Government Operations Committee on April 3rd, 2014

[The Chair, NDP, Pierre-Luc Dusseault]

Order, please. We will begin immediately.

We have a busy day today. We are hearing from four expert witnesses for our study on open data. I will introduce them right away. We have David Eaves, open data expert, appearing as an individual. We also have Renée Miller, who is a professor at University of Toronto's Department of Computer Science. By videoconference, we will hear from Ginny Dybenko, Executive Director of the University of Waterloo's Stratford Campus, and Mark Gayler, from Microsoft Canada. Since Mr. Gayler is in Vancouver, where it is currently 5:45 a.m., we will show some indulgence toward him.

Without further ado, let's start with the presentations of up to 10 minutes for each witness. Afterwards, the committee members will have an opportunity to ask the witnesses some questions.

I yield the floor to you, Mr. Eaves. Thank you for joining us and for taking the time to testify before our committee today.

[David Eaves, Open Data Consultant, As an Individual]

Thank you.

Just so Mark doesn't get all the credit, I'd like to note that I also just came in from Vancouver, so it's also 5:45 in the morning for me.

[Voices]

Oh, oh!

[David Eaves]

My name is David Eaves. Since I'm listed here as speaking “As an Individual” and don't have the credentials of some of my peers, perhaps I'll start with a little bit of background on me.

For the last five, six, seven years, I have been working to make open data happen in Canada. I wrote the original motion that led to the creation of the open data portal with the City of Vancouver. I then worked behind the scenes with some of the provinces to help them create their open data initiatives. I gently applied pressure on the federal government to persuade them to adopt open data as a policy.

I also work with several governments. I ran the boot camp for the Presidential Innovation Fellows at the White House. I've worked for the State Department and the World Bank. I sit on Mr. Clement's open government advisory panel as well as Premier Wynne's open government task force, which recently released its results. I sit also as an affiliate at the Center for Internet and Society at Harvard University, at the Berkman Center. I also sit on the boards of several non-profits as well as several start-ups, both in the open data space and outside.

I want to share a few thoughts with you about what I think matters about open data, how we're doing, and some of the things we could be doing. Maybe just as a little bit of a backdrop—I imagine everybody is trying to explain to you what open data is and why it matters—I'll give you a simple metaphor.

I'm carrying around with me a Fitbit, a small device that tracks how many steps I take every single day. This is mostly because I have the potentially bad belief that if I take 10,000 steps, I can eat whatever I want. So I try to get to 10,000. When you look at this device, it's tracking some data about me, specifically my movement. Increasingly as you look around, all sorts of data is being tracked about you and created about you, from your bank statement to your mortgage to where you're going. This device happens to know where that happens to be all the time as well. It would be nice to think that you could harness all that information to tell you something useful about your life that could cause you to change your behaviour, or to do something different, or to save a little bit of money.

I'd like to apply that metaphor to the federal government. Right now there are probably about a billion of these types of devices. Whether there are people tracking expenses in Excel spreadsheets, or devices measuring the weather, the temperature, or something else around the country, around the world, all of that data is being collected. Wouldn't it be nice if we had access to it so that we could say something intelligent about this country and about our community, and maybe change some behaviour here, or figure things out that are not going well?

I think the open data initiative is trying to solve the same problem that many people are trying to solve on the consumer page: how do you harness all of this data that's being created, some of which you don't even know is being created, or where? Can you bring it into a central place where it becomes useful, actionable, and leverageable by a community of people?

Hopefully that gives you a metaphor that makes it a little bit easier to understand what open data is and how potentially it can be useful.

I think for me, there are one or two examples that strike me as the most interesting around how far we've come and how far we have yet to go. I'm sure this committee is interested in knowing, as everyone is, how we are doing internationally. I would argue that internationally Canada is doing relatively well. We're not what I would consider to be a front-running leader. We're not like the United States or the United Kingdom. But we're also not a laggard. Maybe only 20% of the countries in the world are thinking very, very hard about open data, and we sit very comfortably in that group.

The real danger I would flag around this is that I think using international comparisons, especially this early on, and any time in government, is always enormously dangerous. I get very frustrated when I see comparisons about how governments are performing in technology and then people becoming satisfied about being at the top of those rankings. Whenever you have a ranking of government performance in technology, what you are actually doing is you're taking all the slowest movers in a space, putting them into one category, comparing yourself with all the other laggards, and then sitting around and congratulating yourself for being really, really fast against the other slow-moving players in this space.

