Hello, I wonder if anyone could shed any light on the following.

My wife's aunt recently treated us to a short stay in an hotel. When it came 
to settling up she offered her debit card for payment - having previously 
ensured that her account contained sufficient funds. She typed her pin into 
the card reader and after a minute or so, and to her acute emabarrasment, 
the machine declined the transaction. My wife then offered her debit card 
and the same thing happened - even though our account also had more than 
sufficent funds. Finally my wife offered her mastercard credit card which 
went through with no problems.

All cards (including the mastercard) were issued by Nat West, so we popped 
over to the local Nat West to see what the cause was. The manager blathered 
on about glitches and how it was always a good idea to carry a credit card. 
However, we were really none the wiser about why this happened - though he 
did assure us that the debit cards should have worked even though the amount 
was around 1000GBP. He also warned us that our debit cards would probably be 
locked and took us to a hole in the wall where he showed us how to unlock 
the card.

So, does anyone know what might have caused this and how it can be avoided?

Thanks
Jeff

"Jeff"  wrote in message 
news:79Cuk.213260$ah4.77059@newsfe15.ams2...
> Hello, I wonder if anyone could shed any light on the following.
>
> My wife's aunt recently treated us to a short stay in an hotel. When it 
> came to settling up she offered her debit card for payment - having 
> previously ensured that her account contained sufficient funds. She typed 
> her pin into the card reader and after a minute or so, and to her acute 
> emabarrasment, the machine declined the transaction. My wife then offered 
> her debit card and the same thing happened - even though our account also 
> had more than sufficent funds. Finally my wife offered her mastercard 
> credit card which went through with no problems.
>
> All cards (including the mastercard) were issued by Nat West, so we popped 
> over to the local Nat West to see what the cause was. The manager 
> blathered on about glitches and how it was always a good idea to carry a 
> credit card. However, we were really none the wiser about why this 
> happened - though he did assure us that the debit cards should have worked 
> even though the amount was around 1000GBP. He also warned us that our 
> debit cards would probably be locked and took us to a hole in the wall 
> where he showed us how to unlock the card.
>
> So, does anyone know what might have caused this and how it can be 
> avoided?
>
> Thanks
> Jeff
>

What type of card was it?
Most hotels do not accept Visa Electron or Solo Debit Cards or so I was 
told.

"Section <31T>"  wrote in message 
news:g9erjc$mhs$1@news.albasani.net...
>
>
> What type of card was it?
> Most hotels do not accept Visa Electron or Solo Debit Cards or so I was 
> told.

They were both regular current account debit cards displaying the maestro 
symbol - not electron or solo. Our cards have the words "servicecard 250" 
whilst my wife's aunt's card is guaranteed to 100 GBP. I wondered if the 
bill (>1000GBP) was the problem but the NatWest manager assured me it 
wasn't.

Neither party has had any problem with their card before and the hotel 
seemed happy enough to insert them into their card reader. As a matter of 
fact my wife bought her car on her card - though it did involve speaking to 
the bank.

Thanks
Jeff

"Jeff"  wrote in message 
news:fvCuk.213608$ah4.177049@newsfe15.ams2...
>
> "Section <31T>"  wrote in message 
> news:g9erjc$mhs$1@news.albasani.net...
>>
>>
>> What type of card was it?
>> Most hotels do not accept Visa Electron or Solo Debit Cards or so I was 
>> told.
>
> They were both regular current account debit cards displaying the maestro 
> symbol - not electron or solo. Our cards have the words "servicecard 250" 
> whilst my wife's aunt's card is guaranteed to 100 GBP. I wondered if the 
> bill (>1000GBP) was the problem but the NatWest manager assured me it 
> wasn't.
>
> Neither party has had any problem with their card before and the hotel 
> seemed happy enough to insert them into their card reader. As a matter of 
> fact my wife bought her car on her card - though it did involve speaking 
> to the bank.
>
> Thanks
> Jeff
>

>>>guaranteed to 100 GBP
Isn't that to do with Cheque Guarantee, not debit card transactions?

