• Re: Welcome to GBR, new rules to screw passengers

    From Theo@3:633/10 to All on Saturday, March 07, 2026 14:51:25
    Tweed <usenet.tweed@gmail.com> wrote:
    Theo <theom+news@chiark.greenend.org.uk> wrote:
    Roland Perry <roland@perry.uk> wrote:
    It's being reported today that refunds for un-used tickets have to be
    claimed by midnight the day before travel (Tardis, anyone?) rather than >> up to 30 days afterwards.

    SNCF and others solved this problem years ago: you need to 'composter' your ticket before you get on the train. No stamp == ticket is not valid.

    A modern version of that would be 'activating' a ticket (paper/electronic) before travel. The stamps includes the time/date so if you activated an e-ticket after you boarded the train then that would be detectable. Really it only needs that logic implemented in the gripper's e-ticket scanner - it knows when the train stopped at station X, so calculate if the ticket was activated after that time and flag if so.

    Activated tickets aren't ordinarily eligible for refunds, non-activated tickets are.

    Why would that not work on our system?

    Theo


    You?d need to abolish split tickets. Current Aztec tickets aren?t activateable. You would need platform scanners or revert to m-tickets (tickets captive in an app that you can activate).

    I don't see why you'd need to abolish split tickets - you'd just need to activate them before the train gets to the starting station of each ticket, either at a station or online. You could activate the whole journey
    in one go.

    I agree that more methods of activation (hardware, online, app, etc) would
    need to be implemented. Passing through the barriers could count as
    activation for one ticket, but you'd need other ways - eg scan barcode using phone/webcam.

    It would need a database of recording ticket activation though, and grippers able to contact it.

    Theo

    --- PyGate Linux v1.5.12
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Tweed@3:633/10 to All on Saturday, March 07, 2026 15:20:33
    Theo <theom+news@chiark.greenend.org.uk> wrote:
    Tweed <usenet.tweed@gmail.com> wrote:
    Theo <theom+news@chiark.greenend.org.uk> wrote:
    Roland Perry <roland@perry.uk> wrote:
    It's being reported today that refunds for un-used tickets have to be >>>> claimed by midnight the day before travel (Tardis, anyone?) rather than >>>> up to 30 days afterwards.

    SNCF and others solved this problem years ago: you need to 'composter' your >>> ticket before you get on the train. No stamp == ticket is not valid.

    A modern version of that would be 'activating' a ticket (paper/electronic) >>> before travel. The stamps includes the time/date so if you activated an >>> e-ticket after you boarded the train then that would be detectable. Really >>> it only needs that logic implemented in the gripper's e-ticket scanner - it >>> knows when the train stopped at station X, so calculate if the ticket was >>> activated after that time and flag if so.

    Activated tickets aren't ordinarily eligible for refunds, non-activated
    tickets are.

    Why would that not work on our system?

    Theo


    You?d need to abolish split tickets. Current Aztec tickets aren?t
    activateable. You would need platform scanners or revert to m-tickets
    (tickets captive in an app that you can activate).

    I don't see why you'd need to abolish split tickets - you'd just need to activate them before the train gets to the starting station of each ticket, either at a station or online. You could activate the whole journey
    in one go.

    I agree that more methods of activation (hardware, online, app, etc) would need to be implemented. Passing through the barriers could count as activation for one ticket, but you'd need other ways - eg scan barcode using phone/webcam.

    It would need a database of recording ticket activation though, and grippers able to contact it.

    Theo


    I don?t see how self scan/activation on the train can be fraud proof. Both
    time and location, especially on Android, can be spoofed. So you could
    leave your ticket unscanned until you saw a gripper and then activate with
    a historical starting point. I suppose you could require all tickets that
    are part of a split to be scanned at the start of the trip by a railway controlled device. But I could foresee chaos at a gateline.

