Debunking SmarTrip myths

The WMATA SmarTrip imbroglio continues with two posts from Sunlight Labs and Transportation Nation. The Sunlight Labs post opens with the following falsehood: “Well, the vendor that our transit planners bought it from has gone out of business, and they’re pulling SmarTrip into oblivion with them: the card incorporates proprietary technology, so it’s impossible to find a new supplier.” Regrettably, Transportation Nation perpetuates the myth, saying “Metro also discovered it has a limited supply of SmarTrip cards and the company that makes them no longer exists. To make matters worse, this non-existent company has a proprietary claim on the SmarTrip technology, so Metro can’t even work with a different company to make more.”

This is patently false. No company has gone out of business, period. Moreover, most of the technology behind SmarTrip is in the software which drives the system, Cubic Nextfare. Nextfare supports ISO 14443-compliant smart cards (ISO 14443 is the international standard for contactless smart cards) as well as the Cubic-proprietary GO CARD, which is what the SmarTrip card is currently based on. There’s no need to “make more” SmarTrip cards (by which we mean GO CARD-based SmarTrip cards), since it is entirely possible to migrate the SmarTrip card to a new smart card platform (that is, any ISO 14443-compliant smart card, like the MIFARE Plus). However, the Sunlight Labs post continues with more FUD; here’s a point-by-point dissection and rebuttal:

  • Will the readers on every bus and subway turnstile have to be replaced? No. Every rail faregate, and every bus farebox on WMATA and the other transit systems which use SmarTrip contains a Cubic Tri-Reader, a multi-standard contactless smart card reader capable of reading Cubic’s proprietary GO CARD as well as ISO 14443-compliant cards. In short, there is no need for new equipment.
  • Will riders have to pay for new cards? No, because there’s no reason for riders to be forced to replace their cards. As riders replace their cards (which can be done using the same procedures the RCSC follows now) due to the card being lost or stolen, or otherwise becoming unusable, they can be issued a new card which will use the new technology.
  • How will balance transfers from the old cards to the new ones work, and how much money will be lost along the way? Obviously I can’t speak to any concrete plans, but the process could be entirely automated with a modern ticket vending machine—you walk up, touch your old card to the reader, the machine informs you that your card is eligible to be upgraded and dispenses a new card, and accepts your old card for recycling. The stored value and passes on your card are transferred losslessly. That said, I doubt there’s any need for a special replacement process—existing SmarTrip cards can work alongside new ISO 14443-compliant cards, so most transfers will occur due to lost, stolen, or malfunctioning cards, for which there are already procedures in place. There is no need to stir up fear about riders losing the money they’ve stored on their SmarTrip cards.

In addition, the Sunlight Labs post fails to address the most important point about open standards in automatic fare collection—it’s not just how you communicate with the card (which is what the ISO 14443 standard addresses), but what you store on the card. If you use Cubic Nextfare with an ISO 14443-compatible smart card, the data formats stored on the card, as well as in the backend systems, are still Cubic-proprietary. Sure, you can theoretically use any vendor for smart cards and card readers, but you have to use Cubic’s software. There is an ISO standard for automatic fare collection, but so far I have yet to see it mentioned by anyone bashing WMATA for failing to use open standards. The standard, ISO 24014-1, was published in 2007, and while I know that Thales has delivered ISO 24014-compliant systems, including Presto for Metrolinx in the GTA, I do not know if it has seen any interest from Cubic, Scheidt & Bachmann, ACS or any other AFC vendors. That said, the development of large infrastructure usually moves excruciatingly slowly, and I would not expect widespread adoption of an ISO standard which is only three years old. But if you want to talk about transparency and open standards in fare collection, that’s where the real vendor lock-in is.

Finally, WMATA did not intentionally deploy a smart card which was not standards-compliant. As far as I can tell, the first edition of ISO 14443 was published in 2000. WMATA launched what would later become the SmarTrip card on February 6, 1995 (per WMATA’s history of the system): “WMATA begins smart-card technology demonstration with its GO CARD program at 19 Metrorail stations, five parking lots and three Metrobus lines.” The pilot program culminated in the public launch of SmarTrip on May 18, 1999. It is unreasonable to expect compliance with a standard which had not yet been published. This last quote from the Sunlight Labs post is thus submitted without comment: “There are open standards for RFID cards. A little more care during the procurement process could have avoided this mess. Unfortunately, it seems that adopting open standards wasn’t a priority.”

5 thoughts on “Debunking SmarTrip myths”

  1. Thanks Kurt, that was a really interesting background around this issue. What I still don’t understand is what the actual, current problem is. It seems like most of the reporting on this has been filtered through non-tech reporters who may not be getting it right.

    From your perspective and expertise, what is the current problem with SmarTrip and why do things need to be changed?

    1. The current problem with SmarTrip is that the underlying card (a product known as the “GO CARD”) is going out of production. This affects other transit agencies as well, including Chicago’s CTA. The GO CARD is an older product, which predates current international standards for smart cards. Because of this, it’s not possible to get a direct replacement from another manufacturer. In order for WMATA to continue issuing SmarTrip cards after the current two-year stock has been exhausted, they’ll need to use a different underlying smart card product, similar to those used by transit systems with newer fare collection systems. There’s no immediate crisis, nor is there a need for any widespread change—the technology WMATA has installed is already capable of accepting new, modern smart cards. All that is necessary is for WMATA to issue SmarTrip cards with a different type of chip inside, once the current stock runs out.

      1. Absolutely, yes. As I indicated in the post, WMATA started the SmarTrip pilot in 1995; the ISO 14443 standard for contactless smart cards wasn’t published until 2000. SmarTrip was the first contactless smart card for transit applications in the United States.

  2. Thanks Kurt, this is some really valuable work you did here. It’s not nearly as sensational as the “OMG EVERYTHING IS BREAKING FOREVER” story I see elsewhere, but certainly more nuanced and thoughtful.

Comments are closed.