On Sun, 31 Aug 2008 20:19:30 +0100, Jeff put finger to keyboard and
typed:

>Hello, I wonder if anyone could shed any light on the following.
>
>My wife's aunt recently treated us to a short stay in an hotel. When it came 
>to settling up she offered her debit card for payment - having previously 
>ensured that her account contained sufficient funds. She typed her pin into 
>the card reader and after a minute or so, and to her acute emabarrasment, 
>the machine declined the transaction. My wife then offered her debit card 
>and the same thing happened - even though our account also had more than 
>sufficent funds. Finally my wife offered her mastercard credit card which 
>went through with no problems.
>
>All cards (including the mastercard) were issued by Nat West, so we popped 
>over to the local Nat West to see what the cause was. The manager blathered 
>on about glitches and how it was always a good idea to carry a credit card. 
>However, we were really none the wiser about why this happened - though he 
>did assure us that the debit cards should have worked even though the amount 
>was around 1000GBP. He also warned us that our debit cards would probably be 
>locked and took us to a hole in the wall where he showed us how to unlock 
>the card.
>
>So, does anyone know what might have caused this and how it can be avoided?

Probably a glitch in communication between the terminal and the bank.
If the authorisation system can't contact the bank's computer to
authorise the payment, it will report it as being declined. The credit
card worked because that's authorised by the card company (Mastercard
or Visa) rather than the issuing bank, so any problems on the link to
the bank won't affect it.

There is nothing you can do about this - the problem is in the
communications network. It's not particularly common, but it does
happen. When I managed an online retail system, we'd get maybe one or
two complaints from customers a month that our website had declined
their card even though it was perfectly valid and there were
sufficient funds in the account. It does happen more often with debit
than credit cards, mainly because the authorisation process is more
complex (a lot more different banks to talk to than different card
companies, and the authorisation process is different for debit
cards). So, essentially, your bank manager was right.

Mark
-- 
My rather pointless blog: http://mark.goodge.co.uk
My less pointless stuff: http://www.good-stuff.co.uk

Mark Goodge wrote:

> On Sun, 31 Aug 2008 20:19:30 +0100, Jeff put finger to keyboard and
> typed:
>>
>>So, does anyone know what might have caused this and how it can be
>>avoided?
> 
> Probably a glitch in communication between the terminal and the bank.
> If the authorisation system can't contact the bank's computer to
> authorise the payment, it will report it as being declined.

That's terrible.  The system should not have just two possible outcomes
(accept/decline), and if it can't get through, it should say so.

In particular, if a merchant advertises that it accepts payment by
debit card, then a customer ought to be able to turn up with no cash,
no cheques, no credit cards, just the debit card, and be able to rely
with confidence that it will be accepted.  A mere communication problem
should not make this impossible, i.e. as with credit cards it should be
possible to revert to a backup system whereby perhaps the merchant has
to phone the bank.

On Sun, 31 Aug 2008 20:38:02 GMT, Ronald Raygun put finger to keyboard
and typed:

>Mark Goodge wrote:
>
>> On Sun, 31 Aug 2008 20:19:30 +0100, Jeff put finger to keyboard and
>> typed:
>>>
>>>So, does anyone know what might have caused this and how it can be
>>>avoided?
>> 
>> Probably a glitch in communication between the terminal and the bank.
>> If the authorisation system can't contact the bank's computer to
>> authorise the payment, it will report it as being declined.
>
>That's terrible.  The system should not have just two possible outcomes
>(accept/decline), and if it can't get through, it should say so.

I'm inclined to agree with that, but AIUI part of the point is that
the system deliberately never gives a reason for declining a
transaction as that could give useful information to a potential
fraudster.

>In particular, if a merchant advertises that it accepts payment by
>debit card, then a customer ought to be able to turn up with no cash,
>no cheques, no credit cards, just the debit card, and be able to rely
>with confidence that it will be accepted.  A mere communication problem
>should not make this impossible, i.e. as with credit cards it should be
>possible to revert to a backup system whereby perhaps the merchant has
>to phone the bank.

The costs of making it 100% reliable (as opposed to merely 99.99%
reliable) would not be worth it, as they would inevitably have to be
passed on to the customers (both vendors and cardholders) and the
customers would not be willing to pay for it.

In any case, phoning the bank isn't going to help if the bank's own
staff can't access the data due to the computer being offline. 

Mark
-- 
My rather pointless blog: http://mark.goodge.co.uk
My less pointless stuff: http://www.good-stuff.co.uk

Mark Goodge wrote:

> On Sun, 31 Aug 2008 20:38:02 GMT, Ronald Raygun put finger to keyboard
> and typed:
>>Mark Goodge wrote:
>>> On Sun, 31 Aug 2008 20:19:30 +0100, Jeff put finger to keyboard and
>>> typed:
>>>>
>>>>So, does anyone know what might have caused this and how it can be
>>>>avoided?
>>> 
>>> Probably a glitch in communication between the terminal and the bank.
>>> If the authorisation system can't contact the bank's computer to
>>> authorise the payment, it will report it as being declined.
>>
>>That's terrible.  The system should not have just two possible outcomes
>>(accept/decline), and if it can't get through, it should say so.
> 
> I'm inclined to agree with that, but AIUI part of the point is that
> the system deliberately never gives a reason for declining a
> transaction as that could give useful information to a potential
> fraudster.