    --- PyGate Linux v1.5.12
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Theo@3:633/10 to All on Saturday, March 07, 2026 15:46:51
    Tweed <usenet.tweed@gmail.com> wrote:
    Theo <theom+news@chiark.greenend.org.uk> wrote:
    Tweed <usenet.tweed@gmail.com> wrote:
    Theo <theom+news@chiark.greenend.org.uk> wrote:
    Roland Perry <roland@perry.uk> wrote:
    It's being reported today that refunds for un-used tickets have to be >>>> claimed by midnight the day before travel (Tardis, anyone?) rather than >>>> up to 30 days afterwards.

    SNCF and others solved this problem years ago: you need to 'composter' your
    ticket before you get on the train. No stamp == ticket is not valid.

    A modern version of that would be 'activating' a ticket (paper/electronic)
    before travel. The stamps includes the time/date so if you activated an >>> e-ticket after you boarded the train then that would be detectable. Really
    it only needs that logic implemented in the gripper's e-ticket scanner - it
    knows when the train stopped at station X, so calculate if the ticket was >>> activated after that time and flag if so.

    Activated tickets aren't ordinarily eligible for refunds, non-activated >>> tickets are.

    Why would that not work on our system?

    You?d need to abolish split tickets. Current Aztec tickets aren?t
    activateable. You would need platform scanners or revert to m-tickets
    (tickets captive in an app that you can activate).

    I don't see why you'd need to abolish split tickets - you'd just need to activate them before the train gets to the starting station of each ticket, either at a station or online. You could activate the whole journey
    in one go.

    I agree that more methods of activation (hardware, online, app, etc) would need to be implemented. Passing through the barriers could count as activation for one ticket, but you'd need other ways - eg scan barcode using
    phone/webcam.

    It would need a database of recording ticket activation though, and grippers
    able to contact it.

    I don?t see how self scan/activation on the train can be fraud proof. Both time and location, especially on Android, can be spoofed. So you could
    leave your ticket unscanned until you saw a gripper and then activate with
    a historical starting point. I suppose you could require all tickets that
    are part of a split to be scanned at the start of the trip by a railway controlled device. But I could foresee chaos at a gateline.

    With a database-based system (which is what e-tickets ultimately are), activation would create a timestamp in the database. It doesn't matter how
    or where you activated it, or what time your device thought it was, it's the creation of the database record that counts - you don't get to spoof the
    server time.

    Activation would be an irrevocable intent that the journey has started. If your activation timestamp was after the train you are gripped on departed
    the station on the ticket, then that could be immediately flagged. If you
    pass through a gateline you would need to have activated your ticket a plausible time in advance (eg if the line speed from A to B is 1 hour, you could not have activated it much later than 1 hour earlier)

    Theo

    --- PyGate Linux v1.5.12
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Sam Wilson@3:633/10 to All on Saturday, March 07, 2026 16:14:16
    Theo <theom+news@chiark.greenend.org.uk> wrote:
    Tweed <usenet.tweed@gmail.com> wrote:
    Theo <theom+news@chiark.greenend.org.uk> wrote:
    Tweed <usenet.tweed@gmail.com> wrote:
    Theo <theom+news@chiark.greenend.org.uk> wrote:
    Roland Perry <roland@perry.uk> wrote:
    It's being reported today that refunds for un-used tickets have to be >>>>>> claimed by midnight the day before travel (Tardis, anyone?) rather than >>>>>> up to 30 days afterwards.

    SNCF and others solved this problem years ago: you need to 'composter' your
    ticket before you get on the train. No stamp == ticket is not valid. >>>>>
    A modern version of that would be 'activating' a ticket (paper/electronic)
    before travel. The stamps includes the time/date so if you activated an >>>>> e-ticket after you boarded the train then that would be detectable. Really
    it only needs that logic implemented in the gripper's e-ticket scanner - it
    knows when the train stopped at station X, so calculate if the ticket was >>>>> activated after that time and flag if so.

    Activated tickets aren't ordinarily eligible for refunds, non-activated >>>>> tickets are.

    Why would that not work on our system?

    You?d need to abolish split tickets. Current Aztec tickets aren?t
    activateable. You would need platform scanners or revert to m-tickets
    (tickets captive in an app that you can activate).

