Government Operations Committee on May 13th, 2014

[Prof. Mary Francoli]

I think our commitments aren't totally out of line with commitments made from other countries, if we look at the context of our national action plan. With the G-8, we were involved in the charter on open data, so we adopted an open data charter with the other G-8 leaders. The charter commits Canada and the other G-8 member countries to a set of norms and standards for the proactive release of more high-quality user-friendly data that's unrestricted in the way people can use it and reuse it. In that sense, it's talking about norms. It's talking about standards and potential for use and reuse.

In terms of my study, I found that across the board various people I interviewed were very worried about the types of data being released. I'm not sure if this is quite what you were getting at, but they were really worried that what we're doing at the federal level is making commitments internationally to transparency and to improving accountability, and those things are good commitments to make. There's potential for better public policy. There's potential for strengthened democracy. But what people are seeing or the way they perceive what's happening right now—and I'll try to say this as neutrally as possible so as to not say that they're right or wrong—you kind of have the rhetoric about transparency and accountability going on, while at the same time we're cutting the origin of the data, so that goes back again to the long-form census. They see the government as saying that they're going to release things, that they're going to try for greater transparency and more accountability, but in doing that, they're very selective about what's being released, because they're cutting the collection of certain types of data, a move perceived by some stakeholders as a bit of information control over what's going on or what's feeding into the portal.

People are very worried about the long-term impact of that. They're worried about the impact of that for transparency and for accountability. There seem to be two things fundamentally fighting with one another there, and they're worried about the impact of that for good policy in the future and for what we'll know about Canadian communities on the smaller scale.

[The Chair, NDP, Pierre-Luc Dusseault]

You have 20 seconds left.

[Anne-Marie Day, NDP, Charlesbourg—Haute-Saint-Charles, QC]

Do you have the impression that there have been improvements?

[Prof. Mary Francoli]

A lot of the data in the data portal is data that was already released to the public. We can think of this as potentially an improvement, although there's work to be done in terms of the actual portal itself. We're seeing the data we have, what we own in Canada, and how it can be structured in a way that makes it more accessible. There's room for improvement, but it's good that those conversations are happening.

[The Chair, NDP, Pierre-Luc Dusseault]

Thank you very much. I have to stop you.

Ms. Day, your time is up.

I will now give the floor to Mr. O'Connor for five minutes.

[Gordon O'Connor, Carleton—Mississippi Mills, CPC]

Ms. Francoli, I have just a quick question.

The Treasury Board claims they have more than 190,000 data sets published, and then I see city and province representatives come here with different numbers. Is there any sense between these numbers? Do they mean the same thing? Are they talking about the same quantities of information or not?

[Prof. Mary Francoli]

I can't speak for the cities. My own research has really focused on the federal level, so I'm not quite sure what the municipalities are defining as their data sets.

It would be logical that the federal government is the owner of quite larger amounts of data than the municipalities. I don't think that would be unusual. There have been some bizarre exercises in counting, if you want to call it that, where I think even at the federal level we've been trying to figure out how many data sets we have.

In my report—and I had provided a link to Marc-Olivier—if we look at various points of time, we can see Treasury Board Secretariat releasing different sets of numbers around the quantity of the data sets that we hold. I can't remember them off the top of my head, but it's up and then it's down, and then it's a little bit up and then it's down again. So I think there is a lot of work to be done around what a data set is. Part of that is figuring out the standards as well. I know that TBS has said that in trying to develop a set of standards, they've worked towards combining some of the data sets and that accounts for some of the fluctuation in the numbers.

[Gordon O'Connor]

Thanks.

Mr. Fortin, we've been briefed from a number of people, and basically the same message keeps coming out that experts are who you're dealing with. Experts go to all the databases. They withdraw the information, manipulate it, etc.

But I believe that our purpose should be the people at large, and I suspect that the people at large don't know about these portals. Certainly they may not know about the federal; maybe they know about yours. They may know, in your city, about transportation or things like that. But I think we have a real problem getting to the people, rather than to the experts.

I just wonder if I could have your opinion.

[Jean-Pierre Fortin]

From the outset, I would say that it is fairly predictable that federal government data are not necessarily directly useful. After all, people are closer to services provided locally, that is, by municipalities or cities. Recreational services, garbage collection and public transit, for example, are things for which cities are responsible. It is normal that, in their daily lives, citizens would rather get information on these types of things.

