On Mar 12, 12:45*pm, john b wrote:
On Mar 12, 8:52*pm, solar penguin wrote:

Given how many faulty Oyster readers there are out there, that's the
most likely explanation. *There's one reader at Gipsy Hill that reads
the card and opens the barrier to let you out, no problems, but
sometimes doesn't always manage to write to tell the card that the
journey's been completed! *Three times it given me an Unresolved
Journey because of that! *I've tried telling the station man about it,
but he doesn't believe and/or understand it!

FWIW, nor do I. If the light's gone green and the barrier's opened, I
don't think it's *possible* for the system to leave your journey
unresolved.


--
John Band
john at johnband dot orgwww.johnband.org

John - it is perfectly possible. The gate / validator, station
computer and hence the central system record the exit (or entry -
equally possible) but the exit (or entry) is not written to the card.
It happens from time to time.

On Mar 13, 7:06*am, Jack wrote:
On Mar 12, 12:45*pm, john b wrote:









On Mar 12, 8:52*pm, solar penguin wrote:

Given how many faulty Oyster readers there are out there, that's the
most likely explanation. *There's one reader at Gipsy Hill that reads
the card and opens the barrier to let you out, no problems, but
sometimes doesn't always manage to write to tell the card that the
journey's been completed! *Three times it given me an Unresolved
Journey because of that! *I've tried telling the station man about it,
but he doesn't believe and/or understand it!

FWIW, nor do I. If the light's gone green and the barrier's opened, I
don't think it's *possible* for the system to leave your journey
unresolved.

John - *it is perfectly possible. *The gate / validator, *station
computer and hence the central system record the exit (or entry -
equally possible) but the exit (or entry) is not written to the card.
It happens from time to time.

And you're saying the process is designed so that, even though there's
a record of your journey being completed in the central system, the
fact that it hasn't been written to the card means that when the
system does its reckoning, the central record gets ignored and you get
penalised for an unresolved journey?

If so, that's a sufficiently stupid outcome that I hadn't even
considered it might be the case. Anyone care to confirm?

JB

On Mar 13, 2:07*am, john b wrote:
On Mar 13, 7:06*am, Jack wrote:





On Mar 12, 12:45*pm, john b wrote:

On Mar 12, 8:52*pm, solar penguin wrote:

Given how many faulty Oyster readers there are out there, that's the
most likely explanation. *There's one reader at Gipsy Hill that reads
the card and opens the barrier to let you out, no problems, but
sometimes doesn't always manage to write to tell the card that the
journey's been completed! *Three times it given me an Unresolved
Journey because of that! *I've tried telling the station man about it,
but he doesn't believe and/or understand it!

FWIW, nor do I. If the light's gone green and the barrier's opened, I
don't think it's *possible* for the system to leave your journey
unresolved.

John - *it is perfectly possible. *The gate / validator, *station
computer and hence the central system record the exit (or entry -
equally possible) but the exit (or entry) is not written to the card.
It happens from time to time.

And you're saying the process is designed so that, even though there's
a record of your journey being completed in the central system, the
fact that it hasn't been written to the card means that when the
system does its reckoning, the central record gets ignored and you get
penalised for an unresolved journey?

If so, that's a sufficiently stupid outcome that I hadn't even
considered it might be the case. Anyone care to confirm?

JB

John - the system adjusts data on the card chip during transactions.
The balance and / or season tickets are held on the card chip, not in
the central system. As you are no doubt aware, there are
retrospective refund processes available, both user-iniated by phone,
email etc and also "automated" refunds for "operational issues".
The scenario described may be picked up by the automated process - I
am not sure. It is not "designed" to fail to write an entry or exit
to the chip. There will always be a minority of transactions that are
not completely successful.
The point is that maximum fares are deducted during the transaction
with the gate or validator and not after the fact.

On Sat, 12 Mar 2011 18:20:00 -0800 (PST), Jack
wrote:

On Mar 13, 2:07?am, john b wrote:
On Mar 13, 7:06?am, Jack wrote:
On Mar 12, 12:45?pm, john b wrote:

[...] If the light's gone green and the barrier's opened, I
don't think it's *possible* for the system to leave your journey
unresolved.

John - ?it is perfectly possible. ?The gate / validator, ?station
computer and hence the central system record the exit (or entry -
equally possible) but the exit (or entry) is not written to the card.
It happens from time to time.

John - the system adjusts data on the card chip during transactions.
The balance and / or season tickets are held on the card chip, not in
the central system. [...] There will always be a minority of transactions that are
not completely successful.
The point is that maximum fares are deducted during the transaction
with the gate or validator and not after the fact.

I would expect it to be designed such that the green light is not
given until the card has confirmed that it has successfully received
the update. That way, the transaction can be attempted again.

One of my old Oyster cards used to work on the Underground but on the
bus, 50% of the time a "Card Communication Failure" would happen
immediately afterwards. I wonder why that was... Of course failing
to touch in and out isn't a problem on the bus but as far as I know I
always paid the fa this wasn't a case of a "green light" followed
by an inability to debit the money, which seems to be the situation
discussed above.

Richard.

In article ,
Richard wrote:
I would expect it to be designed such that the green light is not
given until the card has confirmed that it has successfully received
the update. That way, the transaction can be attempted again.

IIRC the Washington, D.C. metro equivalent of Oyster (SmarTrip) -- and
it is very similar, although much simpler -- has a message that the
gates can flash up which is something like "touch again". I
hypothesise that you get this when a valid card is recognized but the
transaction to write data back hasn't completed (or isn't know to be
completed). The gate doesn't open, the display flashes up "touch
again" and when you do so, all is well.

I'm slightly surprised, on reflection, that Oyster gates don't have a
similar message.

-roy