    I don't see why you'd need to abolish split tickets - you'd just need to >>> activate them before the train gets to the starting station of each ticket, >>> either at a station or online. You could activate the whole journey
    in one go.

    I agree that more methods of activation (hardware, online, app, etc) would >>> need to be implemented. Passing through the barriers could count as
    activation for one ticket, but you'd need other ways - eg scan barcode using
    phone/webcam.

    It would need a database of recording ticket activation though, and grippers
    able to contact it.

    I don?t see how self scan/activation on the train can be fraud proof. Both >> time and location, especially on Android, can be spoofed. So you could
    leave your ticket unscanned until you saw a gripper and then activate with >> a historical starting point. I suppose you could require all tickets that
    are part of a split to be scanned at the start of the trip by a railway
    controlled device. But I could foresee chaos at a gateline.

    With a database-based system (which is what e-tickets ultimately are), activation would create a timestamp in the database. It doesn't matter how or where you activated it, or what time your device thought it was, it's the creation of the database record that counts - you don't get to spoof the server time.

    Activation would be an irrevocable intent that the journey has started. If your activation timestamp was after the train you are gripped on departed
    the station on the ticket, then that could be immediately flagged. If you pass through a gateline you would need to have activated your ticket a plausible time in advance (eg if the line speed from A to B is 1 hour, you could not have activated it much later than 1 hour earlier)

    There?s an operational difference between tickets which rely on an external database for validation and those where the validation is inherent in the ticket. The latter includes m-tickets (IIUC) and physical tickets which
    are cancelled or composted in some way. The former need a lot more support infrastructure. It would be interesting to see an analysis of fraud
    vectors for each kind.

    Sam

    --
    The entity formerly known as Sam.Wilson@ed.ac.uk
    Spit the dummy to reply

    --- PyGate Linux v1.5.12
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Roland Perry@3:633/10 to All on Sunday, March 08, 2026 06:21:42
    In message <bCF*B7WAA@news.chiark.greenend.org.uk>, at 14:51:25 on Sat,
    7 Mar 2026, Theo <theom+news@chiark.greenend.org.uk> remarked:

    You?d need to abolish split tickets. Current Aztec tickets aren?t
    activateable. You would need platform scanners or revert to m-tickets
    (tickets captive in an app that you can activate).

    I don't see why you'd need to abolish split tickets - you'd just need to >activate them before the train gets to the starting station of each ticket, >either at a station or online. You could activate the whole journey
    in one go.

    I agree that more methods of activation (hardware, online, app, etc) would >need to be implemented. Passing through the barriers could count as >activation for one ticket, but you'd need other ways - eg scan barcode using >phone/webcam.

    It would need a database of recording ticket activation though, and grippers >able to contact it.

    And to do that reliably requires better mobile phone connectivity than
    exists on many railway lines.
    --
    Roland Perry

    --- PyGate Linux v1.5.12
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Roland Perry@3:633/10 to All on Sunday, March 08, 2026 06:26:16
    In message <jf0oqkd2uk2oaghvrthe1aiome2enqnnfn@4ax.com>, at 10:47:06 on
    Sat, 7 Mar 2026, Trolleybus <ken@birchanger.com> remarked:
    On Fri, 6 Mar 2026 09:51:38 +0000, Roland Perry <roland@perry.uk>
    wrote:

    In message <l47lqk58fhcrqvtihn5f7lsi98bqpmkk96@4ax.com>, at 09:21:52 on >>Fri, 6 Mar 2026, Trolleybus <ken@birchanger.com> remarked:

    Do any TOCs still run Permit To Travel schemes? On stations with no staff >>>>or ticket machines there was a machine you could stick a random amount of >>>>coinage into and it would present a permit which you would hand in at the >>>>destination station and pay the balance of a normal ticket.

    Northern have Promise to Pay machines that prove your starting point.

    Actually they only prove where you bought the Promise to Pay ticket. You >>could have arrived at that station on an unticketed inbound train,
    knowing that the station in question was the last one where you could
    save money by pulling that trick.

