News for hackers. Stuff that matters.
Re:PIN number analysis by furyg3 on Sunday January 27, 2013 @08:13AM (#5124122)
My bank issues a PIN which is not of my choosing, and since it's a smart card to change it means actually going down to the bank to select a new PIN. I think the majority of people never do this, so it seems like a great way to counter people selecting "1234" or their birthday.

At some point I'd really like to see six to eight digit PINs, though...

Re:PIN number analysis by lifeisstillgood on Sunday January 27, 2013 @12:34PM (#5124880)
Apparently (urban myth time) the bank issued pin is a hash of your bank account number plus some other details that are known without an expensive core database lookup.

This way your pin can be verified without needing to call back to the central bank core - the ATM cloud can just move straight on to transactions.

It may save a lot of cash it may be a myth - anyone got details?

Re:PIN number analysis by itcmcgrath on Sunday January 27, 2013 @06:24PM (#5126157)
I worked at a bank, I did code for generating PINs. I can tell you the urban myth is incorrect. Perhaps a single bank did it, but it sure isn't industry standard.
Re:PIN number analysis by mynameishere on Sunday January 27, 2013 @09:26PM (#5126752)
Did you filter out terrible combinations like 1111?
Re:PIN number analysis by itcmcgrath on Monday January 28, 2013 @12:22PM (#5129418)
We did some sensible things to prevent non-sensible things. I don't think talking about the secret sauce would be appropriate though.
Re:PIN number analysis by StavrosK on Sunday January 27, 2013 @12:41PM (#5124907)
The fact that you can change it disproves it.
Re:PIN number analysis by cjg on Sunday January 27, 2013 @03:28PM (#5125515)
Re:PIN number analysis by benvd on Sunday January 27, 2013 @03:14PM (#5125451)
If enough people never bother to change their PIN, it might still be worth it.
Re:PIN number analysis by StavrosK on Sunday January 27, 2013 @03:31PM (#5125532)
Yeah, but then you have to make the lookup anyway, to check if they did change it. You might as well just authenticate. Unless the original PIN always works, which is not secure.
Re:PIN number analysis by PanMan on Sunday January 27, 2013 @04:10PM (#5125676)
I think the changed PIN is stored on your card as an offset. When you enter your pin, both the offset and the code you entered are transmitted, so they always validate agains the original, non changed, PIN.
Re:PIN number analysis by superuser2 on Sunday January 27, 2013 @08:00PM (#5126489)
My ATM won't complain about an invalid pin until it actually communicates with the bank, at the very end of the transaction (after choosing checking/savings and the amount).
Re:PIN number analysis by gedrap on Monday January 28, 2013 @10:29AM (#5128840)
Every ATM in the UK I've used in 1.5 years did that. Makes sense. Why send 2 requests (confirm PIN -> initiate transaction) if you can combine it into one without losing anything?
Re:PIN number analysis by davedx on Sunday January 27, 2013 @09:32AM (#5124365)
Same here in the Netherlands. I don't think I can change mine at all.

This has however resulted in me forgetting it several times, once I almost got stranded at a station and was lucky to have enough cash to buy a ticket.

I won't tell you what I've done to mitigate against being stranded at random places due to forgetting my PIN but it's not secure. Therefore, I have my reservations about forcing people to remember random numbers.

Re:PIN number analysis by barking on Sunday January 27, 2013 @09:39AM (#5124386)
So I have 1 secret 4 digit number that I use to encrypt all my other pin numbers by adding it to them using modulo 10 addition. I then write the resultant 4 digit pin on the back of the card. I've been doing this for well over a decade since seeing it in a book about encryption. Without it I'd have needed to have been issued with new pin numbers many times
Re:PIN number analysis by scoot on Sunday January 27, 2013 @05:10PM (#5125877)
Interesting. If you chose a pin for your card that was the result of modulo arithmetic with some part of the card number (last four digits, for example, but perhaps something more obscure), you would have to write on the card at all.
Re:PIN number analysis by barking on Sunday January 27, 2013 @05:18PM (#5125903)
That's a very clever suggestion.

The only thing is that if you're going to pick the pin you might as well use the secret one.

Re:PIN number analysis by naftaliharris on Sunday January 27, 2013 @05:49PM (#5126013)
You probably wouldn't want to do that: If the pin was somehow compromised, then the attacker would likely try the same compromised pin number in other places. Using different pins for each service avoids this attack.
Re:PIN number analysis by anonymous on Sunday January 27, 2013 @03:15PM (#5125457)
Ooh, that's neat. I use a similar system, though mine permits several possible ways to encrypt a number and I usually pick an encryption with 5 or 6 digits. It doesn't require me to remember anything but the algorithm, so I don't feel comfortable with sharing it. I know it might be less secure than encrypting the pin with another pin, but I can't trust myself to remember any pin.
Re:PIN number analysis by lifeisstillgood on Sunday January 27, 2013 @12:30PM (#5124868)
Errr ... You have a secret pin.

