newsgroup for Linux users.
This even goes so far as being smart enough not to prohibit Linux just because it can be used for serious network abuse (web proxy, mail gateway, etc. etc.). They recognize that Windows has problems too, and have set up a machine which runs around doing its own port scans to look for abuse. I don't know what happens when they detect it, because my system has never been a problem (up 24/7).
I have had service problems occasionally, but that seems to have been because of a slowly failing cable modem. Since I was leasing one of theirs, they did replace it when it finally failed. Funny thing was the service guy who came was the same one who installed it originally. Was nice to see that kind of continuity.
Also, the people on phone support are a bit lame, as I've found on most phone support lines. Think about it: if they were as smart as you think you are, would they be doing that job? You had better hope not, for your own sake. The best advice for dealing with them is to only call them after you've diagnosed the problem and determined it to be outside your computer. And then, run everything they say through a Windows->Linux translator.
All in all, I think the folks running the system here have done an excellent job emplacing a system that people have been using constantly, and enhancing it as they go along. In spite of the reputation cable companies have for having poor service, I would be hard pressed to switch to a different kind of service. I just hope that the merger deal with AT&T doesn't change this.
Not only is this a bad example because of the nature of the trial, but because of the nature of the information. Why do we need multimillion-dollar satellites to create a computer rendering of the scene? Why not just go over there in another airplane and take a picture? And why should anyone trust a computer rendering over a regular photograph?
This kind of justification triviallizes the whole technology, (cynic mode on) but I guess that's just the sort of justification we need to get the people with the money to buy into hi-res satellites (cynic mode off).
The more I read and hear about the PPC roadmap, the more I drool.
The G4, with the AltiVec additions, makes a fantastic DSP development platform. In case you didn't realize, Altivec supposedly completes a multiply-accumulate every clock cycle minumum (It can actually do 4 per clock cycle because of the SIMD architecture). Doing a single-cycle MAC is almost the definition of a DSP. But to really take good advantage of the DSP aspect, you need an honest-to-goodness RTOS.
Now last I looked, TI's 'C6x architecture was headed that way with speed and scalability, but no one is ever going to write or port a general purpose OS (ie, Linux, Windoze, MacOS, Be) for/to that chip.
The G4, however, already runs at least MacOS and Linux. Unfortunately, neither of these are real-time systems. (Has RTLinux been ported to PPC? A great project, if not!) And it's too bad that the situation with Be prevents them from taking advantage of this great chip.
Anyway, I guess my point is just that having a hosted (versus embedded) DSP development system would be just the coolest, and that the G4 makes this possible for wont of a RTOS.
I have not seen anything yet that state explicitly that the G5 will have AltiVec on it. God, I hope so, though.
Would we have the H-bomb without Teller? I think certainly yes. Most likely not a quickly as we did. Would the Russians have had it first? Don't know... Although that thought does frighten me, just like the thought of the Nazis having the A-bomb first.
The thing is that just about any immoral/self-destructing/bad-for-the-world scheme can be justified when there's fear involved. Ideas like Project Chariot (or draining the Mediterranean to make farmland, or... the list goes on) all have one thing in common: they're boldly outlandish, potentially helpful to humans, but we don't fear what might happen if we don't do them.
The H-bomb was invented because of our fear of Communism. If the thought of fusion never occurred (which is unlikely), then it might just as well have been a giant A-bomb on giant rockets. Think of all the other outlandish ideas which failed. H-bombs just happened to work.
Disclaimer: I don't agree with Teller's politics at all; I'm just trying to make some observations of the world in general.
This is no big deal. Program a computer to generate ads (or poems, or Irish jigs, or images) according to certain rules, and assuming you did a good job programming, the output will always adhere to the rules.
The really big thing missing is that humans evaluated the output to decide what was ``good'' - not the computers. It's no different from feeding words into an anagram generator and only choosing the anagrams that you like.
On the other hand, if they actually had something similar to a chess or backgammon playing program (generates possibilities and evaluates their merit), then we might be in trouble.
First off, I hope this moderation and karma scheme works. I'm kind of afraid that my posts would never make it above 1, just because I don't post every day or even every week.
Secondly, you can do all you want to the ACs, but if they start generating slews of new user IDs, the best way to prevent them from doing their dirty work would be a waiting period before new accounts were allowed to post. Like 3 to 5 days. Like every other solution, this one has its drawbacks, too.