    Yes, exactly the same as Permits to Travel or just a TVM.

    The main difference is that if there's a TVM, you should normally be
    able to buy a ticket, not merely register a "Promise to Pay".

    Unless the TVM is broken in some way, like having lost its connectivity.

    Other failures, like a broken touch-screen or having run out of blanks,
    would mean you couldn't get a PtP either.
    --
    Roland Perry

    --- PyGate Linux v1.5.12
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Roland Perry@3:633/10 to All on Sunday, March 08, 2026 06:27:10
    In message <10oh17l$1f5tm$1@dont-email.me>, at 11:12:53 on Sat, 7 Mar
    2026, Mike Humphrey <mail@michaelhumphrey.me.uk> remarked:
    On Fri, 06 Mar 2026 21:44:14 +0000, Mark Goodge wrote:
    On Mon, 2 Mar 2026 17:02:49 -0000 (UTC), Tweed <usenet.tweed@gmail.com>
    wrote:

    GWR?s website is your friend for getting a reservation in advance of
    the
    travel day without needing a ticket at the point of making the >>>reservation.

    The problem with that is that you then get reservations which are never
    actually used.

    T'wife and I went to London last May. We bought walk-up tickets on the
    day, but when we got on the train practically every seat was reserved.
    But we sat in a pair of reseerved seats anyway, on the basis that when
    their owner turned up we'd move if necessary. But nobody did. We stayed
    in the same seats all the way to Paddington.

    Most booking sites insist on giving you a reservation even if you're
    booking a flexible ticket, with no way to cancel or change it. If you
    don't know what time you'll be leaving, you have to take a reservation on
    a train you're unlikely to use.

    Yes, but this isn't new. It's been like that for 20yrs.
    --
    Roland Perry

    --- PyGate Linux v1.5.12
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Trolleybus@3:633/10 to All on Sunday, March 08, 2026 10:32:17
    On Sun, 8 Mar 2026 06:27:10 +0000, Roland Perry <roland@perry.uk>
    wrote:

    In message <10oh17l$1f5tm$1@dont-email.me>, at 11:12:53 on Sat, 7 Mar
    2026, Mike Humphrey <mail@michaelhumphrey.me.uk> remarked:
    On Fri, 06 Mar 2026 21:44:14 +0000, Mark Goodge wrote:
    On Mon, 2 Mar 2026 17:02:49 -0000 (UTC), Tweed <usenet.tweed@gmail.com>
    wrote:

    GWR?s website is your friend for getting a reservation in advance of
    the
    travel day without needing a ticket at the point of making the >>>>reservation.

    The problem with that is that you then get reservations which are never
    actually used.

    T'wife and I went to London last May. We bought walk-up tickets on the
    day, but when we got on the train practically every seat was reserved.
    But we sat in a pair of reseerved seats anyway, on the basis that when
    their owner turned up we'd move if necessary. But nobody did. We stayed
    in the same seats all the way to Paddington.

    Most booking sites insist on giving you a reservation even if you're >>booking a flexible ticket, with no way to cancel or change it. If you
    don't know what time you'll be leaving, you have to take a reservation on
    a train you're unlikely to use.

    Yes, but this isn't new. It's been like that for 20yrs.

    And it's been relaxed recently. When I book with GA (a Trainline user)
    it now asks me if I want to book a seat, so long as it isn't an
    Advance.

    --- PyGate Linux v1.5.12
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Clive Page@3:633/10 to All on Sunday, March 08, 2026 11:12:19
    On 08/03/2026 06:26, Roland Perry wrote:

    The main difference is that if there's a TVM, you should normally be
    able to buy a ticket, not merely register a "Promise to Pay".

    Unless the TVM is broken in some way, like having lost its connectivity.

    Other failures, like a broken touch-screen or having run out of blanks, would mean you couldn't get a PtP either.

    The main failure mode of TVMs around here is that they lose
    connectivity. This seems to allow them to sell tickets, presumably
    without being able to verify credit cards with any server, but not to
    access tickets bought in advance. And it happens quite often.