That's as may be, but inability to contact the bank is hardly useful
information, and in any case inability to contact the bank is not
really a valid reason to decline.

>>In particular, if a merchant advertises that it accepts payment by
>>debit card, then a customer ought to be able to turn up with no cash,
>>no cheques, no credit cards, just the debit card, and be able to rely
>>with confidence that it will be accepted.  A mere communication problem
>>should not make this impossible, i.e. as with credit cards it should be
>>possible to revert to a backup system whereby perhaps the merchant has
>>to phone the bank.
> 
> The costs of making it 100% reliable (as opposed to merely 99.99%
> reliable) would not be worth it, as they would inevitably have to be
> passed on to the customers (both vendors and cardholders) and the
> customers would not be willing to pay for it.

The vendor will need to balance that putative cost against the cost
of losing a sale when a customer's payment is refused.

> In any case, phoning the bank isn't going to help if the bank's own
> staff can't access the data due to the computer being offline.

I take your point, but just because the computer system is unreachable
by the normal merchant authorising route doesn't mean that it is
totally offline and unreachable by all routes including from the
bank's own staff.

Ronald Raygun <no.spam@localhost.localdomain> writes:

> That's terrible.  The system should not have just two possible outcomes
> (accept/decline), and if it can't get through, it should say so.

Especially when, as has happened to me, the merchant (a catalogue
company in this case) makes a charge for declined payments (whether
credit/debit card or cheque)

Ronald Raygun wrote:

>> The costs of making it 100% reliable (as opposed to merely 99.99%
>> reliable) would not be worth it, as they would inevitably have to be
>> passed on to the customers (both vendors and cardholders) and the
>> customers would not be willing to pay for it.
> 
> The vendor will need to balance that putative cost against the cost
> of losing a sale when a customer's payment is refused.

The cost of making it 100% reliable is infinite.  In act it is beyond 
infinite, as it is just not possible.

On 31 Aug, 20:19, "Jeff"  wrote:
> Hello, I wonder if anyone could shed any light on the following.
>
> My wife's aunt recently treated us to a short stay in an hotel. When it came
> to settling up she offered her debit card for payment - having previously
> ensured that her account contained sufficient funds. She typed her pin into
> the card reader and after a minute or so, and to her acute emabarrasment,
> the machine declined the transaction. My wife then offered her debit card
> and the same thing happened - even though our account also had more than
> sufficent funds. Finally my wife offered her mastercard credit card which
> went through with no problems.
>
> All cards (including the mastercard) were issued by Nat West, so we popped
> over to the local Nat West to see what the cause was. The manager blathered
> on about glitches and how it was always a good idea to carry a credit card.
> However, we were really none the wiser about why this happened - though he
> did assure us that the debit cards should have worked even though the amount
> was around 1000GBP. He also warned us that our debit cards would probably be
> locked and took us to a hole in the wall where he showed us how to unlock
> the card.
>
> So, does anyone know what might have caused this and how it can be avoided?
>
> Thanks
> Jeff

Could it be that they have security checks on the pattern of usage? I
usually only use my debit card as a cash-machine card - maximum
permitted withdrawal £200 per day. The first time I tried to use it
for online purchase of unit trusts - several thousand pounds - it
failed. Had to contact the bank who said their random security checks
had stopped the payment - they had to put a release on the thing so
that I could complete the transaction. It seemed that some unusual,
particularly large, transactions might or might not be stopped. Next
couple of times I was going to do a similar transaction, I called them
beforehand to make sure it wouldn't be trapped. Now they've changed
their security policy and pre-authorisation by phone isn't accepted.
Same bank also used to recommend you phoned them before foreign
holidays, to let them know the destination and period so that an
unusual pattern of payments wouldn't freeze the card - they now say
this is no longer necessary. There are problems with this sort of
security if you are the type of person who mainly carries a card to
ensure they can make unusual and exceptional payments in an emergency
- you don't want the card to fail just when you need it.

