Smart Card Wars

With the election upon us, some candidates have decided that transit Smart Cards are an issue they can use to say “I’m the best candidate” by hitching their star to Ontario’s Presto card program.

The Star reports that the TTC will proceed with a tender for an Open Payment system later this year with the intent of a 2011 rollout.  Mayoral hopeful Rocco Rossi has his own scheme called “Presto Plus”.  Can the TTC actually commit to a new system with the current regime still in office?  Rossi’s campaign confirms that he would cancel the TTC’s scheme if he were to become mayor.

Metrolinx would love to see the TTC sign on to Presto, but many questions remain about just what Presto can do for a truly integrated transit system.

Smart Cards are yet another example of the way that transit technology wars in the GTA get in the way of solving fundamental transit problems.  The technology choice becomes more important than the service it provides.  Here are a few questions anyone with “the answer” should consider.

What Will It Cost and Who Will Pay for the Technology?

Estimates of the Presto implementation for the TTC have ranged well over $400-million, and we don’t have a reliable figure for comparative purposes.  The scheme was originally expected to come in at about a third of that price.  We need a credible implementation cost with no hidden add-ons or scope creep.

The cost estimate has grown over time:

  • November 2000 Fare Collection Study:  $140-million
  • March 2004 Canada Strategic Infrastructure Funding announced:  $140-million shared between Ottawa, Queen’s Park and Toronto
  • June 2007 Business Case Report:  $260-million (2006$) [Full report]
  • 2010 Capital Budget:  $416-million plus costs related to Transit City and the two YUS subway extensions totalling $489-million

It is unclear how long the CSIF money will remain on the table, or if it was ever “real” funding.

Queen’s Park wants to pay 1/3 of the cost.  Bully for them.  The last time I looked, Queen’s Park hasn’t got two pennies to rub together and hopes that advisory panels of businessmen, civic luminaries and activists will carry the can for hard decisions on transit funding.  Who will pay the other 2/3?  Depending on the total cost, that’s a lot of money to spend on fare collection technology.

Over at the TTC, advocates of Open Payment systems claim that the implementation will be paid for by the banking system.  I would like to see that in writing, and whether there’s an upper limit to what the banks will invest in this rollout.  Of course, we won’t know this until after a Request For Proposals has made its way through the mill, and the details may still be considered as confidential commercial information.

What Can The Technology Do?

Presto is a stored value system, in the sense that you load money onto your card’s account, and as you travel around the network, Presto deducts the appropriate fare.  Open payment systems track your riding, and bundle this together in a bill periodically charging you (say via credit card or bank withdrawal) for usage.  That leaves a nice “float” in Presto’s hands of all the unspent fares, similar to the value of unused tickets and tokens for which the TTC has already been paid.

Presto has an automatic reload feature, but this presumes that you have an account that can be billed for that reload.  If reloads using cash are to be supported, then the whole infrastructure of add-value machines and cash handling must be added to implementation and operating costs.

Presto operates by deducting the value of “one fare” (whatever that may be locally) from the stored value of the card.  In the case of GO trips, it also calculates the fare appropriate to the distance travelled, and riders can have a “default trip” programmed on their account.  That default allows riders to avoid having to “tap out” at their destination.  It’s a nice idea, but it shows how much Presto is oriented to people making standard commutes, rather than a more complex pattern of travel through the network.

Open Payment systems work on the premise that people already have something that they routinely use, or can use, to identify themselves.  The obvious examples are credit and debit cards which can interact with contactless readers using RFID technology (Radio Frequency Identification).  However, this is also available using cell phones.  Regardless of the medium, a customer registers with the system to permit transit fares to be billed based on their identification.

Fare Policy is Not a Technology

Far more fundamentally, the question we should ask is this:  what types of fare arrangements would each system offer?

A vital question for proponents of either scheme is the menu of fare options that their systems can support.  Already, we know from the FAQ on Metrolinx site that there are some constraints in the fare combinations that Presto can handle.  Are there more?

Presto does not address the fundamental issue of fare policy.  All it does is to duplicate the existing fare systems.  Is this a system constraint, or does the capability of alternate fare models exist waiting only for policy direction to be implemented?

Everyone who crosses the 905/416 boundary complains about multiple fares.  Indeed, one big attraction of subway expansion north of Steeles Avenue is the disappearance of the fare boundary.  This is not a technology issue, it is a policy issue.  How much should we charge people to ride, and should crossing an invisible line incur a completely new fare?

If we give cross-border riders a discount, what will make up the difference in total revenue?  If we change to a zone system or a time based fare, what are the implications for riders elsewhere in the network?

If we attempt to duplicate the TTC’s transfer regulations, the system must track passengers as they move from route to route, and determine when a “new” trip begins.  That’s a lot of effort and infrastructure we could replace simply by moving to a time based fare.

Why is GO Transit a special fare zone?  How does this fit into The Big Move where GO’s rail lines are treated as if they were a regional rapid transit network?  How far out will we push the subway system so that riders can avoid a higher GO fare?

As part of its review of the Open Payment concept, the TTC has asked its staff to report on a variety of fare options so that we can have an informed debate on a future fare structure.  I have yet to hear a word on the issue from Metrolinx even though this is central to fare policy.

These questions are independent of whichever technology is chosen, but they must be answered.  Some aspects of the system implementation cost are directly linked to the fare structure.  A scheme for fare-by-distance, for example, will require that riders “tap out”.  This changes the passenger flow on vehicles and in stations.  Indeed, the “default trip” scheme on Presto was implemented specifically to avoid “tapping out” for the most common trips.

The Fine Print

The Presto system depends on back-end services provided to Metrolinx by Accenture who, as I understand things, bases their fees on transaction counts.  Without the TTC on the system, the potential fees are much lower, and the business model falls apart.

According to the 2007 TTC report, Queen’s Park has already paid $250-million to Accenture for the development of a GTA farecard system, not including any implementation on the TTC, and that contract goes back nearly a decade.  Is there a contractual arrangement requiring Queen’s Park to make up a minimum payment to operate Presto?  How much have we already invested in that system?  Will this be competitive with an Open Payment system?  Is Queen’s Park locked into a high-cost, proprietary technology?