    --
    Clive Page


    --- PyGate Linux v1.5.12
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Roland Perry@3:633/10 to All on Sunday, March 08, 2026 10:51:14
    In message <9vjqqk5uko1dqiv8gsnc9afasc59sa70s5@4ax.com>, at 10:32:17 on
    Sun, 8 Mar 2026, Trolleybus <ken@birchanger.com> remarked:

    Most booking sites insist on giving you a reservation even if you're >>>booking a flexible ticket, with no way to cancel or change it. If you >>>don't know what time you'll be leaving, you have to take a reservation on >>>a train you're unlikely to use.

    Yes, but this isn't new. It's been like that for 20yrs.

    And it's been relaxed recently. When I book with GA (a Trainline user)
    it now asks me if I want to book a seat, so long as it isn't an
    Advance.

    Great. How many other TOCs/booking_sites do the same, though.
    --
    Roland Perry

    --- PyGate Linux v1.5.12
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Tweed@3:633/10 to All on Sunday, March 08, 2026 12:32:02
    Roland Perry <roland@perry.uk> wrote:
    In message <9vjqqk5uko1dqiv8gsnc9afasc59sa70s5@4ax.com>, at 10:32:17 on
    Sun, 8 Mar 2026, Trolleybus <ken@birchanger.com> remarked:

    Most booking sites insist on giving you a reservation even if you're
    booking a flexible ticket, with no way to cancel or change it. If you
    don't know what time you'll be leaving, you have to take a reservation on >>>> a train you're unlikely to use.

    Yes, but this isn't new. It's been like that for 20yrs.

    And it's been relaxed recently. When I book with GA (a Trainline user)
    it now asks me if I want to book a seat, so long as it isn't an
    Advance.

    Great. How many other TOCs/booking_sites do the same, though.

    EMR asks if you want a reservation. Not much use on the IC routes at the
    moment as reservations have been suspended during the class 810
    introduction programme. (Which seems to be going very badly at the moment)


    --- PyGate Linux v1.5.12
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Roland Perry@3:633/10 to All on Tuesday, March 10, 2026 07:03:48
    In message <4anqR.3$X61.1@fx16.ams1>, at 22:09:04 on Thu, 5 Mar 2026,
    Recliner <recliner.usenet@gmail.com> remarked:
    Roland Perry <roland@perry.uk> wrote:
    In message <dCF*WbNAA@news.chiark.greenend.org.uk>, at 17:47:47 on Thu,
    5 Mar 2026, Theo <theom+news@chiark.greenend.org.uk> remarked:
    Roland Perry <roland@perry.uk> wrote:
    In message <10obrc2$3p39q$1@dont-email.me>, at 12:02:10 on Thu, 5 Mar
    2026, Anna Noyd-Dryver <anna@noyd-dryver.com> remarked:
    Roland Perry <roland@perry.uk> wrote:
    In message <8i5qR.185$uJE9.29@fx10.ams1>, at 01:48:52 on Thu, 5 Mar >>>>>> 2026, Recliner <recliner.usenet@gmail.com> remarked:

    [Someone else said, attribution prematurely snipped]:

    The 'repatriations' taking place by air are a symbolic gesture to show >>>>>>>> "Dubai is open!" If they were *actually* interested in repatriating >>>>>>>> people quickly, they would have organised a fleet of coaches to Riyadh.


    Are there large fleets of suitable coaches and drivers (ie, not >>>>>>>the basic
    buses used for Asian slave labourers) in that part of the world?

    I though they had lots of capacity in order to cope with pilgrimage to >>>>>> Mecca. (1.5m foreigners each year).

    That's 1,900km and a 19 hour drive away from Dubai.

    It's a 10hr drive from Dubai to Riyadh. Driving fleets of coaches across >>>> the desert is simply one of the costs of doing this sort of business.

    But these supposed coaches are plying the pilgrim trade near to Medina and >>> Mecca. Which means it's approximately two days drive away from the people >>> who need transport. It would also be picking up passengers in another
    country, and I wouldn't be able to speculate what their laws say about that.

