The problem with the TSA systems (CAPPS I, CAPPS II, whatever) is that they've been trained using data which is 100% non-terrorists. So they fudge some numbers to make some fraction of them be "terrorirsts" just because otherwise the system would naturally declare everyone not a threat.
Now you can fault the airlines or the government for having accessed all our private information just to train and calibrate the systems, but there's a more fundamental problem: they didn't usefully train or calibrate those systems at all. They just wasted time and money. And they give at least some people a false sense of security when all it really is, is mumbo-jumbo.
In the late 90's a company named "Color Kinetics" was granted a patent on making arbitrary colors by mixing R, G, and B sources. Since then they've gone around driving other companies out of the business.
We all learned from the six million dollar man episode where the Venus probe that crashed in the LA river wreaks destruction here on earth, didn't we? Why can't NASA learn the same lesson?
Actually the article has little to do with technical possibilities/impossibilities but seems to be all about politico-economic posturing. Something notably missing in my post:-).
So what's the point? New technologies (actually spread-spectrum has been around since WWII, and Hedy Lamarr of all people has her name on the patent) make 2.4GHz useful to the average Joe. That's great. But not all services (and therefore not all chunks of spectrum) can follow those rules quite yet. So what are we supposed to do? There's a fundamental tradeoff: let people broadcast more power, then they get more bandwidth, but they crowd out other users. It works exactly the same for spread-spectrum as it does for more traditional modulations.
Remember DARPA's terrorism futures?. This can get controversial sometimes. Actually, it's probably good when it's controversial... putting things into dollars gets around all the policitical hyperbole.
Disclaimers: My PhD advisor was a member of JASON and one of my girlfriends in college was there at the very beginning of the Iowa Electronic Market.
They seem to brag about how large their programs are and how much memory they take.
"All it has to do" (note I put it in quotes) is change a handful of control surfaces to put its GPS coordinates back on the desired flight path. This is something that is even simpler than a old-fashioned mechanical autopilot had to do using some gearboxes and hydraulics (Those generally did inertial navigation, and had to do more work because GPS wasn't available half a century ago...).
Admittedly old autopilots did little more than fly in straight lines, but the problem isn't all that much different.
I see continual dictionary attacks against the domain names I own. When it happens I put the IP address of the spammer in the filter, but this only works for a short while because they're always moving around.
In my limited experience, most of the dictionary attacks come from IP's that traceroute back to Singapore. Just blocking all incoming SMTP from Singapore IP's would be smart but I don't know how to do something like that.
Dell offered Red Hat Linux circa 2001 along with their desktop machines. We bought several dozen machines this way. Then, of course, this option mysteriously disappeared. (I think it may have only ever been available for corporate customers to begin with).
Yes, the new option is different. What I see is not so much "testing" but something being there and then disappearing, and then something different being there and disappearing. I'm not going to count on Dell supporting any particular distro, but I think that it is nice to be able to buy a machine without a Microsoft tax.
Does the Magnusson-Moss act have any relevance here? (This is what lets you do your own oil changes instead of having to take the car to the dealer...) Or does the presence of a "license" contract signed by the customer somehow void these guarantees? Or (GASP!) does the DMCA somehow override the Magnusson-Moss act?
Remember all the companies in the late 90's announcing that backbone bandwidth will be cheaper than air? (QWest, etc.) Here we are in 2004, and Slashdotters reading the press releases seem to assume that their cost for bandwidth to the backbone will be nearly free, yet a T-1 (1.5 Mbps, a tenth of what the article is about) is still realistically several hundred dollars a month. Yeah, if you buy a T-3 or OC-whatever that peers to the Internet then it gets cheaper in volume, but not by a whole lot.
I mean, it's great that we are making progress in bandwidth and reducing cost to get from the phone office to the house, but with connectivity to the backbone still costing as much as it does, do we honestly believe that the effective bandwidth to what we now call "the Internet" backbone will be so cheap that we can ignore it?
I see this as just a way for the phone companies to become another media company and sell the usualy junk on commercial and cable TV, with the phone company now getting some of the profits (where some == "as much as they can gouge the user for").