Beyond Fares

If the transit smart card’s use extends to other services, this moves into an area where the banking industry is already well established.

An important difference between Presto and Open Payments is that the latter does not require reinvention of the banking system.  Banks already exist, and they know how to handle payments through a wide variety of outlets.  If Presto is to become a payment standard, then every store, every agency, must support Presto in addition to the standard bank payment systems, or alternately Presto must adhere to payment industry standards.

This is a case of the tail wagging the dog.

A related issue is “the unbanked”, those who operate only with cash on a day-to-day basis.  The banking industry views these people as a large untapped market.  Moving them to a smart card lifestyle is challenging, but it should be solved by the banking industry as a whole, not as a special function of the transit system.

Where Do We Go From Here?

It’s easy for mayoral candidates and provincial Pooh-Bahs to slag the TTC for its foot-dragging on smart card implementation.  [Yes, I know, “Pooh-Bah” almost by definition cannot be made plural.]  However, the question of “who will pay” has never really been answered, and on that basis the TTC has demurred at each opportunity.

Now with the experience of other systems, notably New York City, in moving to Open Payment technology, the TTC is starting to budge, encouraged by the possibility that they won’t have to pay for or operate a goodly chunk of the system.  I hope that we will see a reworked “business case” soon to substantiate this claim.

Our new streetcars, due to appear in only a few years, will force the TTC to shift away from pay-as-you-enter fare collection.  These cars will load at all doors, and passengers will generally not use a farebox or transfers.  This sets the TTC on a path to a complete revolution in its fare collection, and this will involve some flavour of smart card regardless of which technology is chosen.

Meanwhile, Queen’s Park has insisted that Presto be implemented on the Transit City lines.  The first of these, Sheppard, won’t be in operation until after new LFLRVs hit the city streets, and we will see which fare technology is actually implemented on Sheppard and other Transit City routes.

The TTC and Metrolinx must sort out a viable, common fare collection scheme, but more importantly the entire region must sort out an integrated fare policy.  This brings us directly to the question of operating subsidies from local governments and Queen’s Park, an issue nobody wants to touch.

Battling it out with razor-edged smart cards at dawn is so much simpler, but so much less productive.

44 thoughts on “Smart Card Wars

  1. You are right that payment methods is no replacement for fare policy, but it is easier to talk about shiny new technologies than paying for a sustainable transit system. But there is a question that bothers me about the TTC announcement.

    Now that Visa and Mastercard have their own smart card interfaces, have they made any deputations to TTC board to implement smart card systems? It’s not like the TTC has made the smart card a secret issue since 2006. If these companies do have any working product (and they do), they could have shown it to the TTC and Metrolinx to see if it could be implemented.

    I am only asking because the trend for North American transit authorities is to have a proprietary system (Charlie Card, Presto, OneCard, BreezeCard). Is it just a lost opportunity for open payment systems, or is there something else?

    Steve: The lobbyists for proprietary systems have been working very hard in North America. Meanwhile, the next generation of systems using open payments are just starting to show up here. Those with proprietary technologies need to milk their product lines while they still can.


  2. At the end of the day I really don’t buy into the open payment system. By the TTC’s own admission they are still going to need Presto (or some other stored value system) for those without credit or debit cards. As far as I can tell this just makes open payment an additional payment technology that by it’s very nature has to be in place alongside others. With Presto on the other hand it would be quite feasible for the TTC to a system using only cash and presto; open payment will serve only to add another payment system to be managed.

    It is not even as if the TTC’s proposed system would offload all processing to the banks. The account based billing proposed for aggregating transactions is going to require nontrivial processing by the TTC (since just getting retail credit card systems would lead to a separate credit card transaction for each fare), something that even Presto will not (although only by virtue of contracting that processing to Metrolinx).

    There may have been a case for open payments as the basis of Presto in the first place, but that not having happened I don’t see a particularly strong case for switching now, let alone for the TTC trying to force the surrounding areas and GO to change systems now (and more likely getting stuck with a separate fare technology). Ultimately Presto is workable system that is capable of handling any imaginable fare system for the TTC which has already been largely implemented. At this point changing systems seems only likely to drive the cost up further.

    I would also like to see how the TTC is managing to come up with Presto implementation cost estimates so much higher than surrounding systems (per vehicle, obviously total cost will be higher for a larger system). It can’t be a question of wiring the subway since they could quite easily continue with the strategy at the stations using the system now of installing bus type card readers without physical network connections. I suspect that the TTCs general unenthusiasm for capital spending and distrust of outside projects is coming into play here.

    On a related note, I have to disagree that Presto somehow excludes riders not making regular commutes. This is the first time that GO has had anything like a flexible discount fare, and one can hardly claim that tapping out is a particularly arduous requirement, being the standard for just about every smart card system in the world. Having the option to set a default not needing the second tap is an additional feature for the daily commuter true (and I’ll admit not one that any systems I can think of have), but hardly one that disadvantages the rest of us (even in number or location of card readers seeing as Presto uses the same readers for both tap on and tap off).

    Steve: I didn’t say Presto excludes riders not making regular commutes. My point is that the model of “typical rider” for GO/Presto is a lot simpler than for the TTC.

    As for comparative costs, I think that both sides need to be a lot more open about what the costs will be. If Metrolinx can prove that the TTC overstates the cost of Presto, why haven’t they already done so? You say that Presto can handle any imaginable TTC fare, even though it can’t even handle all of the fares GO has now without reprogramming. I simply want some honest answers about what the “out of the box” capabilities are and whether there are any hard limits to the system. One important point is that any system depending on stored value and charging on a pay-as-you-play basis may not be keeping track of fares at a level where it can dynamically offer volume discounts.

    It’s time for Metrolinx to publish detailed specs on what their system can and cannot do.


  3. Would it be worth the time to look at what Montreal has implemented with OPUS? From my friends who live there, there are occasional problems with the technology. (cards have an anecdotally high failure rate, completely unscientific but it’s a common complaint from what I’ve heard) However, by personal experience, the entire system is neat. You “tap in” whenever or wherever you transfer, and from a rider’s perception, it automatically keeps track of your transfers. I don’t think it extends out to the AMT system, though, but the OPUS cards can act as either cash cards or monthly/weekly passes, another bonus. I’m hoping both these features are considered absolute requirements for the TTC. I’m not sure if they already exist in Presto, as I very rarely travel the Go network and thus do so by tickets.

