October 24, 2018

The DC Metro and the Invisible Hand

My friend Tom Lee has been pestering the Washington Metropolitan Area Transit Authority, the agency that runs the DC area’s public transit system, to publish its schedule data in an open format. That will allow companies like Google to include the information in products like Google Transit. It seems that Google has been negotiating with WMATA for months to get access to the data, and the negotiations recently broke down, depriving DC-area transit users of the opportunity to use Google Transit. Reading between the lines, it appears that the sticking point is that WMATA wants Google to cough up some money for access to the data. It seems that WMATA earns some advertising revenue from its existing website, and it’s afraid that Google will undermine that revenue source.

While as a taxpayer I’m happy to see WMATA worrying about its bottom line, this seems like a pretty misguided decision. For starters, this really isn’t about Google. Google has been lobbying transit agencies around the country to publish data in the Google Transit Feed Specification. Although it may sound proprietary, the GTFS is an open standard. This means that absolutely anyone can download GTFS-formatted data and put it to new uses. Of course, Google has a small head start because they invented the format, but with Google making open-sourced tools available for manipulating GTFS files, the barrier to entry here is pretty small.

WMATA seems to have lost sight of the fact that it is a government agency accountable to the general public, not a profit-making business. It’s laudable that the agency is looking for new revenue sources, but it’s silly to do so in the way that’s contrary to its broader mission. And the amount of money we’re talking about here—DCist says the agency made $68,000 in ad revenue 2007—is truly trivial for an agency with a billion-dollar budget. Scuttling WMATA participation in Google Transit looks especially shortsighted when we consider that making schedule information easier to access would almost certainly bring additional riders (and, therefore, additional revenues) to the system.

Finally, and most importantly, WMATA should remember the point made by my colleagues in their recent paper: the most important uses for public data are often the ones that no one expects at the outset. Google Transit is great, and Metro riders will enjoy immediate benefits from being able to access schedule information using it. But there may be even more valuable uses to which the data could be put. And not everyone with a good idea for using the data will have the resources to negotiate directly with the WMATA for access. This is why it’s crucial that WMATA not only release the data to Google, but to make it freely and widely available to the general public, so that other private parties can get access to it. To its credit, Google has asked WMATA to do just that. WMATA should say yes.

Comments

  1. So whose job or which transit routes do you suggest WMATA eliminate so that Google can add some undetermined amount to their profits? Sure, WMATA could probably find the money somewhere to support a free feed in Google Transit format and the resulting lost revenue, but why should they, when they’re dealing with a company that has more than ten times their revenue stream?

    OK, that’s a bit harsh, but whenever I see someone saying “this government agency should give us a bunch of data for free so that we can make a bunch of money selling it in competition with their offerings,” I’m not immediately convinced as a taxpayer that that’s a good idea. It’s a targeted subsidy, and should be examined as such.

    (In this case, if I recall the discussion elsewhere correctly, there were also some substantive issues about update frequency; WMATA aparently didn’t want to be in a position where Google Transit was going to be telling people things that were no longer the case. Although that could obviously have been resolved, you just know that the average user wouldn’t be blaming Google when they went to catch a bus that wasn’t there.)

    • Read the damn article. First, the amount of ad revenue involved is *trivial*, so your “So whose job or which transit routes do you suggest WMATA eliminate” is pure BS. Second, they *already have this data*, it’s not like publishing it in open format would be significant work.

    • Read the damn article. First, the amount of ad revenue involved is *trivial*, so your “So whose job or which transit routes do you suggest WMATA eliminate” is pure BS. Second, they *already have this data*, it’s not like publishing it in open format would be significant work.

    • Read the damn article. First, the amount of ad revenue involved is *trivial*, so your “So whose job or which transit routes do you suggest WMATA eliminate” is pure BS. Second, they *already have this data*, it’s not like publishing it in open format would be significant work.

  2. rp: because it demonstrates the WMATA’s lack of foresight. Everyone who is thriving in the data space is doing so on the concept of open data formats.

  3. lost in DC says:

    @rp
    There’s no evidence to suggest that WMATA would lose any of it’s $68K web advertising revenue if it also made data available to Google. That’s an assumption based on fear, not fact. And, for whatever it’s worth, as of Dec 22 @ 2:33pm EST I don’t see any outside advertising on the homepage of the WMATA. So even if they were in the advertising game as well as the public transportation game, as they apparently claim to be, they’re not being very smart about their choices.