See, I told you there was no way that a single guy, in just six months, could write an original operating system
(backspace xxxxxxxxx)letter to the US patent office. He clearly pirated his letter from Microsoft's letter! See, Microsft came first! And Linux used the same alphabet as Microsft did! And Microsoft has thousands of lawyers, so how could a single guy do the same thing! In, mind you, just six months!
But Num Lock is where God intended the PF1 (aka Gold) key to go.
Can anyone compete at $27/month ???
on
Becoming a CLEC?
·
· Score: 2, Insightful
Can anyone compete with $27/month broadband
access? Say that's 384kbit or 512kbit...
that's one third of a T1. How can you sell
1/3 of a T1 for that little?
Yes, I know the answer is "oversubscription", and maybe you can justify getting a T3 where you'll get some discount, but still it's a tough business. I have a very hard time seeing how the big boys stay in business, much less the small-timers.
I always thought Sun's only realistic market for "Throughput computing" (maximizing ops/watt) was dense server farms (e.g. blades). Now it is true that this market is playing out quite like everyone (especially Sun) wanted it to, but it is a real market.
For thin-client stuff, while low power consumption is a priority, it's not a big enough one to warrant the amount of money that Sun and others have spent on it. Maybe, just maybe, as a spinoff.
These "find a market for our new processor" discussions are getting a little depressing. I remember being excited about the DEC Alpha for embedded applications, but since then it all feels hollow.
Is grsecurity GPL'ed or not? I always thought
it was, which just means that the guy's involvement and leadership will be shut off, not those of others... it's a pain when the CVS tree and mailing list archives are gone but usually resuming development from a late snapshot isn't too bad. Maybe others had mirrored the CVS tree?
As I understand the patent, the unique thing is that it's for editing sound and video "live" as the stuff goes to media. Doesn't Tivo pretty much do this, with an ability to edit out commercials at recording time? Of course Tivo is generally used to record off-the-air instead of live, but that doesn't seem particularly relevant - an AV source is an AV source, right?
I understand that nothing is foolproof, especially when someone tries malicious tricks.
For example, you could imagine a SCO-wannabe taking their commercial code (that nobody is buying anymore but which they for some reason believe has real IP value), and putting one line (seemingly innocuously, effectively no-op'ed by some never-happens if cases) in an obscure kernel module (maybe a driver for some crufty ancient device). Then repeat (possibly under the guise of a different developer). Soon the module is working, with all the sleeper code inside. Then submit a patch that gets rid of all the intervening lines and voila, a big chunk of
proprietary code is in the kernel and nobody noticed.
There are probably simpler ways to sneak stuff in if you want to be malicious. Maybe I've been watching "The Manchurian Candidate" too often!
Slashdot Mirror
Now you can fault the airlines or the government for having accessed all our private information just to train and calibrate the systems, but there's a more fundamental problem: they didn't usefully train or calibrate those systems at all. They just wasted time and money. And they give at least some people a false sense of security when all it really is, is mumbo-jumbo.
In the late 90's a company named "Color Kinetics" was granted a patent on making arbitrary colors by mixing R, G, and B sources. Since then they've gone around driving other companies out of the business.
We all learned from the six million dollar man episode where the Venus probe that crashed in the LA river wreaks destruction here on earth, didn't we? Why can't NASA learn the same lesson?
Actually the article has little to do with technical possibilities/impossibilities but seems to be all about politico-economic posturing. Something notably missing in my post :-).
So what's the point? New technologies (actually spread-spectrum has been around since WWII, and Hedy Lamarr of all people has her name on the patent) make 2.4GHz useful to the average Joe. That's great. But not all services (and therefore not all chunks of spectrum) can follow those rules quite yet. So what are we supposed to do? There's a fundamental tradeoff: let people broadcast more power, then they get more bandwidth, but they crowd out other users. It works exactly the same for spread-spectrum as it does for more traditional modulations.
Disclaimers: My PhD advisor was a member of JASON and one of my girlfriends in college was there at the very beginning of the Iowa Electronic Market.
So now the whole world is out to get Microsoft. Isn't such paranoia a classic schizophrenic symptom?