    And, although it’ll never happen here, I am always enamoured by what I saw in Tokyo with their “IC” smart cards. Not only can the cash value be used for transit, it can be used at vending machines or convenience stores. It’s a handy little item to have around. (they also work incredibly fast compared to Montreal’s OPUS).

    Steve: The folks from AMT did a presentation here last year. AMT has not consolidated its fare systems among the constituent operators in its region. What you do is to purchase up to six (if memory serves) separate services to put on your card. Most people get by with one or two covering their regular travels.

    They figure out overall travel (for planning purposes) on the presumption that most people will take the same path on their return commutes. This avoids the need to tap out for tracking purposes (the fares are independent of distance, and so that info isn’t needed for fare calculations).

    One really important point I forgot to make is the idea that the TTC will have this vast pool of data on passenger travel patterns. They will get much of this even without a tap out feature. Meanwhile, I need hardly point out that they have had vehicle tracking data for the better part of three decades, but never bothered to analyze any of it until I started doing this a few years ago, and as far as I know are still not doing it with the same level of detail. Anyone who tells me that we should have a complex system to track passengers hither and yon doesn’t take into account that the TTC will simply not consider data analysis important enough to waste their time on.


  4. The OPUS system in Montreal is the poorest use of smart card technology I’ve ever seen – all it did was get rid of tickets and transfers from individual agencies and put them on a card. As Steve mentioned, when one loads up their OPUS, they aren’t filling an ‘e-purse’, they are buying virtual tickets and/or passes for specific agencies (AMT trains are included) that are loaded as credits on your card, so, let’s say you’re an STM Montreal customer who wants to use their OPUS on the AMT train – you have to buy AMT tickets and put them on your card first; you can’t just tap on at an AMT reader and go, as what was on your card previously would have been specific to the STM and not valid on the AMT. Some combinations of tickets/passes don’t even co-exist properly, requiring some customers to carry more than one card. Got a headache yet? Every use of transit smart cards I’ve ever seen had the ‘e-purse’ and OPUS is the lone exception.


  5. What will happen to my MetroPass? From everything I’ve heard and read about these local smartcard implementations, I fear that it will be cheaper to drive than to take the TTC (especially while they’re debugging version 1.0), even with all the expenses of maintaining and parking a car or two. The new system has the potential to overcharge in such a way that users cannot recover their overpaid fares–sort of like that freeway north of T.O.

    Steve: Systems around the world have implemented smart cards, and the issue with fare structures is an important part of making things work acceptably. That’s the whole point I am trying to make — rather than pontificating about the importance of Presto, mayoral candidates should be asking questions about what each of these systems can and cannot do. An obvious requirement is to support a Metropass type of functionality, but some schemes for new fare structures would make this difficult or more expensive for riders.


  6. Out of curiosity Steve, just to see if there is a possible hidden motivation for the TTC going with the Open Payment system over Presto, do you have any idea if transactions under the Open Payment system would count as cash fares and whether Presto would use the tickets/tokens fare price?

    Steve: There are two issues. One is to ensure that there is flexibility in designing the fare structure to the TTC’s and Toronto preferences, not to whatever limitations Metrolinx builds into (or accepts) in Presto. Right now, Presto charges the equivalent of a token fare for TTC, but if you want Metropass functionality, you’re out of luck. Hidden motivation? How about having the implementation costs paid for by the vendor? There’s a few hundred million on the table there, more than Queen’s Park’s total spending on Presto to date.


  7. The TTC is going to regret the complexity and passenger confusion that all of this will cause. Replacing tickets, tokens, and cash with Presto or an open payment system but still adhering to the scheme where we pay a flat fare as we enter to a collector or driver is something I can see happening, but I don’t think a POP/all-door loading scheme will survive on the streetcar/TC system. Fare evasion would skyrocket.

    PS … I’m pretty sure the origin of “pooh-bah” is from that episode of the Flintstones.

    Speak oh Imperial Grand Poo-Bah!

    Women of the ladies auxiliary, we are faced with an emergency. The hotel staff has quit, and I hereby call upon the women of the ladies auxiliary to pitch in and help out.

    Group A will do the washing and ironing.
    Group B will do the cleaning and dusting.
    Group C will do the cooking.
    I HAVE SPOKEN. Has anyone anything to say ….

    “Yeah, you’re outa yer cotton’ pickin’ mind” … and then eggs start flying.

    Steve: The hotlink from my article is to Wikipedia’s piece on Gilbert & Sullivan’s “The Mikado” where the “Lord High Everything Else” first made his appearance. The Flintstones were a much more recent use of the term.


  8. Does the Open Payment system make some sort of provision for discounted fares, similar to buying tokens or a Metropass, or would it just take out the full fare each time it was used?

    This is possible with smart cards like the Oyster in London, but I am not familiar with Open Payment.

    Steve: The fare charged depends on the options built into the back-end system for any technology. One option that has been discussed is an automatic sliding scale of charges whereby the rate you are charged depends on your level of usage. Once you get up to a riding level that’s equivalent to a pass, your costs would be capped. This eliminates the marketing barrier where people don’t opt for passes because they think they might not break even, or may be out of town for part of the month.

    The Oyster Card in London caps your daily payments at the value of a daily travelcard, and even allows for whether your travel was peak or offpeak.

    The issue for both the TTC and Metrolinx is to ensure that their back-end systems have the greatest possible flexibility in calculating the fare billed to riders. If someone comes up with a great idea for a new fare (which might even be a temporary promotional fare), the last thing we need to hear is that “the computer can’t do it”, or that “we can’t implement it without six months’ worth of programming and testing”.


  9. How in the name of god can a system slaved to Mastercard or VISA be deemed “Open Payment” (in leading capitals mind and everything).

    Their lobbyists did some fine selling when they persuaded the Admiral and the TTC (and apparently our host) that Presto is “proprietary” but Paypass or Paywave is “open”. As for transaction fees – ask any small business owner how much it costs to accept cards – that 1% cashback or air miles or whatever doesn’t come from thin air.