Toom

David Woolley wrote:

> Ronald Raygun wrote:
> 
>>> The costs of making it 100% reliable (as opposed to merely 99.99%
>>> reliable) would not be worth it, as they would inevitably have to be
>>> passed on to the customers (both vendors and cardholders) and the
>>> customers would not be willing to pay for it.
>> 
>> The vendor will need to balance that putative cost against the cost
>> of losing a sale when a customer's payment is refused.
> 
> The cost of making it 100% reliable is infinite.  In act it is beyond
> infinite, as it is just not possible.

It depends what you mean by 100%.  Clearly the scenario (of a dialup
machine failing to get through to the bank to obtain authorisation
automatically) occurs often enough that many people know someone to
whom it has happened.  That's often enough, I think, to make it worth
a merchant having a Plan B in place for cases when a customer has no
alternative means of payment.

Now, that plan B could be as simple as trying again, and if it still
fails, phoning for authorisation.  The cost of that is hardly infinite.

A plan C might be be to carry the risk oneself (I think large merchants
do this routinely anyway, using a "floor limit"), and accept payment
without authorisation, or to ask the customer to send a cheque when
they get home, or to come and pay a day or two later.  If the merchant
knows the customer, the risk involved will be as good as zero.  If not,
the risk of becoming a fraud victim is still very low, because:
(a) it is low anyway,
(b) the fraudster would have to have prior knowledge of the connection
being compromised in order to exploit this weakness,
(c) the risk is further diluted heavily as a result of the scenario
itself (of the connection being down) being relatively infrequent.

Again, hardly infinite.

In article ,
 Graham Murray  writes:
> Ronald Raygun <no.spam@localhost.localdomain> writes:
> 
> > That's terrible.  The system should not have just two possible outcomes
> > (accept/decline), and if it can't get through, it should say so.
> 
> Especially when, as has happened to me, the merchant (a catalogue
> company in this case) makes a charge for declined payments (whether
> credit/debit card or cheque)

Er.  Isn't that, like, completely outrageous?
-- 
SAm.

sam@ssrl.org.uk (Sam Nelson) writes:

> In article ,
>  Graham Murray  writes:
>> Especially when, as has happened to me, the merchant (a catalogue
>> company in this case) makes a charge for declined payments (whether
>> credit/debit card or cheque)
>
> Er.  Isn't that, like, completely outrageous?

I think so, but that was in their T&Cs. They sent a monthly statement,
which could be paid by sending a cheque in the post, making a payment by
cash or cheque at a bank, or by credit/debit card online or by automated
phone line. They treated a failed card payment in the same was as a
bounced cheque and added a charge to your account. The problem being, as
has been stated in this thread, that card payments can fail through no
fault of the cardholder.

Graham Murray wrote:

> sam@ssrl.org.uk (Sam Nelson) writes:
> 
>> In article ,
>>  Graham Murray  writes:
>>> Especially when, as has happened to me, the merchant (a catalogue
>>> company in this case) makes a charge for declined payments (whether
>>> credit/debit card or cheque)
>>
>> Er.  Isn't that, like, completely outrageous?
> 
> I think so, but that was in their T&Cs. They sent a monthly statement,
> which could be paid by sending a cheque in the post, making a payment by
> cash or cheque at a bank, or by credit/debit card online or by automated
> phone line. They treated a failed card payment in the same was as a
> bounced cheque and added a charge to your account. The problem being, as
> has been stated in this thread, that card payments can fail through no
> fault of the cardholder.

Ah, but you're being a little unfair in describing this as a charge
for a declined card payment.  It would seem that in fact it is a charge
for not settling your monthly bill.

It's still outrageous, of course, since they ought to have alerted you
that the payment failed, which would (unless you'd already left it to
the last minute) have given you a chance to do something about it
before the deadline, like try again by the same or a different method.

"Ronald Raygun" <no.spam@localhost.localdomain> wrote in message 
news:IKUuk.51209$E41.29095@text.news.virginmedia.com...
> Graham Murray wrote:
>
>> sam@ssrl.org.uk (Sam Nelson) writes:
>>
>>> In article ,
>>>  Graham Murray  writes:
>>>> Especially when, as has happened to me, the merchant (a catalogue
>>>> company in this case) makes a charge for declined payments (whether
>>>> credit/debit card or cheque)
>>>
>>> Er.  Isn't that, like, completely outrageous?
>>
>> I think so, but that was in their T&Cs. They sent a monthly statement,
>> which could be paid by sending a cheque in the post, making a payment by
>> cash or cheque at a bank, or by credit/debit card online or by automated
>> phone line. They treated a failed card payment in the same was as a
>> bounced cheque and added a charge to your account. The problem being, as
>> has been stated in this thread, that card payments can fail through no
>> fault of the cardholder.
>
> Ah, but you're being a little unfair in describing this as a charge
> for a declined card payment.  It would seem that in fact it is a charge
> for not settling your monthly bill.
>
> It's still outrageous, of course, since they ought to have alerted you
> that the payment failed, which would (unless you'd already left it to
> the last minute) have given you a chance to do something about it
> before the deadline, like try again by the same or a different method.



