Greater Greater Washington

Government


RAC member's resignation a symptom of WMATA's opacity

On November 7, WMATA Riders' Advisory Council member Christopher Schmitt tendered his resignation to the WMATA Board. Schmitt resigned after being frustrated at the failure of a series of information requests. This leads to some hard questions about transparency and oversight at WMATA and the role of the RAC.


Photo by Spec-ta-cles on Flickr.

Schmitt first requested information of the agency early in 2011, seeking detailed Metrorail performance data, as well as information on the safety of and changes made to the signaling system in the wake of the 2009 Red Line crash.

WMATA's reasons for denying both requests are flimsy at best, and signal a continued unwillingness to permit oversight of the agency's operations.

In the case of Metrorail performance data, Schmitt sought the raw data used to calculate the summary metrics provided in the agency's Vital Signs Report, including on-time performance. The agency claimed that the dataabout 40,000 records per day, or 29 million for the 24-month period covered by Vital Signswere too voluminous for them to produce.

Dan Stessel, the agency's spokesman, told the Examiner that WMATA objected to "mining 29 million data points", but sought to fulfill the request in other ways. Unfortunately, when it comes to computing on-time performance, there's no substitute for the raw arrival and departure data.

WMATA's fear of the volume of data also betrays a lack of good data management practices on their part; while 29 million rows might seem like a large quantity of data, in this era of "big data" and petabyte-scale databases, it sounds more like a smokescreen.

Schmitt also requested information about the Metrorail signaling system, its safety, and changes made in the wake of the 2009 Red Line crash. Here, WMATA denied the request due to the agency's involvement in ongoing litigation connected to the crash.

Yet, as Schmitt noted in a message to the RAC, a great deal of information on the subject has already been released to the public, through NTSB reports and other publications, and the agency's response is indicative of an unwillingness to sort out information which actually must be protected from disclosure from other information:

In the case of the ATC data, assuming for the sake of argument that a PARP exemption does apply, WMATA has a duty under its policy to segregate exempt from non-exempt information. WMATA has made no effort to fulfill its duty, and it is not credible to claim that every item of information regarding ATC and collision avoidance in WMATA's possession is subject to exemption. At the least, considerable information maintained by WMATA is already in the public domain, such as via NTSB reports. If information is publicly available, it cannot be withheld as exempt. Therefore, WMATA's position is unreasonable on its face.
Regardless of the reasons for the agency's denials, they are merely symptoms of a much larger, more serious problem: an ongoing lack of transparency and meaningful oversight at WMATA.

WMATA's board is unable to provide independent oversight, and, under the agency's "performance-based management" program, receives only summaries of performance data. The WMATA Office of Inspector General conducts audits, but the recommendations in their reports seem to go mostly unheeded. The Tri-State Oversight Committee, a group which became prominent only after the 2009 Red Line crash, is independent of WMATA, but lacks the staff and funding to conduct the most detailed of investigations; beyond that, their remit covers only safety issues.

After an accident, the National Transportation Safety Board has the jurisdiction to conduct an investigation and issue a report; their criticism of WMATA is routinely scathing, yet the accidents continue. The NTSB cannot compel compliance with their recommendations, and so the same recommendations to WMATA appear in NTSB report after report.

That leaves the Riders' Advisory Council, the group Schmitt sat on. The RAC lies somewhere between being merely a glorified focus group and an actual oversight body; they lack independence from WMATA and, in the wake of these events, apparently have no real investigative powers of their own.

In his resignation letter, Schmitt criticizes the RAC for having gotten too cozy with WMATA and for failing to press hard enough for real transparency. He's perfectly right to do so, and yet at the same time, there's no way that a group created by an act of the WMATA board, operated and funded by the agency, can ever have real independence. As Schmitt notes in his letter of resignation, however, the agency desperately needs independent oversight:

Taken together, the denial of the information requests, the resultant loss of the RAC's independence, and the unwillingness of the RAC to defend its own interests mean that there is no institution within Metro dedicated to transparency and aggressive pursuit of information that is of significant rider interest.

Despite some limited improvement, Metro remains a deeply troubled agency, rife with problems that demand accountability and transparency, and which are too numerous to enumerate here. I hope that in the future, a greater dedication to meaningful information access will develop. This would be a starting point for necessary reforms, as well as for establishing the RAC as an independent, robust voice on behalf of riders.

Given that the organizational structure of the RAC effectively precludes meaningful and independent oversight, transit riders in the region should consider forming a completely independent oversight and advocacy group akin to the Straphangers Campaign in New York City.