    Steve: The “open” in the name refers to an international standard for payment systems that already exists. It keeps vendors like the TTC from being locked into a proprietary technology (readers, back-end processing systems, etc) that only work with the closed system’s vendor’s equipment. Imagine buying a car that only ran on fuel from one oil company.

    One other point – will data from Paywave/Paypass terminals be available to data miners, so that in addition to knowing you eat at Swiss Chalet and shop at Best Buy, they know you take transit from Old Mill and from St Andrew on weekdays?

    Steve: This sort of issue has come up as a concern of the Privacy Commissioner, and gets into the implementation of the billing system. The description given as part of an Open Payments presentation at the TTC goes something like this.

    When you tap whatever card, PDA, etc., this data is collected by a central system at the TTC. At that point, your card has been used only as a means of identification. Later (say, monthly), the system at the TTC figures out the value of all the riding you did and sends a consolidated bill to whatever agency/account you have registered for your payment. The trip details stay with the transit agency.

    There is a similar barrier in the way Presto operates in that it is a stored value system that tops up your Presto account as needed from, say, your bank account. The trip details stay with Presto, and only the periodic top-up charges show up at your bank.

    This is an important design issue in any smart card implementation — separating the billing machinery from the trip details.

    Presto is probably not the best system we could have, and Rocco Rossi’s ideas of turning into a payment card could have financial regulation impact so I’m wary of that, but the TTC has no track record in going its own way to be proud of – look at the trip planner / Google Transit fiasco.

    Steve: With all of the political baggage surrounding this, I think it’s important to recognize that pressure for both the move to open data at the TTC (which will allow Google to provide trip planning) as well as the open payment scheme comes from the Chair’s office, and more generally from the political level of the TTC.

    Politicians who attack the TTC, and through it try to take swipes at Adam Giambrone and David Miller, are fighting the wrong battle. Some of them, however, have a vested interest in making Queen’s Park look good, at least by comparison.


  10. Maybe I’m being difficult, but I really don’t want to keep track of 2 different ‘pre-load’ transit cards. I’m already juggling my monthly TTC pass and the Presto card on the daily commute. I can’t imagine the TTC choosing a card solution that didn’t integrate with GO given the number of GO riders who also use the TTC at one end of their commute (or both! in my case). The additional flexibility of Presto for use in outlaying GTA areas makes it a natural fit. I’m not saying its a great system, but its the one most of the GTA has already adopted.

    Steve: There is no reason that both systems could not co-exist other than political infighting, and whatever arrangements might exist where Presto might not be viable on its own without the TTC’s volume of transactions. I would point out that there are vastly more TTC users than GO and other-GTA system users, and if Presto means that I have to pay more to fit into a regional system, I have no interest in using that system. Metrolinx talks about technology, but not about fare structure.


  11. Steve, has anyone filed an FOI to get the Accenture-Ontario agreement and contract for Presto? Or is this considered confidential because of the commercial aspects?

    Steve: I doubt anyone has asked for it, and almost certainly the important details would be withheld.


  12. Dave wrote, “What will happen to my MetroPass?”

    In theory (and in practice in many other systems), you would be in a better place with either a smart card or open payment system, but as Steve has pointed out, Presto currently only charges the token rate for each trip, thus totally ignoring the MetroPass concept.

    I say a ‘better place’ because it would eliminate the need to decide at the start of the month whether or not to purchase the pass or use tokens. If one buys the pass, and then something happens in their life where they don’t use transit for half the month, they feel they overpaid. Similarly, of one opts to go with tokens for the month and then something happens where they need transit every day, they will likely end up paying more than the cost of the pass.

    With a ‘smart technology’ system, it starts off the month charging you the token price per trip, but maxes out when you have spent the price of the pass for that month. A flexible enough system, combined with a fare structure that includes day and week passes, would also take those into account as well.

    The optimist in me expects this to be the case, but the realist in me is not so hopeful.

    Steve: An important part of any design is to decouple the billing process from the day-to-day usage of a card. That can be a problem with stored value systems. It is important that once a fare system “knows” who you are and has a fairly good idea that you will pay (likely by verification up front of a credit card number), then all that should happen as you travel around the system is that your trips are recorded. Later, when it’s time to bill you, the system can figure out how much you owe based on your actual travel and any applicable discounts, bulk purchase arrangements, etc.


  13. How long is the Presto contract with Accenture? Couldn’t they contract out the actual payment processing part of it out to bank or credit card company, so it’s basically like a bank card or pre-loaded credit card and then fully compatible with open payment? Accenture or whoever could get a contract for the installation and maintenance for the physical card readers and reloading stations (or I guess they’re just ATMs really) . I don’t see why with a little cooperation, Presto and open payment could be fully compatible, although I guess that’s a lot to ask from all the organizations involved…

    I’m a supporter of GTA-standardized distance-based fares using tap-in/tap-out, so people pay something like $1 per 5 km (point-to-point distance) regardless of transit method (Go, subway, VIVA, bus, streetcar) to a capped daily/weekly/monthly maximum. I think this could also help to eliminate the duplication of services between transit operators between regional boundries, and station designs could start getting simpler, without having to worry about fare-paid areas between mode transfers.

    It seems like the technology today can do so much, and the TTC just needs to embrace it. Also seems to me that it could actually save a lot of money in the future by really simplifying things.

    Steve: As far as I can tell, the TTC is happy to embrace the technology, provided that someone pays the freight for implementing it, and doesn’t cut back on other subsidies to make up for all the money spent on fare collection. Until the TTC started talking about Open Payment, Presto was content to muddle along with their closed system. If nothing else, the TTC’s actions have forced Metrolinx to admit that their system could not support the same range of payment media as other systems people cite when talking about where Toronto should be going.


  14. Steve: The fare charged depends on the options built into the back-end system for any technology. One option that has been discussed is an automatic sliding scale of charges whereby the rate you are charged depends on your level of usage. Once you get up to a riding level that’s equivalent to a pass, your costs would be capped. This eliminates the marketing barrier where people don’t opt for passes because they think they might not break even, or may be out of town for part of the month.