It's just a little funny to see a/. item about head transplants and Einstein's HEAD right next to it. How about THAT head transplant? Now if we could just find Marilyn Monroe's body...
1) With regard to link reliability, we really need to see the link budget and the % availability they expect to achieve at 5 km. And just because it's raining, snowing or foggy doesn't mean the system is useless. Attenuation is not binary, and athough the link may go away at 5 km, it still may be useful at 500 m, because there's 20 dB more power in the link budget. I can imagine a semi-permanent link between buildings in NYC, for instance, that would expect and get 99.9% availability with this sort of link, and that might be just fine.
Birds are not so much a problem because if it's important, a TCP-like net connection is being used, and retransmissions will occur. Now a whole flock of birds, well...
2) Encryption? Well, it is hard to intercept because you'd have to be in the line-of-sight. Now if you're between the Tx and Rx, you'd probably have a good intercept, but you might ruin the link for the legit user. If you were behind the legit Rx, then that receiver would be blocking you, and you as well might be out of range for the link.
Still, if it's important, the user(s) that need encryption will do so as necessary on their connection(s) only.
3) No good for anything but line-of-sight (LOS)? This is still a big market for data carriers. There's bandwidth all over the place between 2 GHz up to 38 GHz for point-to-point use, and these pretty much have to be LOS-only. At around 28 GHz, there's LMDS, which is point-to-multipoint; still, it's LOS-only (in spite of what some might say).
This laser solution is clearly LOS-only, and will require proper aiming and all that at each end. And, it isn't very mobile, but Navy ships could certainly afford the required autotrack mechanism to make this useful even with gentle rocking of the ship.
Shafik, you beat me to it. So I'll just add that for those 240 million sheep out there, it might be an easier intro to Chomsky to watch the documentary ``Manufacturing Consent.'' Check out this for some online info about the film. If ten people read this message and see the film, I'll be very happy. (If a million do, I'll consider that proof that the time of rapture is near)
Heck, it might already be answered in an FAQ. Every time I check out news on comp.os.linux.hardware, there's some thead about it.
Anyway, it depends on how much memory you have in your machine and what you plan do run on it. Here's my rule of thumb. For 8 MB up to 64 MB RAM, make a swap partition twice the size of the RAM. For larger RAM, just make swap 127 MB, since it's the limit for a single partition. If you have more than that amount of RAM, 127 MB swap is still fine - remember, you've got lots of RAM.
(One person's opinion was just go ahead and start off with 127 MB swap no matter how much RAM you have because it's too hard to increase the size after it's been created. I wish I had heard this before I configured my machine.)
However, if you really need to have several LARGE programs running at the same time (Netscape and StarOffice and KDE and... a few other big things) then you need more RAM than 64 MB (try 192 for starters). If you really need all that swap, adding more is only going to slow things down more, even if it's on another disk - after all, a drive is slower than RAM...
Still, I'd like to know myself what rules of thumb there are for large memory machines (256 MB and higher). I can't imagine the rules for swap in small RAM machines continues to hold.
Slashdot Mirror
But anyway, WEP was broken as designed. There are a dozen ways it sucks. Go to the IEEE 802.11, Task Group I web page (http://grouper.ieee.org/groups/802/11/Reports/tgi _update.htm) and read some of the reports there. Look at "Papers given" under each meeting report.
Task Group I has been working on a replacement for WEP that's "really secure". It's built around using 802.1X as an authentication and key distribution protocl, and AES-OCB as a data encryption/authentication algorithm. So far as I have read, it looks very good. .1X allows for session keys and key aging. AES-OCB allows for efficient, one-step encryption and authentication (no need to encrypt and then compute a MAC on the result).
Cracking RC4 should accelerate ratification of this standard, I would think. If you wanted to write a software upgrade, I'd recommend following TGi's progress. And for now don't do anything that requires passwords on your wireless laptop (like log in to a NT domain).
He said they couldn't turn on 80 for just me. He claimed that they'd turn on port 80 again for everyone in a few days, but wouldn't commit to a time. They've got to get rid of the worm and virus first.
I asked if they were going to block other ports, for instance, 10000, if I started using that one for a server. Again, somewhat surpised reaction, but he said "no" they wouldn't. Makes sense, the worm uses 80.