The campaign, a project of NYPIRG, independently advocates on behalf of New York City's transit riders. But more importantly, the Straphangers are able to conduct their own investigations, including a survey of announcements in the subway and a survey of payphones in the subway.

Though the Straphangers Campaign performs political advocacy, similarly to groups like the Action Committee for Transit and Greater Greater Washington do locally, it also has the ability to collect its own data and form its own judgments, rather than being limited to carefully-selected and carefully-screened presentations like those the RAC gets now.

This means that the campaign is able to independently assess the performance of the transit authority over time, rather than being dependent on the agency's own (potentially flawed) data.

As an agency formed by interstate compact, WMATA is subject to even less scrutiny than the average transit authority in this country. While an independent riders' group would be no more able to compel action than the RAC is now, they would have the advantage of being able to conduct their own investigation and advocacy work, independent of WMATA and its constituent jurisdictions.

Christopher Schmitt sought for the RAC to become "an independent, robust voice on behalf of riders," but riders may be better off finding their own voice.

Kurt Raschke is an information technology professional and transit enthusiast interested in how technology can improve the usability of transit systems. A car-free resident of Silver Spring, he is a frequent user of Metrorail and Metrobus. He also blogs at Raschke on Transport. All views expressed here are his alone. 

Comments

Add a comment »

Okay, I'm interested. What's the first step?

by yatesc on Nov 21, 2011 12:36 pm • linkreport

29 million rows doesn't even scratch the surface of "Big Data."

29 million rows of data would probably fit on a CD, and could easily be produced and sanitized with a custom view (assuming WMATA are using a SQL or SQL-like database).

by andrew on Nov 21, 2011 5:08 pm • linkreport

@yatesc

The first step is to organize, and not just with a blog, but a real organization, similar in form to the Straphangers Campaign. Blogs come and go, but the Straphangers Campaign has been in operation since 1979. Considering that WMATA spans multiple jurisdictions, I'm not sure how well such an organization would fit into the PIRG system; it might have to be built from the ground up.

There's a lot of useful statistical information that can be gathered through crowdsourcing, but rather than relying on (and placing the burden on) a single person, having an organization operate such a program would provide better continuity. FixWMATA, for example, did valuable work, but now they're gone.

Beyond that, some information (like the reports produced by the Straphangers Campaign) may be easiest to produce not by mining WMATA data or crowdsourced reports, but rather conducting an actual boots-on-the-ground survey. In the case of the Straphangers' pay phone survey, for example, this means that Straphangers Campaign volunteers actually dropped coins in 740 phones across the New York City subway.

The surveys may be conducted by volunteers, but the work of the Straphangers still needs funding; as they put it, donating to the Straphangers Campaign helps them "produce reports, provide useful services, and organize riders". I don't know that people will donate to one person with a blog. But show them an organization that does more than rant—that gathers data, uses the media and political tools effectively, and advocates competently—and that may get their attention.

@andrew

Precisely. WMATA could produce the data; they choose not to. All told, the agency has about 100 TB spread across various Oracle and SQL Server databases, but that's still really not all that much. If Oracle's claims about compression hold up with real-world data, the whole thing could fit in an Exadata rack with room to spare.

by Kurt Raschke on Nov 21, 2011 9:52 pm • linkreport

I was going to say that there could be real data management issues - 29 million rows is too much to analyze in Excel, for example, but WMATA could provide a random sample of its data, or you could get a copy of SAS or STATA and a programmer (maybe hire a college student, or team up with some university's statistics department).

The real point, though, is that WMATA has already analyzed the data and provided summary statistics. It could be that they are lying about being unable to just provide the raw data. Alternatively, it could be that they haven't really done any analysis and just pulled their summary stats from thin air. In any case, this is definitely fishy.

by Weiwen on Nov 22, 2011 11:45 am • linkreport

@Weiwen

I hadn't thought that someone would try to analyze the data in Excel; I agree that that would be unproductive.

Dumping the data into a database to do some initial data reduction (which makes sense, since it came out of a database), and then R or gnuplot would probably turn out okay.

In any event, WMATA's public information policy makes no distinction between requesters who have the capability to analyze the information they're seeking and those who don't.

by Kurt Raschke on Nov 22, 2011 8:55 pm • linkreport

Add a Comment

Name: (will be displayed on the comments page)

Email: (must be your real address, but will be kept private)

URL: (optional, will be displayed)

Your comment:

By submitting a comment, you agree to abide by our comment policy.
Notify me of followup comments via email. (You can also subscribe without commenting.)
Save my name and email address on this computer so I don't have to enter it next time, and so I don't have to answer the anti-spam map challenge question in the future.

or