Further, I agree that, for data to be useful, they must be relevant and easily accessible to citizens. I say this while being aware that these data are extremely useful to administrators. It is important that the government have statistics, so that it can understand its own processes, and that it can understand how its various programs perform. This type of data is not of direct interest to citizens, but it is for the government.

If data is to be accessible and useful to citizens, it must be processed and it must be turned into information. This data must also be collected, used and applied within the framework of applications. These applications could be on a platform provided by government. This would allow for all kinds of questions and analyses, if need be, but also for applications in everyday life, as those mentioned a little earlier by Mr. Deslauriers. For example, these applications could relate to parking, that is, where parking spots are available in a neighbourhood. This is really something that's tangible. This information could be made accessible in real time.

So by using existing systems, we could extract data in real time and transfer them to applications which provide information to people who need it. For example, it could be information about parking, as in this situation, but it could also be information about registration for recreational activities. Any data contained in our systems could be used in applications provided by either a public organization, or by third parties supported by government.

With regard to the infamous programming marathons, Mr. Deslauriers did not talk about their futility, but the fact that their usefulness seems questionable. Indeed, they result in the development of applications which have a limited shelf life. But also, there is the issue of quality, since sometimes these applications are developed too quickly, and also the fact that once these applications have been created, people start to lose interest. So the data will have to be used in applications which will be of lasting usefulness. Of course, I am talking about those applications which will be deemed to be useful and effective.

[The Chair, NDP, Pierre-Luc Dusseault]

Thank you, Mr. Fortin.

Mr. Simms, you have five minutes.

[Scott Simms, Liberal, Bonavista—Gander—Grand Falls—Windsor, NL]

Thank you, sir.

Mr. Hollett, I certainly appreciate where you're coming from in talking about communities and everything else, but even the smallest communities are trying to use this data for several purposes—obviously, for attracting industries, for abandoned plants, that sort of thing. But you said something that caught my attention earlier, which was that they were surprised you would buy this from Statistics Canada and then release it to the public for free.

How does that work? What exactly are you buying? What kind of contract do you have?

[Alton Hollett]

I was thinking specifically of the data that we have in the community accounts, and of course, there's a wide range of data there because there are a large number of domains—income, employment, demographics, and on and on. We buy any data that we don't have readily available. We have historically bought the data from Statistics Canada in a generally pretty raw form, and then we've prepared it for the community accounts. So it's basically any data that we needed for the community accounts that were required to respond to the framework that we've established.

[Scott Simms, Liberal, Bonavista—Gander—Grand Falls—Windsor, NL]

You mentioned the term “well-being framework”. Obviously this relates to what is user-friendly regarding open data for people to use. In many cases, I agree that there are tools set up, but people don't know they are there and therefore don't utilize them enough.

I think about certain industries and companies that want to set up in certain areas, accessing a concept that we talk about, which is a skill set inventory. There has always been the traditional circumstance in which people try to find work by going to a particular website to find data for open jobs. But how can commercial enterprises find out what skills are available in a certain area in terms of people working in that region? Much of this stuff just exists as raw data. Nobody really packages it as something accessible.

I guess I'm asking all three of you this question. What are some of the best practices by which municipalities or the province or the federal government are packaging this data in such a way as to allow people to use it to better their communities?

[Alton Hollett]

That's an interesting question. This was one thing that led us, when we were building the community accounts, to put in an organizational framework for the data. When you talk about people who are interested in labour market-related data, for instance, that was one area we specifically gave a lot of thought to.

It goes back to what I said earlier. If you ask people what they want, quite often they don't know. Economists have ways of thinking about this. We look at our labour market and our labour market participants in a certain way. A lot of this is quite simplistic, as you'd be aware: males, females, age, wage rates, and so on. There are many other categories.

We began to set the data up so that those kinds of data were readily available. Instead of leaving people saying that they want to know what the unemployment rate is but would like to know more about the labour market and skills and what type of people are in the community and so on, we set it up along the lines of how an economist would think of it, which is basically as a description of the way the world works. That was very helpful to people for understanding what kind of occupations people are in now in our communities.

There are a couple of other things that we're doing as a government. One of them is obviously.... With our tight labour markets in recent years, it's a big issue. What we've been doing is developing occupational profiles for people in different occupations. We're also now building occupational projection models, whereby we take the forecasts we do at our branch, look at labour demand, and then look at labour supply and at where the mismatches are and provide occupational profiles—for individuals, to say where the jobs might be; and of course for companies, to give them a better sense of where the labour markets might be tight.