    The Oyster Card in London caps your daily payments at the value of a daily travelcard, and even allows for whether your travel was peak or offpeak.

    It would be madness for any modern system not to do this… madness… but sadly, from what I’ve seen of the TTC since getting here less than a year ago, I can’t say I’m confident they’ll actually make this happen.

    I’ve used the Oyster card in London–it’s AMAZING.

    The TTC fare structure and their methods of payment are so backwards it’s hard to describe (I seriously can’t use a credit card to buy my Metropass?!). The only thing I can think of that they do right is they have weekly passes. But even those are fixed-date (as opposed to San Francisco and New York, where you can get a 7-day pass and “start” it on any day of the month).


  15. NCarlson writes: “one can hardly claim that tapping out is a particularly arduous requirement, being the standard for just about every smart card system in the world”

    I don’t think you want to mean this literally. Just because every system makes you do this doesn’t mean it isn’t arduous, only that it’s not an exceptional requirement. Maybe tap-out can be made non-arduous, but I’m much more certain that tap-out can be (mis-)designed and (mis-)implemented to make it ridiculously arduous. Not that we’d do that here!

    Tapping in and out is fine if each end of your ride is not at a major transfer point. But what of major transfer points, such as Finch station? Okay, maybe you don’t need to track people going through the station allowing free transfers between subway and bus without tapping in or out.

    If we don’t track each link of a ride, then fare-by-distance schemes get on the one hand more complicated (what distance??), and on the other hand simpler (assume ‘as the bird flies’ between the two points?).

    As someone who has had billing disputes with utilities, I don’t want to touch pre-authorized withdrawals for a variable utility service (e.g. hydro, gas) and I would put transit in the same category. It’s all beautiful if the charges are correct, but as soon as something goes awry — and it rarely goes awry in your favour — then you might be in for months of phone calls. If the utility has your money already, good luck getting it back in a hurry if they’ve made a mistake.

    One thing I would *not* like is an electronic system where I would have to go on a web site to figure out how my monthly charges were determined. And I would also want to see the bugs found and worked out by other guinea pigs.

    Lessons from the 407 billing department, and from people hit with $30,000 phone roaming charges, apply.

    P.S. I know it’s not French, but, “Poohs-Bah”?


  16. Steve said: “Hidden motivation? How about having the implementation costs paid for by the vendor? There’s a few hundred million on the table there, more than Queen’s Park’s total spending on Presto to date.”

    Just to clarify. What I was trying to get across is that if the open payment system is well designed, the TTC might be counting that some would be willing to switch to the cash fare due to its convenience over tickets/tokens. As a result, the hidden motivation would be a perceived increase in revenue if they decide to to charge the cash fare for all or part of the transactions handled by the open payment system compared to what they would get with implementing Presto. The implementation costs being covered by the vendor is a little too obvious to be hidden.

    Steve: Whether they get “cash” fares depends a lot on how the system is implemented. If it keeps track of use, and treats previously unknown visitors to the same discounts as regular users, then they should get the benefit of various discounts and passes automatically.

    It should even be possible to avoid the “first time authorization” delay by saying “oh, hello, a new card” when a previously unknown card is detected. The validation process can take place in the background to determine if it’s a good card, and the next time it is used, it would be accepted or rejected. At most the system would lose one fare, but would gain speed and ease of use.

    Obviously something that is not, in itself, a payment medium such as an RFID enabled cell phone, would have to be registered with the system to link it with an account for payment.


  17. Given that we are now switching to chip and pin for credit and debit cards, I assume they be waiving PINs for TTC rides. I wonder who will be responsible for fraudulent transactions — the TTC, the banks/credit card companies or the consumer?

    Steve: For RFID transactions, you don’t enter a PIN. Once a card is reported as lost/stolen, then it would cease to be valid for transit use (subject to the delay time between reporting and update of the transit system’s invalid card list).


  18. My previous job sometimes included reporting on the pros and cons of various technology options available to clients, which meant I would spend time chatting with the technology vendor. Something I *always* asked was “what it can’t it do?”. Given that we already have Presto, the question I ask any Open Payment proponents is “What can it do that Presto can’t?”

    There does seem to be an issue with co-fares for those who have made 36+ GO trips in a month. This does sound like some sort of software issue that shouldn’t happen.

    It is instructive to look at Oyster. That started out as a purely TfL (transport for London – Tube and within-London-bus operator). London already had a sophisticated system of magnetic swipe cards, and the primary motivation for switching to contactless smart cards was speed… it takes a couple of seconds to insert and retrieve your magnetic ticket, while Oyster requires less than half a second. It doesn’t sound like much, but it increases the throughput at busy Tube stations significantly. TfL then lobbied to get Oyster available at mainline rail stations as well, to make things for convenient for commuters. The use of it in shops was more a useful extra than a serious intention.

    I think the key difference is that it started out in London, and spread outwards… ideally, Presto would have started with TTC, then spread outwards, but can you really see the TTC ever implementing something so different of their own accord? I can’t.


  19. At some vendors (Valuemart for example … perhaps Loblaws … my purchases there are never small) you already don’t have to either sign or use a PIN for small dollar purchases. They simply slide the VISA card through, and you’re done. Quicker than cash. I’ve been seeing this for a few months now … and it seems to work for them, as I now use my VISA card for a $20 purchase, when I would have used to have cash, because it used to be quicker.

    So the PIN/signature issue isn’t an issue.


  20. M. Briganti said:

    I’m pretty sure the origin of “pooh-bah” is from that episode of the Flintstones.

    and you commented:

    The hotlink from my article is to Wikipedia’s piece on Gilbert & Sullivan’s “The Mikado” where the “Lord High Everything Else” first made his appearance. The Flintstones were a much more recent use of the term.

    You are quite right.

    The Oxford English Dictionary (Online) includes the word Pooh Bah with the definition :

    A person who holds many offices at the same time; a person or body with much influence or many functions; (also in extended use) a pompous or self-important person.

    and shows the etymology as :

    [< Pooh Bah, the name of a character in William Schwenck Gilbert and Arthur Seymour Sullivan’s operetta The Mikado (1885). Compare POOH int. and BAH int.]

