Someone had posted about online banks requesting those last three digits on the back of your credit card.
I looked at Gandi, and they looked good. Totally automated.
I signed up, and the bank they used to handle the credit payment wanted those three digits. http://www.creditmutuel.fr/
They had a little radio button to select "There is no 3 digit code." However, it isn't an actual radio button; It just pops up a window, with a picture of where to find the three numbers. You can't continue without typing some numbers in the box.
So, I called CitiBank. They said the number was internal and I should never be needed, and never given out. (It is apparently on the magnetic strip though, so...) The operator at Citibank thought it was fishy.
I did want to complete the transaction. I just didn't want to give out those three numbers. Mozilla claimed the encryption was valid, and the key proper from Verisign. So I used my credit card, but filled the last 4 numbers of the card in instead of the "verification code." Low and behold it works fine. I didn't try again with a random number.
Before anyone coments, the card is set to be canceled next month.
Do credit cards in Europe have some use for this number? Is this normal for a French bank? They can't have used it during the validation. It didn't even complain that I gave them 4 numbers.
However, if we define this degredation as a necessary property of blackholes, then we've another ball of wax.
If a security measure has a built in time limit, do we count waiting until the limit is over?
This does bring up a discussion of entropy, and the relevance of degredation. We usually think of time as increasing the ability of a secret to be discovered because of two things: The agregation of attacks, and improvements in technology. We think in terms of implementing stronger defences. However, over time things break down on their own. If we're dealing with satelites, then there is a lot of radiation, which can flip a lot of bits. More importantly, it's very hard to swap in new parts.
The thing is that this isn't solely an issue with satelites and similar "distant" resources. The tech mindset is based in a replacement mentality; Supplacement is assumed and possibly necessary for the current growth of our abilities. We need to keep everthing on a time scale. How long does this need to survive? How long might it last? What do we do when it fails? What do we do when it doesn't fail? What do we do with the residue (space junk/computer landfills)?
Time is an issue we rarely incorporate in our designs.
Demanding WinTel ties you to a lot of extra costs. You can't sell a $200 PC that comes with a $100 OS.
Similarly, backward compatibility is a major cost issue. For people with old equipment, doing away with it will cost them. For the manufacturer keeping it adds a lot to the cost, and degrades performance.
If you want small, fast, and cheep you're better off looking at a beefed up embeded system. Note that you can keep Intel, if you want.
If they have piles of cash, why don't they hire a programmer with it? Each company with old 3000's can support one member of the development team, and call it a support contract.
Ever notice that "hyperactive" kids will stare at a screen non-stop for hours? Maybe they'll draw for hours; or read SF, or code. It is only when you point them at a textbook that they have an attention defficit.
ADD people can't pay attention to boring stuff. It's a matter of stimulus level and skill training. Games are easy to progress at so kids become experts and addicts. Usefull skills have a hard initial learning curve but once you get past it they can be as addictive as games.
ADD is only a disorder if you haven't learned to cope with your brain emersed in the world.
> A day to encrypt one doc? Think about this. > I can encrypt a doc in 5 seconds. > For it to take a day, say, half a day, 12 hours, I would have to say... computers would > have to be 8640 times faster today than they were in 1994.
Actually no. Because you can trade some of the time for space. So, assuming that you can design the alg to use all of your space, My 64Meg P233, has >20,000 times the SpaceTime capacity of a C64 with a 10mhz (that's a guess) chip and 64k RAM. I would suspect the Gov can do the equivalent for any hardware commonly available in 1994.
You benchmark a server by setting it up on a network, and connecting many clients to it. The clients all run exactly the same stuff. The clients collect the data. The reason that these tests are expensive is that it takes a lot of clients to load down a large server.
The entire point of benchmarks is to determine if one machine (or OS on the same hardware) is significantly faster than the other. The tests under discusion are run on the same hardware.
On a desktop machine, what works is what you are most comfortable with as a user. On a server what works is not having your clients call you on the phone.
Slashdot Mirror
Someone had posted about online banks requesting those last three digits on the back of your credit card.
I looked at Gandi, and they looked good. Totally automated.
I signed up, and the bank they used to handle the credit payment wanted those three digits. http://www.creditmutuel.fr/
They had a little radio button to select "There is no 3 digit code." However, it isn't an actual radio button; It just pops up a window, with a picture of where to find the three numbers. You can't continue without typing some numbers in the box.
So, I called CitiBank. They said the number was internal and I should never be needed, and never given out. (It is apparently on the magnetic strip though, so...) The operator at Citibank thought it was fishy.
