Tough To Get Data From San Diego’s Open Data Initiative
Thursday, February 4, 2016
It's been more than a year and a half since the city of San Diego launched its open data initiative to make public all sorts of information — from city salaries to parking ticket details — but not much is there for the viewing.
The city of San Diego launched its open data initiative to much fanfare in July 2014, promising an era of transparency and robust citizen engagement around massive amounts of information the city collects and stores.
The promise was that information on everything from employee salaries to parking tickets would be easily accessible to regular San Diegans.
More than a year and a half after that launch, city officials have compiled a list of some 2,000 data sets, half of which they consider to be public. However, despite official requests from inewsource, the city has refused to release the list. inewsource has not asked for the data itself.
What’s more, it is not clear if San Diegans will ever get an accounting of all the data. It’s a question inewsource has been asking since October, with little to show for it.
Peter Scheer, a First Amendment expert, was skeptical of the city’s way of doing business.
“I can’t see any legitimate reason for not releasing a list of data sets — an inventory or master list of the data that the city possesses,” Scheer said.
The City Council members who advocated for the open data initiative — Mark Kersey and Sherri Lightner — declined to be interviewed about the transparency of the process. Their representatives said they were satisfied with how the initiative was progressing.
City officials who head departments working to implement the initiative said the data collected so far was in draft form and not subject to public disclosure. They also cite cyber security risks. None could explain, however, why a list of the nearly 1,000 data sets collected by the city and deemed public could not be released.
The city’s data policy, approved by the City Council in the fall of 2014, calls for posting “an inventory and a summary description of public data sets under the control of each city department.”
The policy defines a public data set as information “that is available for inspection by the public in accordance with any provision of law and is maintained on a (city) computer system.”
The policy then lists eight broad exceptions to that definition such as “any portion of a data set to which the City may deny access under applicable federal, state or local law, rule or regulation.”
In a July update to the City Council, the city’s Performance and Analytics Department — the department charged with implementing the open data initiative — reported that it had compiled an initial list of all city data sets and that “departments already consider nearly half of the reported datasets to be public.”
inewsource filed a Public Records Act request for that list of data sets in October.
That request was denied and a second, amended request for the list was met with a request for a 60-day extension and no promise that the list would eventually be released.
Under the Public Records Act, agencies must generally determine whether a record is releasable within 10 business days of receiving a request but may ask for an extension under unusual circumstances.
Almis Udrys, director of the Performance and Analytics Department, declined an interview request but answered questions about the denial and delay by email.
Udrys said that the “initial inventory process” wasn’t complete. He said his team needed more time to “validate the self-reported information” provided by city departments to eliminate redundancies, check for accuracy and ensure no private information was included in the data sets.
He said that the department intended to post a limited selection of the data inventory on the city’s website in the coming months so the public could comment on which initial datasets to publish to a data portal scheduled for launch by July 1.
Udrys wrote that the two-month delay in responding to inewsource’s second request for information was necessary because “we have not worked through all of the potential concerns and are proceeding very carefully to make sure we are expeditious without exposing the City to unnecessary risk.”
He did not explain how releasing a list of data set names could expose the city to “unnecessary risk.”
The two City Council members who drafted the open data initiative dismissed questions about the process, saying they were satisfied with the progress of the program.
Jennifer Kearns, director of communications for Council President Lightner, said the councilwoman felt Performance and Analytics was “moving forward in a really timely manner” and was “really satisfied with the progress they’ve made so far.”
When asked what justification the city had for delaying release of the list of data sets’ names, Kearns stressed that she hadn’t read inewsource’s records requests and offered two possibilities.
First, she echoed Udrys’ concerns over the accuracy of the list.
"You want to make sure the list itself is valid,” Kearns said, noting that somebody could have named a data set in a misleading way or that some data sets in the city’s possession could be left off the list.
Second, she offered that the list might not be releasable at all.
"It's still a work-in-progress, and things that are considered a work-in-progress are not subject to a PRA [request]."
Gina Jacobs, director of communications for Councilman Kersey, provided a statement from the councilman in which he said he was pleased with the implementation of the open data Initiative.
“My understanding is we are on track to release the inventory and data sets when expected,” the statement read. “There will be regular reports to Council and I will be able to ask questions about data sets and the decision making process. If I think there is something amiss or don’t agree with their findings, we can discuss it at that time.”
Scheer, the executive director of the open government advocacy group the First Amendment Coalition, was hard-pressed to come up with a reason the city couldn’t release a list of data set names.
“That’s a separate question from whether they have to, if requested, release the underlying data sets in their entirety but we’re not talking about that,” Scheer said.
“It will almost always be the case that the city can’t necessarily vouch for the accuracy of all records or all data it may hold in one way or another, in one department or another, but that is not a basis for refusing to let the public know that the data set exists, which is what they seem to be trying to do here.”
To view PDF documents, Download Acrobat Reader.