Mmmm....  Interesting stuff.

Not quite the same scenario, but....  We got caught out when buying a big 
ticket item from DABS.  Barclaycard's anti-fraud system clicked in - so the 
xaction wasn't authorised until I had spoken to them.  I then incurred a 
(fairly modest) surcharge from DABS.

My view was (and is) that DABS had extra hassle, so were entitled to charge 
something for that.  Equally, I've found B/card's anti-fraud procedures very 
reassuring and competent on a number of occasions.

For the record, I hadn't read any T+Cs.  But was anyone being outrageous in 
this example...?


-- 
Martin

Thanks to everyone for the helpful replies and discussion.

It does sound like there was a 'glitch' between the merchant and the NatWest 
banking system somewhere along the line - which would explain why both 
NatWest cards were declined.

One worrying fact that I didn't mention earlier is that the Hotel said they 
wouldn't be able to accept a checque from my wife's Aunt if it was drawn on 
the same account as the declined debit card! Since she refuses to have 
anything to do with credit cards that would have placed her in a very 
difficult position if we hadn't been there.

Thanks again
Jeff

"Jeff"  wrote in message
news:nvWuk.103391$6p1.91464@newsfe19.ams2...
> Thanks to everyone for the helpful replies and discussion.
>
> It does sound like there was a 'glitch' between the merchant and the
> NatWest banking system somewhere along the line - which would explain why
> both NatWest cards were declined.
>
> One worrying fact that I didn't mention earlier is that the Hotel said
> they wouldn't be able to accept a checque from my wife's Aunt if it was
> drawn on the same account as the declined debit card! Since she refuses to
> have anything to do with credit cards that would have placed her in a very
> difficult position if we hadn't been there.
>
> Thanks again
> Jeff


Thankfully, it doesn't  work like that... it's the hotel which would have 
been left in a very difficult position...!!

-- 
Martin

At 21:10:43 on 31/08/2008, Mark Goodge delighted uk.finance by
announcing:

> On Sun, 31 Aug 2008 20:19:30 +0100, Jeff put finger to keyboard and
> typed:
> 
> > Hello, I wonder if anyone could shed any light on the following.
> > 
> > My wife's aunt recently treated us to a short stay in an hotel.
> > When it came to settling up she offered her debit card for payment
> > - having previously ensured that her account contained sufficient
> > funds. She typed her pin into the card reader and after a minute or
> > so, and to her acute emabarrasment, the machine declined the
> > transaction. My wife then offered her debit card and the same thing
> > happened - even though our account also had more than sufficent
> > funds. Finally my wife offered her mastercard credit card which
> > went through with no problems.
> > 
> > All cards (including the mastercard) were issued by Nat West, so we
> > popped over to the local Nat West to see what the cause was. The
> > manager blathered on about glitches and how it was always a good
> > idea to carry a credit card.  However, we were really none the
> > wiser about why this happened - though he did assure us that the
> > debit cards should have worked even though the amount was around
> > 1000GBP. He also warned us that our debit cards would probably be
> > locked and took us to a hole in the wall where he showed us how to
> > unlock the card.
> > 
> > So, does anyone know what might have caused this and how it can be
> > avoided?
> 
> Probably a glitch in communication between the terminal and the bank.
> If the authorisation system can't contact the bank's computer to
> authorise the payment, it will report it as being declined. The credit
> card worked because that's authorised by the card company (Mastercard
> or Visa) rather than the issuing bank, so any problems on the link to
> the bank won't affect it.

Nonsense.  Mastercard and Visa do not authorise transactions for UK
issuers.  If the issuer cannot be contacted then the decision will be
made by the acquirer (or the terminal acting as its proxy) and the card
based on their risk management settings.