You modulo that with the pin your bank gives you.

You write the result on the back of the card.

If you forget the pin you can work backwards to the bank issued pin?

Ok I get it now. Sounds crazy.

Re:PIN number analysis by arrrg on Sunday January 27, 2013 @10:10AM (#5124458)
My bank (in Germany) just recently introduced a feature on their ATMs that allows you to easily change your PIN.

I think it's pretty obvious that's a bad idea, given all the huge posters also hanging everywhere in the bank, sternly warning people not to pick a dumb PIN. I'm sure that'll work.

4-digit PINs are easy to remember. It's real simple to always come up with some sort of mnemonic. There is no need whatsoever to pick your own.

Re:PIN number analysis by nmcfarl on Sunday January 27, 2013 @12:53PM (#5124960)
I disagree - with all my personal and business accounts I have around 15 pins. I of course have to write the down.
Re:PIN number analysis by scoot on Sunday January 27, 2013 @05:05PM (#5125850)
"it seems like a great way to counter people selecting "1234" or their birthday."

It would be, were it not for the fact that the pin is randomly assigned. I've had a card issued with "2468" as the pin.

"I'd really like to see six to eight digit PINs, though..."

Whilst the default length is 4 digits, you can change it to something longer if you choose.

Re:PIN number analysis by cryptoz on Sunday January 27, 2013 @06:27PM (#5126166)
> Whilst the default length is 4 digits, you can change it to something longer if you choose.

Not really. You'll break international standards if you do. I used an 8-digit PIN for many years, but it is 100% impossible to use anywhere I tried in Europe. You will be completely without any way to pay other than cash. No machine will allow more than 4 digits in my experience.

Re:PIN number analysis by tripzilch on Monday January 28, 2013 @10:21AM (#5128805)
How long ago was your experience? This may explain why (over the past 5-10 years or so) I've encountered more and more machines that require to press OK after entering the four digits (before, some machines would submit after you type the fourth, without confirmation).
Re:PIN number analysis by flux_w42 on Monday January 28, 2013 @02:27AM (#5127565)
Here in Belgium they do
Re:PIN number analysis by kamkha on Sunday January 27, 2013 @01:06PM (#5124999)
My Bank of America account has a six digit PIN, and it has worked at every ATM (BofA or not) I've encountered.
Re:PIN number analysis by LogicX on Sunday January 27, 2013 @10:33PM (#5126974)
Mine is a 10-digit pin, of my choosing.

When dealing with 'foreign' ATMs (whether domestic or foreign), my pin is sometimes truncated -- anywhere from 8 to 4 digits.

I've also come across ATMs where it won't work.

Bank of America advises you to change it to a 4-digit PIN before traveling internationally (and actually have a paper handout they'll give at the teller window saying such if you tell them you're traveling internationally.

Re:PIN number analysis by alanctgardner2 on Sunday January 27, 2013 @09:15AM (#5124322)
You can do up to 6 in Canada. Nobody ever told me, but when I was signing up at RBC I accidentally typed 5 and it worked. Now I know why they don't deterministically say 'OK, he's done' after you hit four digits.
Re:PIN number analysis by 3pt14159 on Sunday January 27, 2013 @12:13PM (#5124811)
Watch out! I also live in Canada, and a close friend of mine while travelling was frozen out of his accounts because there were no banks that could handle 6 digit pins there.

For that reason I recommend having a 4 digit pin.

Re:PIN number analysis by dbaupp on Sunday January 27, 2013 @01:38PM (#5125094)
Where's "there"? (Presumably not Canada?)
Re:PIN number analysis by derpmaster on Sunday January 27, 2013 @04:17PM (#5125700)
I couldn't use a 6 digit pin in Italy. Some ATMs in London,UK didn't like it either.

Singapore it's mandatory for a 6 digit pin so have no idea if a 4 digit pin will work, anybody been there?

Re:PIN number analysis by lvturner on Sunday January 27, 2013 @10:38PM (#5126993)
In China, 6 digit pins seem to be the norm - sometimes I can't take money out on my international card because my 4 digit pin is deemed by the machine as too short/invalid.
Re:PIN number analysis by rodly on Sunday January 27, 2013 @09:29PM (#5126761)
I had to get a new debit card relatively recently (was robbed) and it came with a 6-digit PIN not of my choosing (Bank of America). Just saying, they now exist.
Re:PIN number analysis by nwh on Sunday January 27, 2013 @08:16AM (#5124129)
They allow them (8+ digit PIN) in Australia, so I assume they are available elsewhere too. Parts of Europe are the exception according to my bank.