So I guess we can all curse MS for this one, and us non-MS users can be just a little smug and annoyed for the moment. And let's hope MS's screwup doesn't ruin the whole game for the rest of us.
Don't forget - IBM has the record for the most patent one year. How many more are like this one???
Personally I don't have a problem with people owning copyrights. The image of a strugging artist expecting to get paid for their efforts seems just to me. The problem for me comes with the current system which so strongly favors the "distribution channel" over the artist.
Yeah, you'd do a lot better with lasers, and it certainly is not the first time someone thought of it. The problem with lasers is (a) rain, and (b) thermals in the atmosphere, which make the beam wander around. This probably happens to RF, too, but since the beamwidth with a laser is so small, it really messes things up. Just ask youself why stars twinkle...
God, the effort I went through just to get graphical output. That was in the days when CPU time was accounted for. My advisor dropped me a note in my department mailbox, telling me I had used something like $2300 of CPU time under account "misc" - "I hope there is a good reason for this". Well, there was, I was writing cwplot. The thought of CPU time accounting seems silly now.
But the COMND JSYS was the greatest. It made the command interpreter pretty much self-documenting. When I first encountered Unix, I was extremely annoyed that the "man" command didn't even do that really. It took a year of solid use to finally "get" Unix (and C programming, too), over 5 years after my experience with TOPS-20. tcsh tries to emulate COMND, but essentially fails. How can one go adding documentation for every command on the system? The very thought is ridiculous.
BTW, anyone else remember PCL? You could write clever hacks in that - kind of a scripting language for "exec," the command processor. But you had to run the "exec" in the directory. And PCL docs were something you printed out on the wide line printer and then cut and bound yourself. Had a lot of fun with PCL...
And I still remember a little TECO: "y" would irrevocably erase a buffer in EMACS. Of course, EMACS is permanently wired to my fingers now. I've only had to stop using it when I was stuck with Macs, DOS or Windoze...
My first abortive experience with C was on TOPS-20. A CS guy at WPI had written a compiler for it, and because of the 36 bit word, all the "bytes" were 9 bits! I stuck to Pascal anyway - thought it was the greatest. Didn't learn C until 4 years later.
Thanks for the great article. It explained a lot of little things I never knew, most probably because WPI was cut off from the ARPANET, supposedly because of some transgressions committed before I got there...
I have to say I laughed out loud when I got to the part about the Flourinert turning to gel! In fact, that was the main reason I kept reading anyway. I have never heard of flourinert before, but I do know that there aren't too many things still liquid at LN2 temperatures. Jeeze, guys! Nothing to do, not too smart, too much money and too many toys. An accident waiting to happen.
Talking to the customers is the right thing too, but there the argument needs to be taken to every potential customer, in a negative marketing effort.
I don't know - perhaps a billboard campaign that doesn't mention Pinkerton, but is lots more specific than "love your child" messages.
My four-year-old son has this to look forward to? He's already bigger and taller than everyone else his age, and he's smart, too. He also likes to "play rough." Jeeze, he fits the profile already.
Next, Firewire is a bus protocol, Ethernet is a network protocol. Ethernet vs. Firewire is like saying Ferrari vs. Mac Truck. Even Firewire vs. USB is a rather silly, although I'm sure Intel sees it as a contest.
The next thing I have a problem with is this urge to turn all interconnect busses into a home networking solution. 150 foot USB! 10 meter wireless IDE! Yippee!
Then again, with Firewire, it does actually make sense. It would be so cool to have a house wired for Firewire. Then, I would need only one sat receiver or cable box downstairs in the den and simply hook up my bedroom (or outdoor or basement or kitchen) TV to the Firewire connection and we'd be in business. With the right accessories, you could control channels through Firewire too.
Obviously, if two people wanted to watch different things at the same time, there'd be a problem. But the solution now is to rent another cable or sat box (big bucks), or carry the damn thing to the bedroom and hook it up there.
Think also what it could do for home surveilliance. Put your little vidcams everywhere and hook them up to Firewire. Use the monitor to switch to different views. Heck, maybe even tell the cameras to deliver lots of little low-res views so the bandwidth doesn't saturate.
And since Firewire is designed for this, it would be far superior to Ethernet even if there were a 400 Mbps version. Firewire delivers the Quality of Service needed. There might even be a market for Ethernet to Firewire bridges, so you could use the Firewire cable to network the house for computers, or to hook Firewire up to your old cable modem.