    Given there's a hundred thousand (or whatever) people stranded, and
    little sign that flights will resume any time soon, then these are small
    problems to overcome.

    Emirates is probably getting something like 20,000 people a day out. The >activity at DXB now seems to be building up, presumably as the Iranian
    drone activity declines. But QR still seems to be grounded.

    But that doesn?t address the problem of the many more thousands of people
    who were planning to fly to or from Europe to Asia, Australia or Southern >Africa via a Gulf hub. Yes, there are other routes, but they have much less >capacity.

    Ironically, the daily non-stop Perth-London flight is now having to go via >Singapore.

    Because of closed airspace and having to take a longer routing. The
    stopover at Singapore adds 3hrs. However the Paris-Perth direct flight
    is still operating (16hrs 15min).
    --
    Roland Perry

    --- PyGate Linux v1.5.12
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Recliner@3:633/10 to All on Tuesday, March 10, 2026 14:27:05
    On Tue, 10 Mar 2026 07:03:48 +0000, Roland Perry <roland@perry.uk> wrote:

    In message <4anqR.3$X61.1@fx16.ams1>, at 22:09:04 on Thu, 5 Mar 2026, >Recliner <recliner.usenet@gmail.com> remarked:
    Roland Perry <roland@perry.uk> wrote:
    In message <dCF*WbNAA@news.chiark.greenend.org.uk>, at 17:47:47 on Thu,
    5 Mar 2026, Theo <theom+news@chiark.greenend.org.uk> remarked:
    Roland Perry <roland@perry.uk> wrote:
    In message <10obrc2$3p39q$1@dont-email.me>, at 12:02:10 on Thu, 5 Mar >>>>> 2026, Anna Noyd-Dryver <anna@noyd-dryver.com> remarked:
    Roland Perry <roland@perry.uk> wrote:
    In message <8i5qR.185$uJE9.29@fx10.ams1>, at 01:48:52 on Thu, 5 Mar >>>>>>> 2026, Recliner <recliner.usenet@gmail.com> remarked:

    [Someone else said, attribution prematurely snipped]:

    The 'repatriations' taking place by air are a symbolic gesture to show
    "Dubai is open!" If they were *actually* interested in repatriating >>>>>>>>> people quickly, they would have organised a fleet of coaches to Riyadh.


    Are there large fleets of suitable coaches and drivers (ie, not >>>>>>>>the basic
    buses used for Asian slave labourers) in that part of the world? >>>>>>>
    I though they had lots of capacity in order to cope with pilgrimage to >>>>>>> Mecca. (1.5m foreigners each year).

    That's 1,900km and a 19 hour drive away from Dubai.

    It's a 10hr drive from Dubai to Riyadh. Driving fleets of coaches across >>>>> the desert is simply one of the costs of doing this sort of business. >>>>
    But these supposed coaches are plying the pilgrim trade near to Medina and >>>> Mecca. Which means it's approximately two days drive away from the people >>>> who need transport. It would also be picking up passengers in another >>>> country, and I wouldn't be able to speculate what their laws say about that.

    Given there's a hundred thousand (or whatever) people stranded, and
    little sign that flights will resume any time soon, then these are small >>> problems to overcome.

    Emirates is probably getting something like 20,000 people a day out. The >>activity at DXB now seems to be building up, presumably as the Iranian >>drone activity declines. But QR still seems to be grounded.

    But that doesn?t address the problem of the many more thousands of people >>who were planning to fly to or from Europe to Asia, Australia or Southern >>Africa via a Gulf hub. Yes, there are other routes, but they have much less >>capacity.

    Ironically, the daily non-stop Perth-London flight is now having to go via >>Singapore.

    Because of closed airspace and having to take a longer routing. The
    stopover at Singapore adds 3hrs. However the Paris-Perth direct flight
    is still operating (16hrs 15min).

    Yes, it's about an hour shorter, so probably just within range, despite the longer routing. Also, perhaps those flights
    aren't as full, so there's less to lose through having to restrict payload.

    --- PyGate Linux v1.5.12
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)