"All it has to do" (note I put it in quotes) is change a handful of control surfaces to put its GPS coordinates back on the desired flight path. This is something that is even simpler than a old-fashioned mechanical autopilot had to do using some gearboxes and hydraulics (Those generally did inertial navigation, and had to do more work because GPS wasn't available half a century ago...).
Admittedly old autopilots did little more than fly in straight lines, but the problem isn't all that much different.
First thing you do when reading the article is search it for your name. Doubly true when the article is about statistics of citations...
That said, it's enough (if compressed data) to have the Social Security number of all US Citizens, or all their credit card numbers, etc.
In my limited experience, most of the dictionary attacks come from IP's that traceroute back to Singapore. Just blocking all incoming SMTP from Singapore IP's would be smart but I don't know how to do something like that.
- Everybody has to buy new hardware
- Everybody has to buy all the records all over again on the new media
- Better RIAA/MPAA control of the world's consumers - including the possibility of remotely making stuff unplayable.
. Of course, it's also vaguely possible that a free, standards-based recording medium will come about. (Ha!)Yes, the new option is different. What I see is not so much "testing" but something being there and then disappearing, and then something different being there and disappearing. I'm not going to count on Dell supporting any particular distro, but I think that it is nice to be able to buy a machine without a Microsoft tax.
Does the Magnusson-Moss act have any relevance here? (This is what lets you do your own oil changes instead of having to take the car to the dealer...) Or does the presence of a "license" contract signed by the customer somehow void these guarantees? Or (GASP!) does the DMCA somehow override the Magnusson-Moss act?
My kids browse the web by telnetting to port 80 from my model 33 teletype. What am I supposed to be filtering out?
I mean, it's great that we are making progress in bandwidth and reducing cost to get from the phone office to the house, but with connectivity to the backbone still costing as much as it does, do we honestly believe that the effective bandwidth to what we now call "the Internet" backbone will be so cheap that we can ignore it?
I see this as just a way for the phone companies to become another media company and sell the usualy junk on commercial and cable TV, with the phone company now getting some of the profits (where some == "as much as they can gouge the user for").
Just me being cynical.
But Num Lock is where God intended the PF1 (aka Gold) key to go.
Yes, I know the answer is "oversubscription", and maybe you can justify getting a T3 where you'll get some discount, but still it's a tough business. I have a very hard time seeing how the big boys stay in business, much less the small-timers.
For thin-client stuff, while low power consumption is a priority, it's not a big enough one to warrant the amount of money that Sun and others have spent on it. Maybe, just maybe, as a spinoff.
These "find a market for our new processor" discussions are getting a little depressing. I remember being excited about the DEC Alpha for embedded applications, but since then it all feels hollow.
Is grsecurity GPL'ed or not? I always thought it was, which just means that the guy's involvement and leadership will be shut off, not those of others... it's a pain when the CVS tree and mailing list archives are gone but usually resuming development from a late snapshot isn't too bad. Maybe others had mirrored the CVS tree?
As I understand the patent, the unique thing is that it's for editing sound and video "live" as the stuff goes to media. Doesn't Tivo pretty much do this, with an ability to edit out commercials at recording time? Of course Tivo is generally used to record off-the-air instead of live, but that doesn't seem particularly relevant - an AV source is an AV source, right?
Here is a technical comparison. They seem more alike than different to me.
I know of a few very high-powered geologists who cross-check GPS with GLONASS. Having a third system would seem to only help.
For example, you could imagine a SCO-wannabe taking their commercial code (that nobody is buying anymore but which they for some reason believe has real IP value), and putting one line (seemingly innocuously, effectively no-op'ed by some never-happens if cases) in an obscure kernel module (maybe a driver for some crufty ancient device). Then repeat (possibly under the guise of a different developer). Soon the module is working, with all the sleeper code inside. Then submit a patch that gets rid of all the intervening lines and voila, a big chunk of proprietary code is in the kernel and nobody noticed.
There are probably simpler ways to sneak stuff in if you want to be malicious. Maybe I've been watching "The Manchurian Candidate" too often!