At 09:24:25 on 01/09/2008, Ronald Raygun delighted uk.finance by
announcing:

> David Woolley wrote:
> 
> > Ronald Raygun wrote:
> > 
> >>> The costs of making it 100% reliable (as opposed to merely 99.99%
> >>> reliable) would not be worth it, as they would inevitably have to
> be >>> passed on to the customers (both vendors and cardholders) and
> the >>> customers would not be willing to pay for it.
> >> 
> >> The vendor will need to balance that putative cost against the cost
> >> of losing a sale when a customer's payment is refused.
> > 
> > The cost of making it 100% reliable is infinite.  In act it is
> > beyond infinite, as it is just not possible.
> 
> It depends what you mean by 100%.  Clearly the scenario (of a dialup
> machine failing to get through to the bank to obtain authorisation
> automatically) occurs often enough that many people know someone to
> whom it has happened.  That's often enough, I think, to make it worth
> a merchant having a Plan B in place for cases when a customer has no
> alternative means of payment.
> 
> Now, that plan B could be as simple as trying again, and if it still
> fails, phoning for authorisation.  The cost of that is hardly
> infinite.
> 
> A plan C might be be to carry the risk oneself (I think large
> merchants do this routinely anyway, using a "floor limit"), and
> accept payment without authorisation, or to ask the customer to send
> a cheque when they get home, or to come and pay a day or two later.
> If the merchant knows the customer, the risk involved will be as good
> as zero.  If not, the risk of becoming a fraud victim is still very
> low, because:  (a) it is low anyway,
> (b) the fraudster would have to have prior knowledge of the connection
> being compromised in order to exploit this weakness,
> (c) the risk is further diluted heavily as a result of the scenario
> itself (of the connection being down) being relatively infrequent.
> 
> Again, hardly infinite.

When dealing with debit cards, one also has to consider the fact that
the customer may not have the funds available.

On Mon, 1 Sep 2008 01:23:42 -0700 (PDT), Toom Tabard
 wrote:

>Could it be that they have security checks on the pattern of usage? I
>usually only use my debit card as a cash-machine card - maximum
>permitted withdrawal £200 per day. The first time I tried to use it
>for online purchase of unit trusts - several thousand pounds - it
>failed. Had to contact the bank who said their random security checks
>had stopped the payment - they had to put a release on the thing so
>that I could complete the transaction. It seemed that some unusual,
>particularly large, transactions might or might not be stopped.

Just as a matter of interest, does anyone know if some banks are more
prone to stopping customers' unusual transactions than others?  I keep
hearing about people falling foul of over-sensitive transaction
pattern-matching, yet I've never experienced any rejections or
referrals, despite buying the occasional car with a debit card and
using ATMs in countries off the beaten trail.

I'm just curious about whether my bank (Nationwide) has a somewhat
laxer approach than others or whether they recognise a quality
customer who shouldn't be inconvenienced :-)

Mike.

"Mike"  wrote in message 
news:gnkob4lk409tbv6l3a50558rnfkt5054d6@news.kempston.net...
> On Mon, 1 Sep 2008 01:23:42 -0700 (PDT), Toom Tabard
>  wrote:
>
>>Could it be that they have security checks on the pattern of usage? I
>>usually only use my debit card as a cash-machine card - maximum
>>permitted withdrawal £200 per day. The first time I tried to use it
>>for online purchase of unit trusts - several thousand pounds - it
>>failed. Had to contact the bank who said their random security checks
>>had stopped the payment - they had to put a release on the thing so
>>that I could complete the transaction. It seemed that some unusual,
>>particularly large, transactions might or might not be stopped.
>
> Just as a matter of interest, does anyone know if some banks are more
> prone to stopping customers' unusual transactions than others?  I keep
> hearing about people falling foul of over-sensitive transaction
> pattern-matching, yet I've never experienced any rejections or
> referrals, despite buying the occasional car with a debit card and
> using ATMs in countries off the beaten trail.

same here (without the cars)

tim

Ronald Raygun wrote:
> David Woolley wrote:
> 
>> Ronald Raygun wrote:

> It depends what you mean by 100%.  Clearly the scenario (of a dialup

I meant 100.000000000000000000000000000000000000000000000000000%

> 
> Again, hardly infinite.

Hardly 100% either.  A 100% solution has to consider the possibility of 
being struck by a large meteorite during the transaction, or very 
improbable, but theoretically possible statistical or quantum mechanical 
effects.

You'll often see this difference in the way that financial service 
companies and engineering ones specify computer system performance.  The 
engineering company will specify that the performance be met 99% of the 
time, but the financial service company will insist that it be met 100% 
of the time.  That difference can make a huge difference to cost, or 
even to feasability.