It could be the beginning of a new (excuse me) paradigm. As a bus gets better at working long distances, maybe it will supplant the LAN in some settings. And Firewire is great because you don't need a computer in the middle of it. Making Firewire work over these long enough distances is the only thing that makes sense. Then someday there will be wireless...
Useful list, but you never mentioned SRP! Anyway, last I read about SRP it was only a means for secure authentication and had nothing with encrypting the link. SSH, of course, does a whole lot more, like providing secure host-to-host TCP tunnels, in addition to secure authentication. The key improvement of SRP was that compromise of the remote host's "password" database would be of no use to the thief.
The SETI people have explained their position as well as they need to, IMHO. The post you are replying to explains it again. And as a person who has a Ph.D. in spectral estimation and 20 years in programming (ie, ``smarter than the average bear''), I have contributed 1500 results and have no intention of leaving.
I heard that Connection show. So much of a big deal about Enigma. Everyone wants to know about Enigma. Well, it was a big deal, and much talked about. But no mention is ever made about the codes used on the Allied side. Having recently read Leo Marks' fascinating and entertaining memoir, "Between Silk and Cyanide: A Codemaker's War," I was sensitized to this. Check it out - it's a good read. Mr. Marks made his own contribution toward shortening the war by three months.
Over 5 years ago I was de-facto sysadmin for an IBM RS-6000. There was a program on it which would do just what you're talking about. I was very impressed, because this was BTW (Before The Web). You clicked a button and it would download the bug fix database or whatever, and you could select pertinant fixes for your machine. Very slick. Should be even easier in this web-enabled day and age.
Lots of people responding who don't want to see Linux cater to the masses seem to have forgotten something: If you don't want it, don't install it and don't use it!
There are plenty of distributions for experts, and since making a version of Linux that's luser-friendly will most likely mean an X overlay that hides the command shell, why worry that the core Linux will become dumbed-down? Things like WinLinux 2000 aren't there to replace e.g., Debian, so relax!
Better yet, go join one of those luser-friendly companies (or form your own) and make a friendly Linux, and make a lot of money off it! Except for the part about lots of money perhaps, that's what has been going on...
I'd just like to confirm how good the service seems to be around here. And, they've been pretty receptive about Linux, which is quite a difference from regular telephone ISP. Perhaps things are better now, but I could never figure out how to get all those "chap secrets" working for PPP... DHCP is so much easier.
I won't say they support Linux, because they say they don't. The good thing is that, instead of meaning Linux is prohibited, they just mean, don't ask us for help. They even set up a local newsgroup for Linux users.
This even goes so far as being smart enough not to prohibit Linux just because it can be used for serious network abuse (web proxy, mail gateway, etc. etc.). They recognize that Windows has problems too, and have set up a machine which runs around doing its own port scans to look for abuse. I don't know what happens when they detect it, because my system has never been a problem (up 24/7).
I have had service problems occasionally, but that seems to have been because of a slowly failing cable modem. Since I was leasing one of theirs, they did replace it when it finally failed. Funny thing was the service guy who came was the same one who installed it originally. Was nice to see that kind of continuity.
Also, the people on phone support are a bit lame, as I've found on most phone support lines. Think about it: if they were as smart as you think you are, would they be doing that job? You had better hope not, for your own sake. The best advice for dealing with them is to only call them after you've diagnosed the problem and determined it to be outside your computer. And then, run everything they say through a Windows->Linux translator.
All in all, I think the folks running the system here have done an excellent job emplacing a system that people have been using constantly, and enhancing it as they go along. In spite of the reputation cable companies have for having poor service, I would be hard pressed to switch to a different kind of service. I just hope that the merger deal with AT&T doesn't change this.
This kind of justification triviallizes the whole technology, (cynic mode on) but I guess that's just the sort of justification we need to get the people with the money to buy into hi-res satellites (cynic mode off).
The G4, with the AltiVec additions, makes a fantastic DSP development platform. In case you didn't realize, Altivec supposedly completes a multiply-accumulate every clock cycle minumum (It can actually do 4 per clock cycle because of the SIMD architecture). Doing a single-cycle MAC is almost the definition of a DSP. But to really take good advantage of the DSP aspect, you need an honest-to-goodness RTOS.
