Journalists have been making requests for electronic data to federal departments for a good decade and a half. And while the fights have sometimes been long, such as the CBC’s battle for adverse drug reaction data or my fight for aviation safety information, the end result has been that the data was released.

But a new trend is creeping in, and has been for well over a year. Departments, citing concerns about possible inadvertent release of sensitive information, are using software to convert ATIP release packages to images.

This works fine for paper records. An image of a piece of paper, as long as it is legible and printable, is about the same as the piece of paper itself. It may actually be better because it saves on photocopy charges to the requester.

But when you convert data to an image, you fundamentally change its character. Suddenly, it’s not data anymore, but a picture of data. You can’t import it into a database or spreadsheet to look at it unless you first go through a time-consuming and imperfect process of using an optical character recognition program to extract the text from the image.

I know of several people–I am one of them–who have run into this roadblock.

The problem seems rooted in one or more requests, not data requests I might add, in which mistakes were made by civil servants processing ATIP requests, and sensitive information was left in the electronic files in a way that it could be recovered.

In late 2007, the Treasury Board Secretariat, which oversees access to information in the federal government, advised departments to “cease responding to ATIP requests in electronic format until they are certain that any potential risks (of inadvertently releasing severed information) have been addressed.”

Treasury Board officials told me last year there was no blanket ban on releasing electronic information, and indeed one department sent me an Excel file last week. But others are balking, and if this continues, this has the potential to set back access to electronic records to the dark days of the 90s when officials tried to make the argument that database records aren’t records under the act. One department made that very argument last week in justifying its image pdf policy.

I suspect this issue is going to end up before the information commissioner and maybe, in time, the federal court. How it is resolved will be pretty important to anyone asking federal departments for data. 

I should note that so far at least, the image trend appears to be limited to Ottawa, although some provincial departments and municipalities still resist releasing electronic files at all. I will keep you posted.