    And there is no problem with the plural (and it is Pooh-Bahs, not Poohs-Bah as was suggested) as there are several such usages shown in the short selection of quotes used in this entry, the earliest being from 1888 and the latest 1994.

    Steve: Except, if someone is “The Lord High Everything Else”, it would be hard for there to be two of them!


  21. We know from the FAQ on Metrolinx site that there are some constraints in the fare combinations that Presto can handle

    Hang on, where on their website? I really can’t find it – could you provide a link please?

    Steve: Er .. ah .. you go to the Metrolinx website where there is a nice big link to PRESTO .. and over there right at the bottom of the navigation bar is FAQ.


  22. Steve Said:
    “One important point is that any system depending on stored value and charging on a pay-as-you-play basis may not be keeping track of fares at a level where it can dynamically offer volume discounts.”

    Take a look at the actual GO fares under Presto now, this is already being done.

    From GO’s fare pages:
    Rides 1 – 35, 7.5% off the single adult GO fare
    Rides 36 – 40, 87.5% off the single adult GO fare
    Rides 41+, – 100% off the single adult GO fare

    Now, in all fairness, I’m not entirely sure if this is always as good as passes are now, but it is flexible, which nothing GO has offered in the past has been. More significantly though, it demonstrates that dynamic bulk discounting is already in place (and my understanding is that Oakville Transit intends to use it immediately as well). As for the TTC, I somehow suspect that the approach if a decision was made today would be that such a system would only result in “lost revenue”, but that really has nothing to do with the system itself. In fact, I have yet to see any evidence that Open Payment will handle dynamic discounts, although I suspect it could in one way or another.

    Steve: GO’s policy on discounts, approved in February, is that 10-ride tickets are discounted 7.5% off of 10 single fares, and monthly passes are discounted 17.5% off of 40 single fares. If we start with a fare of, say, $10, you would pay $9.25 for trips 1-35 ($323.75) and $1.25 for trips 36-40 ($6.25). The total is $330, or 17.5% off. Therefore Presto is handling pass pricing today, at least for GO. Elsewhere in this thread people have noted discount structures for co-payments on local transit systems.

    Life gets a bit trickier for the TTC because there are fare caps within a day, weekly passes, and special arrangements for group travel on weekends and holidays.

    You also say that there are existing GO fares that have problems with the card, what are they? I might not have been following this closely enough to notice, I can’t think of anything off hand that it doesn’t do now, apart from concession fares.

    Steve: Their own site says that it cannot handle people who use the combination local transit plus GO fares more than 35 times a month, and that existing monthly pass programs are not integrated. Obviously, they are hoping that all systems will move to the “frequent rider” model.

    I pay a discounted rate to ride the local bus to the GO Station. Will the system deduct that lower co-fare? Starting May 1, 2010, GO Transit will introduce a revised fare discount program, and due to current PRESTO system limitations, the lower co-fare can only be deducted for up to and including the first 35 trips in a calendar month. Full functionality for co-fares with local transit systems will be available at a later date. Until then, customers who travel by GO Train combined with local transit more than 35 times per month should continue to use their current type of GO ticket or pass. Co-fares are not applicable to the TTC.

    Will monthly/weekly passes still be available? Each transit authority will continue to set its own fare policy. Some will continue to offer a monthly pass, while others will be moving to a “frequent rider” program where the fare is reduced after a set number of trips have been taken. Watch for more information on this from your transit service provider.

    On the transfer tap issue Ed brought up, it will be interesting to see how GO handles them once the buses start getting Presto; the norm does seem to be tap out and back in for each vehicle, and I suspect GO will take that path. That said, most places (outside Toronto that is) aren’t already set up for barrier free transfer, and it really wouldn’t make sense for the TTC to start requiring multiple taps at places not needing transfers now. The only reason a tap would be needed somewhere currently barrier free would be to get routing data from individual passengers and even GO’s existing fare by distance doesn’t depend on knowing the routing, but is based entirely on zone pairs (with an infuriatingly unpublished zone map).

    Steve: GO has a fundamentally different model of passenger handling on its buses from the TTC’s surface and rapid transit networks. If the TTC had dwell times at stops as long as GO buses do, the Queen car would leave Neville on Monday and arrive in Long Branch on Tuesday. My complaint about Presto is that it (a) duplicates existing products, (b) does not address all of the complexities of a GTA wide fare structure, and (c) until a quite recent announcement, was using a technology that is not compatible with international standards.


  23. The Canadian Public Transit Discussion Board has a thread on Presto.

    The glitches and gotchas listed there reinforces my previous comment: until it’s working well, I would not trust it to withdraw from my bank account. And customer service in case of problems appears to be an issue.


  24. I’ve used smart card based systems in Montreal, Vancouver, NYC and most recently the new ORCA Card in Seattle. I don’t understand why Presto sounds so poorly designed, but there’s nothing stopping the system from being sane.

    The ORCA Card in Seattle supports both a stored value ePurse as well as numerous passes for the various transit systems in the region. The systems in the region make use of all of the following features and they all integrate just fine on one card: timed transfers, ride free zones, zoned fares, different fares between systems and distance based fares on the new LINK Light Rail.

    Only the distance based fares require a tap out.

    Additionally varying kinds of monthly or weekly passes can also be stored on the same card. Instead of offering a fixed “Metropass” like in our TTC, you can purchase a monthly pass at at any cash value, e.g. $2.50, if some element of your trip requires the equivalent of a $3 fare, an additional $0.50 is deducted from your ePurse and you are given the equivalent of a $3 2-hour transfer.

    I don’t see why (if Presto were properly designed) a similar system wouldn’t work great for the TTC.

    As far as this Open Payment goes, it seems to heavily depend on deferred back-end reconciling, which in my experiencing working on systems dealing with inventory tends to be a bad thing if it can be avoided. Any sort of system that requires separate reconciliation is bound to cause confusion and complaints from customers who will look at their credit card bill and wonder why they were charged one amount when they expected a different amount. If the TTC wants to provide detailed trip/payment information to help the customer understand why their bill looks the way it will, it sounds like it would likely duplicate much of the infrastructure required to do it at payment-time like a properly designed smart card system would.