Now last I looked, TI's 'C6x architecture was headed that way with speed and scalability, but no one is ever going to write or port a general purpose OS (ie, Linux, Windoze, MacOS, Be) for/to that chip.
The G4, however, already runs at least MacOS and Linux. Unfortunately, neither of these are real-time systems. (Has RTLinux been ported to PPC? A great project, if not!) And it's too bad that the situation with Be prevents them from taking advantage of this great chip.
Anyway, I guess my point is just that having a hosted (versus embedded) DSP development system would be just the coolest, and that the G4 makes this possible for wont of a RTOS.
I have not seen anything yet that state explicitly that the G5 will have AltiVec on it. God, I hope so, though.
(drool-mode off)
The thing is that just about any immoral/self-destructing/bad-for-the-world scheme can be justified when there's fear involved. Ideas like Project Chariot (or draining the Mediterranean to make farmland, or... the list goes on) all have one thing in common: they're boldly outlandish, potentially helpful to humans, but we don't fear what might happen if we don't do them.
The H-bomb was invented because of our fear of Communism. If the thought of fusion never occurred (which is unlikely), then it might just as well have been a giant A-bomb on giant rockets. Think of all the other outlandish ideas which failed. H-bombs just happened to work.
Disclaimer: I don't agree with Teller's politics at all; I'm just trying to make some observations of the world in general.
The really big thing missing is that humans evaluated the output to decide what was ``good'' - not the computers. It's no different from feeding words into an anagram generator and only choosing the anagrams that you like.
On the other hand, if they actually had something similar to a chess or backgammon playing program (generates possibilities and evaluates their merit), then we might be in trouble.
Secondly, you can do all you want to the ACs, but if they start generating slews of new user IDs, the best way to prevent them from doing their dirty work would be a waiting period before new accounts were allowed to post. Like 3 to 5 days. Like every other solution, this one has its drawbacks, too.
It's just a little funny to see a /. item about head transplants and Einstein's HEAD right next to it. How about THAT head transplant? Now if we could just find Marilyn Monroe's body...
Birds are not so much a problem because if it's important, a TCP-like net connection is being used, and retransmissions will occur. Now a whole flock of birds, well...
2) Encryption? Well, it is hard to intercept because you'd have to be in the line-of-sight. Now if you're between the Tx and Rx, you'd probably have a good intercept, but you might ruin the link for the legit user. If you were behind the legit Rx, then that receiver would be blocking you, and you as well might be out of range for the link.
Still, if it's important, the user(s) that need encryption will do so as necessary on their connection(s) only.
3) No good for anything but line-of-sight (LOS)? This is still a big market for data carriers. There's bandwidth all over the place between 2 GHz up to 38 GHz for point-to-point use, and these pretty much have to be LOS-only. At around 28 GHz, there's LMDS, which is point-to-multipoint; still, it's LOS-only (in spite of what some might say).
This laser solution is clearly LOS-only, and will require proper aiming and all that at each end. And, it isn't very mobile, but Navy ships could certainly afford the required autotrack mechanism to make this useful even with gentle rocking of the ship.
Yes, I do this for a living, only at RF.
Shafik, you beat me to it. So I'll just add that for those 240 million sheep out there, it might be an easier intro to Chomsky to watch the documentary ``Manufacturing Consent.'' Check out this for some online info about the film. If ten people read this message and see the film, I'll be very happy. (If a million do, I'll consider that proof that the time of rapture is near)
Anyway, it depends on how much memory you have in your machine and what you plan do run on it. Here's my rule of thumb. For 8 MB up to 64 MB RAM, make a swap partition twice the size of the RAM. For larger RAM, just make swap 127 MB, since it's the limit for a single partition. If you have more than that amount of RAM, 127 MB swap is still fine - remember, you've got lots of RAM.
(One person's opinion was just go ahead and start off with 127 MB swap no matter how much RAM you have because it's too hard to increase the size after it's been created. I wish I had heard this before I configured my machine.)
However, if you really need to have several LARGE programs running at the same time (Netscape and StarOffice and KDE and... a few other big things) then you need more RAM than 64 MB (try 192 for starters). If you really need all that swap, adding more is only going to slow things down more, even if it's on another disk - after all, a drive is slower than RAM...
Still, I'd like to know myself what rules of thumb there are for large memory machines (256 MB and higher). I can't imagine the rules for swap in small RAM machines continues to hold.