That too is good news, but I think it is important to identify the source of the licensing language and the larger issue at play. First, the licensing terms, including the disrepute provision, have been used by the government for several years. The licence terms at Agriculture and Agri-Food Canada, which has offered open data for several years, features the same language on a webpage that was last modified in 2008. In fact, the GeoConnections program, which disseminates geographic data, published a 184 page best practices guide in 2008 (and that was version 2) that discusses licensing terms in great detail and includes several samples. In each case, the licence includes the disrepute provision. While it may be true that few people ever read the licence – Transport Canada published the new GC Open Data Portal licence weeks before yesterday’s launch and no one seemed to notice – the terms are important both because they can be used to later restrict activities and because they reflect the government’s view of the rights of Canadians to their data.
The government may revise the licence by removing the disrepute term, but I think a larger issue will remain.
But open government data is about more than public access and use. It is also about a change in control, where governments recognize that they are largely relinquishing control over much of their data and how it is used. For governments accustomed to crown copyright and the ability to assert control (the U.S. has the most open licence because it started from the premise that it does not have control over public data), this is an even bigger change than public access.
It is readily apparent that many governments have recognized the shift in control – Australia and New Zealand both publish many of their datasets using Creative Commons licences, which are understandable to the average person and require nothing more than attribution. The right to control or impose restrictions and conditions on how the information is used is gone. The UK open government licence contains a few more conditions, but it too resembles the Creative Commons model.
The Canadian model, on the other hand, is a six page, 2,000 word licence agreement that is incomprehensible to the average citizen. If licences could talk, this one would say “this is our data and here is how we the government will allow you the public to use it.” But open government means accepting that government data is the public’s data and that the government’s obligation is not to control it, but to make it as freely and unconditionally available to the public as reasonably possible. The right approach in addressing concerns over the new Canada open data portal is not to make a small change in the licence terms by dropping the disrepute provision. It is to drop the current licence altogether, instead adopting a simplified, open licence that tells Canadians it is their data and (subject to reasonable attribution requirements) they are free to access, use, and reuse it without restrictions.
Dr Geist, is there anything out there that would give a rationale for why the “disrepute” clause is there in the first place? For instance, is there, under Canadian law, a legal liability that the government has with respect to works derived from government furnished information? I agree that this clause shouldn’t be there, but a bit more background may be beneficial for your readers.
The commons problem for Canada
Thanks for the post Michael. My most immediate reaction to the huge licensing agreement is that restrictive agreements like this are characteristic of the current federal government. I was impressed that you and David and the OpenGov folks have got this far in pursuing an open model. Congrats on that front. What I’m not surprised about is how difficult it is to distribute/use the information. Given the current quandary the feds are in over secrecy/obstruction and the like, this is just another case where they look like they’re doing the right thing, but it comes with huge conditions.
This very issue of information control is playing out on our TV sets as we speak. No one likes to give up control, ahem .. **AA, and governments are not different.
Confused about this….
If it is open, even with restrictive terms, you can still comment on the data and do analysis under fair use can’t you? If the data is there to back up your analysis then you wouldn’t have anything to worry about….
Or am I being nieve….
It’s our data, we don’t need to “license” our own data. The license terms existing in the first place is illegal.
Where’s the Data?
I was just at the Open Data site, and decided to Browse the collections. There is nothing in them except item headings. I clicked the category, eg Biology and Ecology, a list of items shows up, and the clickable links lead to “The Page Cannot Be Found” messages. I guess it’s open because there is nothing to take…
It’s our data, we don’t need to “license” our own data.
The license terms existing in the first place is illegal.