  25. It’s quite ironic that there is all this resistance to open payment, especially from people that (one would think) want to see more private-sector involvement. Open Payment (if properly implemented) is a very benign opportunity for private-sector involvement in the public transit business.

    Any P3 model (…or AFP… pick your acronym; a rose by any other name…) should be about getting the most efficiency and, by extension, the best value for money. Open Payment to some degree opens the door to an indirect effect of downsizing, but not in real terms (staff will no doubt be reassigned, not laid off, which allow said staff to do useful things instead of redundant things). Open payment allows the actual cash handling (albeit in a digital medium) to go to a private-sector partner, whether a credit card company or Interac or the banks themselves, meaning the TTC just needs to keep the records of trip accounts, and need not actually collect any cash.

    For those that don’t have bank accounts, it can reasonably be argued that it should be the banks’ or Interac’s responsibility to offer a non-bank or non-credit card alternative fare medium. As was mentioned when the TTC did a presentation on open payment, these institutions are looking that public transit universe as a place to expand their market. Let them provide a card for the kids, or a card for temporary visitors, or a card for those that for whatever reason don’t have a bank card of any kind. It is these companies that get the transaction fees at the end of the day, after all.

    In a political climate where they are trying to squeeze every last dollar, the Province should drop Accenture on the spot – Presto isn’t going to help transit agencies’ bottom lines. Open Payment allows a good sized chunk of transit agencies’ resources devoted to fare collection, processing, and security to be taken off their shoulders, and have all agencies effectively outsourcing fare collection to the same financial institutions. That lowers operating costs and increases productivity (like providing more service?), as redundancies drop by a noteworthy margin.

    Smart Cards in Japan, for comparison purposes, work fantastic as an in-house development because Japanese railway companies are mega-empires with very diversified business activities that go far beyond providing transportation services. In some cases, the railway company itself also has a bank company under its umbrella. Others typically have some kind of financial business that isn’t necessarily a bank. So it’s easy for them to do Smart Cards in-house, they have the resources in-house for other purposes already. TTC is a whole other kettle of fish, and outsourcing fare collection to where those resources already exist is where the best results are going to be had in terms of efficiency and resources management. Only Open Payment allows that; Presto is a different fare medium, but does little to actually improve efficiency of resources (human or technological) of transit agencies. There is a report from a few years back that compared electronic fare collection to the existing “gravity system,” and it validates this point.

    The Province seems to always turn a blind eye to the best P3 opportunities, like the Ambassador Bridge for expansion in crossing the Detroit River, and Open Payment for transit fare collection.


  26. “When you tap whatever card, PDA, etc., this data is collected by a central system at the TTC. ”

    Sounds to me like the only real difference between Presto and “Open” Payment is Presto is prepaid and OP is postpaid. There is still a black box between the passenger and the payment.

    As for Presto’s deemed inadequacies – would you spend a bunch of time programming for the needs of a customer who is plainly looking to avoid purchasing your product, or concentrate during initial rollout on the customer is actually rolling it out?

    Steve: You might try making sure that your system works for your single largest potential client. What’s more, you might reasonably ask why Ontario has to invent its own system. We have a long tradition of using transit for dubious industrial development schemes, some of which allow Ministry staff to justify their existence and pay fat fees to consultants.


  27. There’s another, much more fundamental consideration that seems to be overlooked. In Canada, our payment cards don’t actually do any offline transactions. All transactions phone home for confirmation. Canada is unique in this regard. In most countries, offline transactions (which are a bit of a hybrid between online and stored value) were common years before transit payment came along. I’m not sure that payment cards would work as quickly, smoothly or as ubiquitously here as they have elsewhere.

    Steve: Considering that it’s the banking industry that is strongly advocating an open payment scheme, I think they might have worked out this problem by now.


  28. I don’t think Presto is re-inventing the wheel. It looks like a duplicate of smartcard systems used elsewhere in the world, I strongly disagree with Steve’s statement that “Ontario has to invent its own system”, because it hasn’t! The technology and systems have all been used elsewhere before. If anything, Ontario is just being a copycat, rather trying something new.

    Thanks for the link Steve – I was taking the Presto website as being a separate thing from the Metrolinx website. The “no co-fare over 35 rides” thing is a worry, but it’s being portrayed as something it can’t do *yet*, rather something it can’t do at all. That said, I’d love to know exactly why things get stuck after 35 rides. I have a feeling it’s to do with how discounted rides are deducted from your e-purse.

    Steve: I am meeting with the technical folks from Presto on July 26 and will pursue these and many other questions about their system.


  29. Accenture’s history (of how they do business with public AND private sector) indicates that ANY changes to fare policy will be very costly to the taxpayer (and/or riders). Personally, I would not trust Accenture with access to my bank account. (NOTE: Accenture was once upon a time called “Andersen Consulting” the consulting branch of Arthur Andersen, who were the accountants for Enron). Presto has the potential to become QP’s next “eHealth fiasco”.

    Steve: I believe that Accenture will tell anyone who asks that the consulting side of the firm (now called Accenture) was completely separate from the auditors who helped Enron to its ill-gotten gains.

    Fare policy has been totally ignored by the Metrolinx (and GO, TTC, MT, YRT and DRT) bureaucrats since it’s inception. Riddle me this; how come one can travel from Stouffville to Stoney Creek on a single YRT or vice versa on a single HSR fare, but to cross the Rouge costs DRT + GO + TTC fares (over $10) apply??? Or a trip from 14th and McCowan to Finch and McCowan costs close to $6 ???? Short rides on a GO bus costs $4.50 while “all day parking included” trip on GO can be less than $6?

    Then again, how can we expect Metrolinx to understand electrification if they cannot grasp the simple s**t like designing a proper schedule? placement of bus stops? ad nauseum


  30. It seems to me that Metrolinx is creating barriers within the technology in order to drive fare policy…

    Steve: I might agree with you, except that I don’t think Metrolinx even has thought about “fare policy” beyond GO Transit.


  31. What is the repercussion of the “no co-fare over 35 rides” issue with Presto?

    Is it that after 35 rides, it charges one the regular local fare instead of the 50 or 60 cents (depending on the local agency) or does it just not charge anything after 35 rides? Or, does it just beep an error indicating that it cannot process it and the rider must either pay cash or get off and use the “two foot trolley” (walk).