    Besides the hassle, what would prevent me from compiling the bus and rail info and formatting it FOR Google? Is it copyrighted?

  4. This is a situation where the Feds should get involved. WMATA gets funding from them, and it benefits to have an integrated national transportation system.

    I lived in Germany in 2005. I could go to the Bahn website, input departure and destination, and I would get every connection, including bus, subway, local, regional, and national trains, time schedules and locations to get from and to anywhere in the country. Fantastic (espcially for a clueless yank).

  5. An issue here is to evaluate exactly how an agency like WMATA is funded. First, their core competency, and the reason they are funded is to provide a usable transportation network. The presumption is that a network benefits the entire metro region. For anyone who follows urban planning, that isn’t at all presumptuous, but is grounded in fact.

    That being the case, if WMATA funding were closely tied to their ridership, they would be strongly motivated to do anything that would increase ridership,, like making the information re: schedules freely available.

  6. An issue here is to evaluate exactly how an agency like WMATA is funded. First, their core competency, and the reason they are funded is to provide a usable transportation network. The presumption is that a network benefits the entire metro region. For anyone who follows urban planning, that isn’t at all presumptuous, but is grounded in fact.

    That being the case, if WMATA funding were closely tied to their ridership, they would be strongly motivated to do anything that would increase ridership,, like making the information re: schedules freely available.

  7. An issue here is to evaluate exactly how an agency like WMATA is funded. First, their core competency, and the reason they are funded is to provide a usable transportation network. The presumption is that a network benefits the entire metro region. For anyone who follows urban planning, that isn’t at all presumptuous, but is grounded in fact.

    That being the case, if WMATA funding were closely tied to their ridership, they would be strongly motivated to do anything that would increase ridership,, like making the information re: schedules freely available.

  8. Having seen the way DC works (or does not work), I would suggest that there is a department in the WMATA that has: 5 employees, generates only $65k and the WMATA does not want to cut these five jobs. This department’s only job is the advertisment associated with the schedule.

    • That department should go anyway. Even if those employees are only getting $10K a year, plus benefits, that department is operating at a loss.

  9. Scott Ainsworth says:

    in 1991, I was the lead designer of the trip planning software that WMATA currently uses. The trip planning software must merge data from multiple sources (bus and training scheduling systems, bus stop lists, GIS systems, etc.) to create a usable database. Since all of these systems and sources predate the “open” standards by at least decade and a half, they cannot produce the data in the “open” formats. So, this becomes a funding issue, which in turn makes it a political issue. Who will pay to write the software to produce the data in the open format?

    Besides is it really an “open” standard? My former employers had several discussions with Google and even supported the idea of open standards. however, even though 20+ years of continuous experience were brought to the table, Google all but ignored the input received. Because of this, there are significant disconnects between Google’s “open” model and the data model currently used at WMATA. (Keep in mind that the original version of the software used at WMATA ran on 386/16s with 16MB of RAM.)

    So again, who will foot the bill to create and maintain the conversion programs? Will it really benefit the taxpayers who will wind up footing the bill?

    I think ultimately it will benefit the taxpayers. The “billion dollar” budget seems large, but running trains is very expensive and very little is left to allocate for software. In fact, most of that department’s money goes to pay wages for the call center operators not for software upgrades.

    So please keep in mind that the problem is much more complex and costly than simply providing the data in an open format. The open format is a wonderful concept, but it requires funding. And in government, funding requires justification and long-term cost planning. What can Google do to help? Given the circumstances, revenue sharing of some sort seems reasonable.

  10. Anonymous says:

    “In fact, most of that department’s money goes to pay wages for the call center operators not for software upgrades.”

    You don’t think that if the schedules were readily available on mobile phones, there’d be fewer calls to the call center?

    San Francisco added a GPS to every bus, reporting where it is at all times. Not just the schedule, but the actual bus. I used to call San Francisco’s 311 info line all the time, asking “When is the next 38 bus arriving at the corner of Market and 2nd?” to avoid standing 10 minutes in the rain (or let me pick a different bus line that was coming sooner). Now I can get it from nextmuni.com, and my Blackberry has dozens of bookmarks that pop me right to the info for my common trips.

    Last time I called 311 was six months ago — reporting a broken stoplight.

    I could see Google — or a pro-transit nonprofit — funding a one-time fee for a consultant to write the conversion program. But not ongoing “profit sharing”.

  11. Anonymous says:

    His post was not about internet marketing. Can you even read?