tim..... wrote on 18 March 2014 21:57:24 ...


"Richard J." wrote in message news
Roland Perry wrote on 18 March 2014 20:47:53 ...
In message , at 20:03:09 on Tue, 18 Mar
2014, Mizter T remarked:

On 18/03/2014 19:34, Roland Perry wrote:

http://metro.co.uk/2014/03/16/thousa...for-bus-after-
oyster-readers-charge-wrong-card-in-contactless-glitch-4610552/

"Just under one per cent of all Oyster and contactless journeys
involve a card clash and we are seeing this number continue to
drop each week."

Sorry, but 1% seems very high.

My experience of card clash is an error 94 & the gates don't open.

I didn't think that gates were enabled for contactless cards yet.

Given there's a not-insignificant trial running, I'd suggest they are.

I'd forgotten about the trial.

But it's very worrying that they are charging the credit cards of people
not signed up to the trial. That seems to me to be completely wrong, on
many levels (for example, it means anyone can join the 'trial' just by
proffering their card).

The Metro story you referenced in the original post refers to buses -
CPC acceptance is not a trial on buses, it's been accepted as a fare
payment method since December 2012.

I know, but someone mentioned "gates" (error 94 etc), which I don't
recall ever seeing on a bus.

FWIW, trying my contactless credit card on validators (both on gates
and standalone) has resulted in an error message, can't remember which
one.

There are now posters and signs around the transport network warning
against card clash, but I think the message should have been delivered
earlier and more forcefully.

That's another message from the newspaper article - the need for such
warnings, which are quite frankly a desperate attempt to cover up a
massive technology failure.

I'm not sure I'd describe it as a *technology* failure. The passenger
has presented a set of cards to the reader, two of which have validity
for the journey. What do you expect to happen in those circumstances?

If the card furthest away from the reader pad is on the edge of the
acceptable range, it may or may not be read. So you might get a card
clash (neither card accepted - try again) or you might have the nearest
card used for the journey.

The failure is the lack of communication for several YEARS, by both TfL
and the banks. It took me some time to realise that the frequent
mis-read of my Oyster (Freedom Pass) at Tube stations was caused by the
RFID Barclaycard in the same wallet. That was long before RFID cards
had any validity on TfL services. Neither Barclaycard nor TfL had
bothered to warn me.

------------------------------------------------------------------------------------------------------------------------

I think you contradicted your own claim here

If both cards are valid for the journey then you are right that the
technology may have difficulty deciding which one to charge (and certainly
wont know it has made a mistake if only one tries to "connect")

but cards that are not valid for the journey interfering with a card that
is, is a failure of the technology and ought to have been designed out at
the start

Or perhaps a limitation of the technology? Are there in fact any
examples of RFID systems which can handle and ignore any non-valid cards
while processing a valid card?
--
Richard J.
(to email me, swap 'uk' and 'yon' in address)

In message , at 21:32:05 on Tue, 18 Mar
2014, Richard J. remarked:
There are now posters and signs around the transport network warning
against card clash, but I think the message should have been delivered
earlier and more forcefully.

That's another message from the newspaper article - the need for such
warnings, which are quite frankly a desperate attempt to cover up a
massive technology failure.

I'm not sure I'd describe it as a *technology* failure. The passenger
has presented a set of cards to the reader, two of which have validity
for the journey. What do you expect to happen in those circumstances?

A properly designed system would instruct the cards to back-off for a
pseudo random period, then interrogate again. At that point it's much
more likely that only one would respond. Yes, I know it doesn't
currently work like that.

If the card furthest away from the reader pad is on the edge of the
acceptable range, it may or may not be read. So you might get a card
clash (neither card accepted - try again) or you might have the nearest
card used for the journey.

Another, lower-tech, option would be to have twin pads. One for Oyster
and another for contactless credit cards.

The failure is the lack of communication for several YEARS, by both TfL
and the banks. It took me some time to realise that the frequent
mis-read of my Oyster (Freedom Pass) at Tube stations was caused by the
RFID Barclaycard in the same wallet.

That can't be the case, because there's a combined Oyster/Contactless
Barclaycard which has worked perfectly well for years as an Oyster on an
Oyster-only pad.

That was long before RFID cards had any validity on TfL services.
Neither Barclaycard nor TfL had bothered to warn me.

Clashes have been known about for years - I used to have a door-entry
rfid card which I could not keep in my regular wallet as it interfered
with the Oyster.
--
Roland Perry

In message , at 02:01:21 on Wed, 19 Mar
2014, Steve Fitzgerald ] remarked:
I didn't think that gates were enabled for contactless cards yet.

