Government Operations Committee on May 13th, 2014

[Prof. Mary Francoli]

I think the geospatial data was already out there. There was already a lot of it. What happened when the data.gc.ca portal went up is that the existing data sets got amalgamated and pushed into that common portal.

A lot of the data on some of the areas that you mentioned and that I referred to earlier are data sets that might be seen as a little more sensitive. They are data sets that often have personally identifying information. I think that issues of privacy and respecting the Privacy Act are some of the main concerns about making that data public. That data needs to be really cleaned to make sure that personally identifying information is not part of the data sets.

[Mathieu Ravignat, NDP, Pontiac, QC]

Do you know of plans to clean that data and make it available?

[Prof. Mary Francoli]

That I don't know. Treasury Board did say, during the course of the interviews, that there are efforts to expand the data on the portal but I don't know exactly what kind of data.

[Mathieu Ravignat, NDP, Pontiac, QC]

I want to come back to what I think was Mr. Aspin's question, a kind of comparative, on how the Obama administration talked about “open by default” government. This government hasn't gone there yet. In the case of open data, do you have a sense of what that might look like if we were to go to a kind of open by default policy?

[Prof. Mary Francoli]

I'm not sure the open by default policy necessarily will have the greatest effect on open data. The open data initiative is really pushing data sets out there so it would be like, okay, if you have new data sets that are developed, then that information will automatically be made public. But there are parameters around that as well, so again, relating to things like privacy and national security and making sure that you're not releasing data that has potential negative consequences.

I think perhaps a bigger impact of that open by default policy will be on more the unstructured information, so information that Canadians would typically get via the access to information system, access to information request. I think that's where you would probably see the primary difference. We do have, in our national action plan, a commitment to developing an open government directive, which, as I understand it, is leading us down a path of that open by default framework that's been adopted already in the United States. This was something that we had committed in our national action plan to having in place earlier this year, but it's one of the commitments we're actually a little bit behind on. We are ahead in some commitment areas and behind in some. That was one that we're behind in.

[Mathieu Ravignat, NDP, Pontiac, QC]

With regard to the results of your research and recommendations going forward, has there been, between yourself and the government, an open exchange of ideas? Have your recommendations been well received and do you expect them to act on some of your recommendations?

[Prof. Mary Francoli]

I hope that they pay careful attention to the recommendations. They're not, specifically, my own feelings about open government, so they were largely based on what I heard from the majority of stakeholders and things that were supported by documentary analysis as well. I certainly hope they pay attention to that.

[Mathieu Ravignat, NDP, Pontiac, QC]

Have you met with Treasury Board officials?

[Prof. Mary Francoli]

I have not met with anybody from Treasury Board.

[Mathieu Ravignat, NDP, Pontiac, QC]

Have you requested to meet with them?

[Prof. Mary Francoli]

I sent a letter. The way the OGP IRM process works is that the national governments get to see a copy of your report prior to it going public. They don't get to veto anything in there but they do get to say, “Hey, there's a factual error or a problem here”. So that process happened and I did get feedback from the Treasury Board Secretariat in response to which I wrote them a letter addressing each of the points that they had made and offering to meet with them to tell them a little bit more about what I had learned and what's going on. I haven't met with them since the report was published. They did schedule a meeting between me and two other people, which they cancelled the morning of and then I never heard anything again.

[The Chair, NDP, Pierre-Luc Dusseault]

Thank you, Ms. Francoli and Mr. Ravignat.

Mr. Adler, it's your turn now for five minutes.

[Mark Adler, Conservative, York Centre, ON]

Thank you, Chair.

Thank you, witnesses, for being here today.

I do want to just begin with Ms. Francoli. We're pretty much in the infancy of open data. Is there anybody, any jurisdiction out there, in your estimation, that so far is getting it just right and is sort of the gold standard at this point?

[Prof. Mary Francoli]

That's really difficult. Different jurisdictions have approached it in different ways and I've been looking primarily at the national levels. I think really interesting things are going on at the municipal and provincial levels but most of my research is focused on the national level. If I look around at what's been going on in other countries, I've certainly seen some other initiatives that were really interesting. The U.K., as one of the committee members noted, led the charge towards the open licence, so we're looking to other countries like the U.K. and the U.S. and modelling some of our initiatives on them. So in terms of best practices we can say that's happening. There's the open licence as an example, and the open by default from the U.S. where we're modelling our open government directive on that.

One thing the U.K. has also done that was really interesting is a big data audit of their data holdings, and in addition to publishing a range of data sets they've also published the list of unpublished data sets. They've been very transparent with society to say, look, this is what you have, this is what you don't have. It gives them a mechanism to refer to and to go through and say maybe we can work on releasing this and this in the future. So there's a very clear understanding of what's there.

[Mark Adler, Conservative, York Centre, ON]

Okay.

Since you've been studying and comparing national jurisdictions, you're probably in a good position to answer this question. In terms of a lexicon, or what standard definitions we may have or what certain sets of data define themselves as, is it uniform what you've been finding across jurisdictions, or does it vary from jurisdiction to jurisdiction? Could that create a problem, if it does vary, and should we come up with a standard definition of what specific words mean and have that universally applied?

[Prof. Mary Francoli]

I have seen glossaries of what the different terminology means when it comes to open data. On the data.gc.ca website there is a list of things, if you want to look up the background of open data.

I think, though, there's a divide that's happened. I think the divide exists between government as well as those involved in public administration and those with expertise in the area of data use. Data scientists are probably the best example of that. Governments, for example, seize data sets in a bit of a smaller way than the data scientists who want to use the data. I think there's room for more ongoing consultation and more ongoing dialogue between those two groups in particular to say, okay, well, what is it you need; what makes a complete data set?

A complete data set isn't maybe just a range of data sets with the same information for different provinces; maybe it's one big data set with all the provincial data in it. That data set can be easily comparable, or combined with other data sets, so that it can be used well. I think that conversation needs to happen a little bit more.

But as you say, we're still in the infancy of this. There's certainly room for that conversation to continue.

[Mark Adler, Conservative, York Centre, ON]

Yes.

You can't just take data, dump it out there, and call it open. Do you find that government has undertaken enough of a public awareness campaign that, first of all, this data is available, and second, what this data may mean and how it could be helpful to you, the citizenry—

[Prof. Mary Francoli]

No, I don't think—

[Mark Adler, Conservative, York Centre, ON]

—and one that cuts across all jurisdictions?

[Prof. Mary Francoli]

Again, I can't speak for the provinces or the municipalities—

[Mark Adler, Conservative, York Centre, ON]

It isn't what you've studied.

[Prof. Mary Francoli]

—but at the federal level, certainly we haven't had that great of a campaign.

The one instance that really pops into my mind where that might have happened in an interesting and good way is the CODE appathon that the Treasury Board Secretariat sponsored to really say to people, here are that data sets we have on our portal, take them, do something interesting with them, and we'll publicize that. That was an interesting initiative.

But there's no communication, really, around the fact that we're a member of the Open Government Partnership, let alone the various commitments we have.

[Mark Adler, Conservative, York Centre, ON]

You feel it would be very helpful if that were something that we undertook.

[Prof. Mary Francoli]

I think so. People need to know, in the context of open government in general, and open data as well, as part of open government, that we're doing interesting things as a country. There's room for improvement, but that improvement won't happen unless there's dialogue between different actors and unless citizens are engaged in a meaningful and ongoing way. Certainly that requires some sort of information or public relations campaign so that people know about it in the first place.

[Mark Adler, Conservative, York Centre, ON]

Great. Thank you.