I did want to complete the transaction. I just didn't want to give out those three numbers. Mozilla claimed the encryption was valid, and the key proper from Verisign. So I used my credit card, but filled the last 4 numbers of the card in instead of the "verification code." Low and behold it works fine. I didn't try again with a random number.
Before anyone coments, the card is set to be canceled next month.
Do credit cards in Europe have some use for this number? Is this normal for a French bank? They can't have used it during the validation. It didn't even complain that I gave them 4 numbers.
Any thoughts from the Europeans?
Hurd can be the New Technology that allows Linux to enter the server space....
LinuXP...?
Seriously though, it would a similar transition.
Did I mix my threads? I thought I read earlier:
Which is closer to what most of the users I know have learned. "Just click 'OK'"
Accepted.
However, if we define this degredation as a necessary property of blackholes, then we've another ball of wax.
If a security measure has a built in time limit, do we count waiting until the limit is over?
This does bring up a discussion of entropy, and the relevance of degredation. We usually think of time as increasing the ability of a secret to be discovered because of two things: The agregation of attacks, and improvements in technology. We think in terms of implementing stronger defences. However, over time things break down on their own. If we're dealing with satelites, then there is a lot of radiation, which can flip a lot of bits. More importantly, it's very hard to swap in new parts.
The thing is that this isn't solely an issue with satelites and similar "distant" resources. The tech mindset is based in a replacement mentality; Supplacement is assumed and possibly necessary for the current growth of our abilities. We need to keep everthing on a time scale. How long does this need to survive? How long might it last? What do we do when it fails? What do we do when it doesn't fail? What do we do with the residue (space junk/computer landfills)?
Time is an issue we rarely incorporate in our designs.
It is possible to make a black hole. The ability to "unmake" one is left as a exercise for the student.
Demanding WinTel ties you to a lot of extra costs. You can't sell a $200 PC that comes with a $100 OS.
Similarly, backward compatibility is a major cost issue. For people with old equipment, doing away with it will cost them. For the manufacturer keeping it adds a lot to the cost, and degrades performance.
If you want small, fast, and cheep you're better off looking at a beefed up embeded system. Note that you can keep Intel, if you want.
ROT13 !!!!
Hey if the DMCA says it's crypto....
http://goatse.cx/
I apologize in advance.
If they have piles of cash, why don't they hire a programmer with it? Each company with old 3000's can support one member of the development team, and call it a support contract.
Software support is a human resources issue.
Wouldn't the adhesive barcode obstruct the motherboards thermal sensor, thus increasing the likelyhood of frying the CPU?
How high a court would be needed to order a review of the USPO?
----
Ever notice that "hyperactive" kids will stare at a screen non-stop for hours? Maybe they'll draw for hours; or read SF, or code. It is only when you point them at a textbook that they have an attention defficit.
ADD people can't pay attention to boring stuff. It's a matter of stimulus level and skill training. Games are easy to progress at so kids become experts and addicts. Usefull skills have a hard initial learning curve but once you get past it they can be as addictive as games.
ADD is only a disorder if you haven't learned to cope with your brain emersed in the world.
My ADD is bad, but my hands are worse.
----
> A day to encrypt one doc? Think about this.
> I can encrypt a doc in 5 seconds.
> For it to take a day, say, half a day, 12 hours, I would have to say... computers would
> have to be 8640 times faster today than they were in 1994.
Actually no. Because you can trade some of the
time for space. So, assuming that you can
design the alg to use all of your space,
My 64Meg P233, has >20,000 times the
SpaceTime capacity of a C64 with a 10mhz
(that's a guess) chip and 64k RAM.
I would suspect the Gov can do the equivalent for
any hardware commonly available in 1994.
-sh
----
The files aren't actually that well
encrypted, it just that they stored
the harddrive on top of some speakers
and randomized it.
"Truely complicated noise is indistinguishable
from random information."
-sh
----
MSweb obviously made a typo.
The index page is labled Dec 2, 1998.
The NT4 page is labled Dec 2, 1999.
-sh
Video about rioting on NewYears?
So when does NBC go off the air?
-sh
You benchmark a server by setting it up on a network, and connecting many clients to it. The clients all run exactly the same stuff. The clients collect the data.
The reason that these tests are expensive is that it takes a lot of clients to load down a large server.
The entire point of benchmarks is to determine if one machine (or OS on the same hardware) is significantly faster than the other. The tests under discusion are run on the same hardware.
On a desktop machine, what works is what you are most comfortable with as a user. On a server what works is not having your clients call you on the phone.
-sh