  32. Daniel Hammond wrote, “Riddle me this; how come one can travel from Stouffville to Stoney Creek on a single YRT or vice versa on a single HSR fare…”

    Where do you get this from?!?

    A single-zone YRT fare will get you from Stouffville to Bramalea City Centre, but your transfer will not allow you to board another Brampton Transit vehicle. There are destinations across boundaries such as this where one does not have to pay extra (unless the vehicle is operated by the TTC), but continuing one’s trip on another vehicle will require another fare.

    I understand the point Daniel was trying to make about the inequities of the GTHA fare structures, but this sort of hyperbole does diminishes the argument.


  33. David Hammond Riddle me this; how come one can travel from Stouffville to Stoney Creek on a single YRT or vice versa on a single HSR fare, but to cross the Rouge costs DRT + GO + TTC fares (over $10) apply?

    Every GTHA system other than TTC has a free transfer policy with its neighbours (ignoring Durham, becuase they only intersect TTC’s), because they figure that is better to allow ‘foreign’ transit users in and get money from them on the way out; than to put them off by charging them on the way in as well.

    TTC presumably just sees the “lost” revenue.

    NB: when Burlington Transit and Oakville Transit enabled Presto, I emailed to find out how inter-municipal transfers will work. If (for example) you board a Burlington bus with Presto, you have a two-hour window from the moment of boarding (slightly better than the old “two hours from when the bus’s run started”). You get a free transfer onto Oakville buses, but the two time limit remains unchanged. I assume this rule will become the norm. So, you’ll no longer be able to go from Stoney Creek to Stouffville on an HSR fare, unless you can do it inside two hours.

    (In rush hour, you can travel by TTC to Rouge Hill GO station, and take the DRT bus from there.)

    Steve: Unlike all of the other GTA bus systems, the TTC has a huge network and many people want to ride it. The cost of accepting neighbouring system’s transfers on the TTC is to, in effect, extend a single fare zone deep into the 905. Somebody has to pay for this, not to mention the riding it will generate on the 905 routes.

    Everyone wants cheap rides, but the last time I looked, the City of Toronto was paying the ENTIRE operating subsidy for the TTC, over $400-million in 2010.


  34. Thom’s note on “online” payment is not quite right. Credit card transactions can be batched offline and then called home when they are collected. The parking machines we have in Toronto were previously offline, but that caused too much trouble when a large chunk of the cards were rejected as expired/blacklisted/etc. when the batches were processed. (Eventually, an online system was implemented, which is why using a parking machine takes longer now that it used to.)


  35. Yes the Stouffville to Stoney Creek trip is a goofy anomaly, but it indicates that in too many areas of the GTA, public transit is dysfunctional. I have personally made Markham to Caledon (2 transfers on Brampton Transit to the boundary and a walk across the street. Newmarket to Stoney Creek (2 zone YRT) and Stoney Creek to Etobicoke trips on single fares. The policy in most of 905 is to provide a fresh transfer upon boarding a new route. The policy is not often communicated well to drivers and even some of the “management” is rather clued out. It also doesn’t help when GO and the various transit agencies REFUSE to allow public examination of fare agreements.

    In the absence of proper direction on fare policy, Presto, Open Payments or what-ever e-Fare system will be a fiasco of proportions that will dwarf Scarboro’s hover-train and crappy Orion buses (hybrid or otherwise).

    This dysfunctional fare system encourages, rather than discourages driving to a GO station. Yes I have used the route 109 on numerous occasions, and it makes me wonder why folk on the “other side” in 905 can get to Rouge Hill GO for 65 cents … while passing neighbourhoods in 416 where the 109 will NOT stop for … they have to wait for a TTC bus and pay full fare. Then again has the TTC even requested a co-fare agreement with GO? I seriously doubt that any such correspondence has ever occurred. Then again, the level of service on the route 109 is so s**ty that driving to Rouge Hill GO is the preferred alternative. Heck why encourage end-to-end transit usage when the possible real agenda of Metrolinx is only to hand out construction contracts, which parking lots, and especially parking garages, make sweet contracts for the “correct” parties.

    And I would like to sign off while on the topic of parking, the minimum fare on GO (including that Rouge River crossing) is now up to $4.20 … yes we have “pure” transit riders cross-subsidizing “free parking”.

    Steve: GO/Metrolinx have repeated stated that they will not enter into a co-fare with the TTC as they don’t want Toronto riders clogging up their trains. There is even a study underway to see if GO can dump some of its train traffic onto the subway to relieve congestion at Union. I’m sure that they assume the TTC has lots of surplus capacity just waiting to be used.

    So much for regional co-operation.


  36. Now … how does this 2 hour thing work if I tap in on Peterborough municipal bus … and then tap out on a GRT or Niagara Falls 4 hours or more later?


  37. Daniel Hammond wrote, “Yes I have used the route 109 on numerous occasions, and it makes me wonder why folk on the “other side” in 905 can get to Rouge Hill GO for 65 cents … while passing neighbourhoods in 416 where the 109 will NOT stop for … they have to wait for a TTC bus and pay full fare.”

    However, if they then take GO to Union and need to get up to Bloor, having paid a TTC fare to get to Rouge Hill, a transfer from that bus in Scarborough is good to get on the subway at Union under the “TTC Times Two” policy.


  38. Daniel Hammond asks: “Now … how does this 2 hour thing work if I tap in on Peterborough municipal bus … and then tap out on a GRT or Niagara Falls 4 hours or more later?”

    There are trips within Toronto that at least approach the two-hour mark. My daily trip from Long Branch to Seneca at Finch/404 was 1:45 if everything worked right in the early morning, but could easily exceed two hours in the afternoon. This wasn’t done for pleasure (I think I was logging over 1,000 km on the TTC every month), and I’m sure there are others in the same unfortunate position, maybe short- or medium-term. GO wasn’t much of an alternative, even if I could have travelled using just the TTC fare. (Trip via GO train Union would be faster, but runs into frequency-of-service issues.)


Comments are closed.