Given there's a not-insignificant trial running, I'd suggest they are.

I'd forgotten about the trial.

But it's very worrying that they are charging the credit cards of
people not signed up to the trial. That seems to me to be completely
wrong, on many levels (for example, it means anyone can join the
'trial' just by proffering their card).

I don't think anyone said they were? They are not charging cards yet,
just reading (and currently rejecting) them when presented.

Bus or tube?

The original article said (about buses):

"Almost 1,800 commuters have been given refunds totalling £11,000 after
Oyster readers charged the wrong card."

Then there's a recent "not-insignificant trial" which from the context
is on the tube. So the tube gates must be activated, and accepting
payments from people on that second trial (iirc the trial is only for
PAYG as season ticket holders didn't qualify).

--
Roland Perry

In message , at 09:33:28 on Wed, 19 Mar
2014, Steve Fitzgerald ] remarked:
I didn't think that gates were enabled for contactless cards yet.

Given there's a not-insignificant trial running, I'd suggest they are.

I'd forgotten about the trial.

But it's very worrying that they are charging the credit cards of
people not signed up to the trial. That seems to me to be completely
wrong, on many levels (for example, it means anyone can join the
'trial' just by proffering their card).

I don't think anyone said they were? They are not charging cards
yet, just reading (and currently rejecting) them when presented.

Bus or tube?

The original article said (about buses):

"Almost 1,800 commuters have been given refunds totalling £11,000
after Oyster readers charged the wrong card."

Then there's a recent "not-insignificant trial" which from the context
is on the tube. So the tube gates must be activated, and accepting
payments from people on that second trial (iirc the trial is only for
PAYG as season ticket holders didn't qualify).

The discussion I referred to (up there) mentioned gates.

That's right. The tube gates which have been recently activated for
contactless credit cards, and which must be contributing to the clashes.

What we don't know yet is how many of the tube gate clashes have
resulted in payments being deducted from a credit card, when the holder
also had an Oyster season. That's the mechanism that's over-charged
travellers at least £11,000 on the buses so far (plus we have to assume
that many such over-charges have gone un-noticed or un-complained
about).

I don't have exhaustive knowledge but I can't ever remember seeing
gates on buses in London recently.

--
Roland Perry

In message , at 12:21:32 on Wed, 19 Mar
2014, Steve Fitzgerald ] remarked:
What we don't know yet is how many of the tube gate clashes have
resulted in payments being deducted from a credit card, when the
holder also had an Oyster season. That's the mechanism that's
over-charged travellers at least £11,000 on the buses so far (plus we
have to assume that many such over-charges have gone un-noticed or
un-complained about).

None, as the gates are not able to charge live cards yet.

Not even the ones belonging to people on the tube-rollout pilot.

http://www.tfl.gov.uk/corporate/proj...mes/28751.aspx

One of my questions being, how do the gates know if a contactless card
presented to them at the moment belongs to a piloter or not?
--
Roland Perry

I'm not sure I'd describe it as a *technology* failure. The passenger
has presented a set of cards to the reader, two of which have validity
for the journey. What do you expect to happen in those circumstances?

A properly designed system would instruct the cards to back-off for a
pseudo random period, then interrogate again.

I still don't understand how that tells it which of the two valid
cards to use.

On the theory that the fairly bad always triumphs over the good, I
expect the solution will be tin foil hat wallete with an external
pocket. My wallet has internal shielding on each card pocket so that
none of the cards in it are visible to external readers. It would not
be hard to make a modified version of such a wallet with one pocket
deliberately left unshielded, so you can use it to touch in without
removing the card from the wallet.

I understand all the reasons such an approach stinks, e.g., for people
who use the Oyster to get to work and a company ID to unlock the
office door, but you heard it here first.

--
Regards,
John Levine, , Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. http://jl.ly

In message , at 14:52:30 on Wed, 19 Mar
2014, John Levine remarked:
I'm not sure I'd describe it as a *technology* failure. The passenger
has presented a set of cards to the reader, two of which have validity
for the journey. What do you expect to happen in those circumstances?

A properly designed system would instruct the cards to back-off for a
pseudo random period, then interrogate again.

I still don't understand how that tells it which of the two valid
cards to use.

If it got an answer from both (with a suitable separation), it could
decide to use the season ticket rather than the PAYG. Doesn't have to
take just the first one, especially just after it had detected a clash.
--
Roland Perry

I still don't understand how that tells it which of the two valid
cards to use.

