(Photo courtesy of Elizabeth Taylor Photography)
Bill Ferguson is something of a darling of Baltimore city’s tech community. It only helps that he’s earnest about publishing more government data.
The first-term Democratic state senator representing Maryland’s 46th District in Baltimore (the city’s southern and southeastern neighborhoods) makes all the right noises. He supports ride-sharing startup Lyft. He’s the Senate chair of the state’s Joint Committee on Transparency and Open Government, a committee created thanks to a bill he and Heather Mizeur (now a candidate in Maryland’s 2014 gubernatorial race) ushered through the General Assembly in 2011. And he wants Maryland to embrace open data.
To that end, Ferguson has introduced a bill into this year’s session that calls for the creation of a statewide Open Data Policy and a Council on Open Data to sift through Maryland’s databases and determine what should be made publicly available, as well as what should remained locked away from the public’s eyes.
If it passes, Ferguson said the policy will be Maryland’s “first formalized step towards an open data regime.”
Maryland was ranked 46th among the states in terms of public access to information in spring 2013. A FOX45 report in fall 2013 revealed that access to certain types of data — such as how tax dollars are spent — is prohibited by the sheer cost of procuring the information.The state sorely needs it, despite the launch of its open data portal in May 2013. While Governor Martin O’Malley is long on talk about StateStat and the extent to which data drives decision making, the results haven’t earned glowing reviews.
We spoke with Ferguson, who is up for re-election this year, about his Open Data Policy (which you can read in full here), and what he hopes will change in an “open data regime.”
This interview has been edited for length and clarity.
TB: What’s the one outcome you hope to achieve from this Open Data bill?
BF: One tangible goal is to create a culture of open data in Maryland state government.
TB: Which means?
BF: One of [our] big findings is there’s no one reason why open data isn’t more prevalent than it is today. There are lots of reasons. Some of them have to do with cultural issues. Many of them have to do with technology. A substantial number of them have to do with privacy concerns. Each department functions differently, so there are many standards that apply.
TB: Does this policy push for the release of any and all data the state now holds — contributions to politicians’ campaigns, say, or the results of departmental audits?
BF: It’s hard to encapsulate all things within one definition. The goal here was to put in state law a single unifying commission that is in charge with moving the state toward more proactively open data.
TB: Let’s talk about that for a minute. Your Open Data Policy calls for the creation of a Council on Open Data. Why is a council necessary?
BF: To do this well, it has to be really thoughtful, and that can only happen with lots of stakeholders at the table, with an identified process, with a permanent set of people who are doing this on a regular basis.
TB: OK, but that council includes more than 30 people. What will prevent someone from stonewalling a request for state data, or making the argument that certain types of state data shouldn’t be available?
BF: We started saying, ‘Let’s list everything we wish we had, and see if we can capture that in some sort of theme.’ The more we started to list, the more additional items became questionable whether they should be regularly open or not. If we have all these stakeholders at the table, they are probably the best ones who can make a data-item by data-item decision about whether information should be open publicly.
TB: Then what are some examples of data lawmakers say shouldn’t be open publicly?
BF: Individualized data about a person’s tolls paid by EZPass and personal information tied to license plates, for example, can be problematic.
The two biggest roadblocks to open data: FERPA and HIPAA [laws governing student data and health data, respectively]. Those two federal laws have very big implications for states and individual actors, and are probably the two most challenging compliance frameworks that make agencies hesitant to move too quickly toward open data. The fear of violating those federal laws prevents the disclosure of lots of issues.
TB: Take HIPAA, for a moment. What data governed by that law couldn’t be disclosed, and what could?
BF: Individualized data about Medicaid reimbursements for medical services, even if the individual is not identified in the data. If date and time of services are provided, it can be matched back to a particular patient, which would be a violation of HIPAA. On the other hand, aggregated data about all Medicaid reimbursements at a single facility would clearly be public and not a violation of HIPAA.
TB: The bill calls for data to be collected in ‘non-narrative’ form. So we’re talking pulling numbers out of summarized PDF reports, right?
BF: The goal would be that we could get all data in a machine-readable format [editor’s note: read this as a format that Microsoft Excel or more advanced software could parse easily] that could be released. The problem is that every agency has very different data retention policies, publication policies. There are lots of legacy systems in the public sector … so there’s a need for greater consistency about how we collect, store and use data in the public sector.
TB: There’s a section of this bill that says data that is not open includes any information that would “impose an undue financial, operational, or administrative burden on a state entity.” Does that mean audits?
BF: A lot of this is Public Information Act stuff. If you have a culture of open data where all the data that can be released by law is being publicly released, then the PIA becomes less relevant.
TB: You represent Maryland’s 46th District in Baltimore city, and I’m sure you’re no stranger to the talk in the city about auditing city agencies. How do you feel about audits?
BF: I generally believe that audits are beneficial to all systems, be it public sector private sector. That said, they are snapshots in time, so they don’t necessarily show historical trends. … And I would say that I default toward auditing being a helpful procedure, but can understand, in an environment of budget constraints where agencies want to focus on getting work done, not having to worry about the balance of taking people off the task of doing work to study the work that’s already been done.
TB: Maryland has received “F” grades in the past for not being forthright when it comes to releasing data. What are the chances of this Open Data Policy passing, and do you think it really will change the culture, as you hope?
BF: I do — I really feel confident. The vast majority of the folks within the public sector of Maryland want to provide more open data on a more regular basis. We know outright that this bill will not be the universal change that we all may hope. But it’s an important first step.-30-