The assumption is that human error is random and will not bias the results. Computer failures tend to be more repeatable and would thus bias the outcome. Couple that with the fact that electronic systems are offer additional avenues for wholesale fraud, and the issue becomes critical.
Go read the writeup on the development of this puppy. What they could have done to the US as of the late '70s is scary. I never really took the propaganda that seriously back then but in hindsight, it was accurate.
I have been using it since v1 on the mac and have used every version for the PC. They did take a *big* step backwards with v5 but they regained most of that with v6. V7 cleaned things up even more. I do, however, only use the vector portion of the program so YMMV.
The CD-R suppliers claim 20-200 years. The upper bound is simply because thay don't know of any real failure mechinisms. The lower bounds is wear and tear of 20 years of careful use. Can someone point us to some *real* data, not just speculations.
A friend who works for a cell phone analysist firm used this yesterday while travelling down the interstate (soemone else driving:-)) and was pulling web pages on her laptop faster than her land line would at work. She even mentioned the 2 Mb figure (yes, bit, not byte).
He claims "Only a select number of terrorist groups..." are of concern. We all know that this is not the case. A couple of host located at disperate locations on the network with a handful of people who can read L0pht and BugTraq are plenty to mount a serious threat to any 'cyber' orginization.
Slashdot Mirror
The assumption is that human error is random and will not bias the results. Computer failures tend to be more repeatable and would thus bias the outcome. Couple that with the fact that electronic systems are offer additional avenues for wholesale fraud, and the issue becomes critical.
Go read the writeup on the development of this puppy. What they could have done to the US as of the late '70s is scary. I never really took the propaganda that seriously back then but in hindsight, it was accurate.
jcl
I know that the Fed. Govt. Agencies can use the DISA team. From those that I have worked with, they have a decent repuatation.
I have been using it since v1 on the mac and have used every version for the PC. They did take a *big* step backwards with v5 but they regained most of that with v6. V7 cleaned things up even more. I do, however, only use the vector portion of the program so YMMV.
The CD-R suppliers claim 20-200 years. The upper bound is simply because thay don't know of any real failure mechinisms. The lower bounds is wear and tear of 20 years of careful use. Can someone point us to some *real* data, not just speculations.
It is actually encouraging that it isn't really hype. Just the facts. It doesn't tell you to or not to use OSS models, just why some people do.
A friend who works for a cell phone analysist firm used this yesterday while travelling down the interstate (soemone else driving :-)) and was pulling web pages on her laptop faster than her land line would at work. She even mentioned the 2 Mb figure (yes, bit, not byte).
He claims "Only a select number of terrorist groups ..." are of concern. We all know that this is not the case. A couple of host located at disperate locations on the network with a handful of people who can read L0pht and BugTraq are plenty to mount a serious threat to any 'cyber' orginization.