If it got an answer from both (with a suitable separation), it could
decide to use the season ticket rather than the PAYG. Doesn't have to
take just the first one, especially just after it had detected a clash.

I suppose, but that still doesn't help in the many cases where there
is no clear reason to prefer one card over another. It seems a poor
use of funds to redo the gates when it will only help sometimes.

Like I said, people will get wallets that more or less solve the
problem.

--
Regards,
John Levine, , Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. http://jl.ly

"Richard J." wrote in message ...

tim..... wrote on 18 March 2014 21:57:24 ...


"Richard J." wrote in message news
Roland Perry wrote on 18 March 2014 20:47:53 ...
In message , at 20:03:09 on Tue, 18 Mar
2014, Mizter T remarked:

On 18/03/2014 19:34, Roland Perry wrote:

http://metro.co.uk/2014/03/16/thousa...for-bus-after-
oyster-readers-charge-wrong-card-in-contactless-glitch-4610552/

"Just under one per cent of all Oyster and contactless journeys
involve a card clash and we are seeing this number continue to
drop each week."

Sorry, but 1% seems very high.

My experience of card clash is an error 94 & the gates don't open.

I didn't think that gates were enabled for contactless cards yet.

Given there's a not-insignificant trial running, I'd suggest they are.

I'd forgotten about the trial.

But it's very worrying that they are charging the credit cards of
people
not signed up to the trial. That seems to me to be completely wrong, on
many levels (for example, it means anyone can join the 'trial' just by
proffering their card).

The Metro story you referenced in the original post refers to buses -
CPC acceptance is not a trial on buses, it's been accepted as a fare
payment method since December 2012.

I know, but someone mentioned "gates" (error 94 etc), which I don't
recall ever seeing on a bus.

FWIW, trying my contactless credit card on validators (both on gates
and standalone) has resulted in an error message, can't remember which
one.

There are now posters and signs around the transport network warning
against card clash, but I think the message should have been delivered
earlier and more forcefully.

That's another message from the newspaper article - the need for such
warnings, which are quite frankly a desperate attempt to cover up a
massive technology failure.

I'm not sure I'd describe it as a *technology* failure. The passenger
has presented a set of cards to the reader, two of which have validity
for the journey. What do you expect to happen in those circumstances?

If the card furthest away from the reader pad is on the edge of the
acceptable range, it may or may not be read. So you might get a card
clash (neither card accepted - try again) or you might have the nearest
card used for the journey.

The failure is the lack of communication for several YEARS, by both TfL
and the banks. It took me some time to realise that the frequent
mis-read of my Oyster (Freedom Pass) at Tube stations was caused by the
RFID Barclaycard in the same wallet. That was long before RFID cards
had any validity on TfL services. Neither Barclaycard nor TfL had
bothered to warn me.

------------------------------------------------------------------------------------------------------------------------

I think you contradicted your own claim here

If both cards are valid for the journey then you are right that the
technology may have difficulty deciding which one to charge (and certainly
wont know it has made a mistake if only one tries to "connect")

but cards that are not valid for the journey interfering with a card that
is, is a failure of the technology and ought to have been designed out at
the start

Or perhaps a limitation of the technology? Are there in fact any
examples of RFID systems which can handle and ignore any non-valid cards
while processing a valid card?

------------------------------------------------------------------------------

This isn't really the point at which the technology has failed

if an RFID can't cope with this then that's the technology that's been
poorly implemented

tim

"Roland Perry" wrote in message ...

In message , at 21:32:05 on Tue, 18 Mar
2014, Richard J. remarked:
There are now posters and signs around the transport network warning
against card clash, but I think the message should have been delivered
earlier and more forcefully.

That's another message from the newspaper article - the need for such
warnings, which are quite frankly a desperate attempt to cover up a
massive technology failure.

I'm not sure I'd describe it as a *technology* failure. The passenger
has presented a set of cards to the reader, two of which have validity
for the journey. What do you expect to happen in those circumstances?

A properly designed system would instruct the cards to back-off for a
pseudo random period, then interrogate again. At that point it's much
more likely that only one would respond. Yes, I know it doesn't
currently work like that.

If the card furthest away from the reader pad is on the edge of the
acceptable range, it may or may not be read. So you might get a card
clash (neither card accepted - try again) or you might have the nearest
card used for the journey.

Another, lower-tech, option would be to have twin pads. One for Oyster
and another for contactless credit cards.

-------------------------------------------------------------------

Oh yeah

I can just see TMOTCO getting that right ;-)

tim,