On that particular subject, it's not difficult to provide general information so that people can understand what is happening in communities from a labour market perspective. But when you start moving forward and asking about companies, or when you dig into the industries and that sort of thing, it requires a variety of measures, really.

[The Chair, NDP, Pierre-Luc Dusseault]

Thank you. The time is up. I'm sorry. Maybe you will have a follow-up question later today.

Mr. Aspin, you have five minutes.

[Jay Aspin, Conservative, Nipissing—Timiskaming, ON]

Thanks, Chair, and welcome to our guests. Thank you for helping us with our study.

I'm going to begin with a question to Ms. Francoli. I'd be interested in getting your assessment of how Canada compares with other countries in terms of open data.

[Prof. Mary Francoli]

That's a bit of a tricky question, given my position as the independent researcher for the reporting mechanism that is part of the OGP. The independent reporting mechanism tries not to rank countries in terms of who is and who is not doing the best.

You can see some reports out there. There is one released in February of 2013 by Capgemini that you might want to look at. I can provide a link to the committee, if you're interested. It did that kind of comparative ranking. It ranked Canada as a trendsetter in the area of open data, but a trendsetter behind the U.K. and behind the U.S.

Their metrics, which I haven't had time to critically evaluate, put us behind those two countries because of such things as the lack of diversity or breadth in the data sets that are out there—that goes back to the point I made earlier about the dominance of geospatial data—and also the lack of a good forum for engagement and lack of a good mechanism for ongoing, sustained engagements. Those two things together set us back a little bit.

[Jay Aspin, Conservative, Nipissing—Timiskaming, ON]

Thank you.

To our other three guests with the province and the municipalities, I'd like a comment from each one of you. Are the provincial and municipal governments collaborating to link the data available on their respective open data sites?

Perhaps we could start with you, Mr. Hollett.

[Alton Hollett]

I think the answer is that's not very well advanced in our province right now. I know that communities use our data an awful lot. I get contacts from mayors throughout our province all the time, quite often thanking us and saying that they wouldn't have too much if we weren't there with our data. But I think there's really a lot to be done in terms of collaboration between the municipalities and our group—for sure—which will be a data leader in Newfoundland and Labrador.

[Jay Aspin, Conservative, Nipissing—Timiskaming, ON]

Thank you.

Mr. Deslauriers.

[Denis Deslauriers]

We are working closely with the provincial government. This initiative brings together many municipalities. In Quebec, the directors of IT services in cities of 100,000 people or more meet every six weeks for different projects. We have united our efforts to meet with the provincial government and to ask for its collaboration.

We therefore already have a common licence for the entire province as far as open data is concerned, as well as a first standardized dataset, which is about events and ideas for things to do. Further, we are in the process of building a common portal, which is managed by an NPO, in which we will deposit our respective data.

[Jay Aspin, Conservative, Nipissing—Timiskaming, ON]

Thank you.

Our friend from Montreal....

[Jean-Pierre Fortin]

I would like to add something to what Mr. Deslauriers from Quebec City said.

There is also an initiative where Canadian cities collaborate. Our group of Quebec cities is also a member of the Municipal Information Systems Association of Canada. This association brings together the people responsible for computer services in Canadian cities, especially the big cities. Within the association, there is a specific interest group on open data. We are in the process of getting organized to at least establish basic standards and criteria with regard to data so that they can be used by everyone.

Over the last few weeks, one person in this group was chosen to work in close collaboration with the Chief Information Officer Branch of the Treasury Board Secretariat of Canada to help Canadian municipalities work together in the interest of creating a common space or, at the very least, of creating conditions conducive to everyone being able to use these data.

[The Chair, NDP, Pierre-Luc Dusseault]

Thank you, Mr. Aspin.

Mr. Ravignat, you have the floor for five minutes.

[Mathieu Ravignat, NDP, Pontiac, QC]

I'd like to dig a little deeper on the lack of diversity in the data sets that are available on the portal, particularly the lack of useful social data, which no doubt—and I fully agree with the stakeholders and with you—would have been better if the long-form census hadn't been eliminated.

You said in your presentation, Madam Francoli, that there was a lack of data on aboriginal people and seniors. I'd like to know why you think that is.