Leadership, for me, is not whether you're the second bull in a herd. The problem with leadership is that you're in the herd to begin with, and real leadership is how you actually break away and go do something that other people are not willing to do. Whether it's within a country or internationally, the herd mentality is so strong that I think it actually prevents leadership from happening. If you benchmark yourself against others, what you're really doing is saying that you just want to be inside the herd, and then asking where you rank in that herd, as opposed to really thinking about what leadership and transformation could look like and doing the things that could potentially really change society in a positive way.

While I think the international metrics matter, I caution you strongly about getting sucked into them and somehow believing that they're a magical metric that should determine whether we're happy with our performance or not, because they're almost invariably very, very poor.

The other question for me is, what's our goal? What are we trying to accomplish? When I look around I see different players doing different things, and they have, I think, a real vision for where they want open data to take them. I think that vision is less clear in Canada. Certainly, we're not realizing our full potential. I suspect that many on this committee are most concerned with the economic benefits of open data. I think those could be significant. I think there's a real risk of overplaying them, and I think there's an enormous amount of hype. I would be very, very cautious about believing every figure that passes by you, or why it's going to have an economic impact—and I'll talk about that briefly. There are also huge opportunities around making government more transparent, which actually has economic benefits in and of itself, as well as more accountable. I wouldn't want those to be lost.

The third is there are huge opportunities in reshaping how public servants work with one another and in using open data to vastly improve the efficiency and productivity of public servants. I wouldn't want that opportunity lost in the pursuit of economic goals.

With that said, I think there are four big ways that open data will serve to transform a society. Let me highlight each one of these, and what I think matters and doesn't matter about them.

The first, which I'm sure you've heard endlessly about, is the opportunity around apps. I'm not the first person...and I don't want to say that apps don't matter. I think apps are enormously important, and there is an enormous opportunity there, but I actually feel it's also the least relevant of all the opportunities before us, especially when it comes to federal government data. A great deal of federal government data is aggregated at such a level that it becomes pretty hard to do anything particularly interesting with it. In addition, the vast majority of the data is created for policy usage and not for day-to-day application usage. There are definitely datasets out there that are very, very interesting. Border wait times, which you heard Colin McKay talk about, I think, is a great example of that.

Operational data is very interesting, but the vast majority of your data is actually geared toward policy analysts, so it's geared to trying to do analysis and understanding what's going on in society or what's going on in the community.

That brings me to the second big place where I think we're going to have impact, and this is the one that I think is the 800-pound gorilla in the room, which is the opportunity for open data to dramatically improve analysis and productivity. The example I like to use is one that I think comes a little out of left field. I'm involved in something called the Canadian Boreal Forest Agreement, which is an agreement between the largest environmental groups in the country and the forestry industry. It's all about trying to ascertain where logging should take place in order to preserve woodland caribou and maximize the benefits for regional communities that make use of logging infrastructure.

This entire agreement is made possible because of enormous amounts of data about knowing where the woodland caribou is, about knowing where different types of trees are, and being able to layer maps over one another to figure out where the places are that we shouldn't be logging and where the places are that we should be logging.

There's no app that's going to come out this. But if you look at the total impact the CBFA could have on the Canadian economy, it could be in the billions. If you talk about no longer having protests against forestry companies by environmental groups, they're actually supporting logging companies as they try to sell their products internationally, having a wood product that is actually seen as ecologically viable and therefore more valuable, about the impact on local communities and the jobs it creates, the impact on shareholder value around all of the different logging companies, it is not hard to imagine that number very quickly running into the hundreds of millions, and even the billions.

That entire project is supported by government data. So for me, rather than focusing merely on apps, thinking about the much larger policy opportunities and the economic opportunities around analysis that open data can provide is the place where I think particularly federal government data becomes enormously valuable and interesting.

The third is the internal use of open data and the way I think it can transform how our government works. I've been around the world talking to people who run open data portals, and invariably you find that roughly 30% of the users of an open data portal come from computers that are located within the government that made that open data available. It's not hard to understand why. The data government creates is most useful to people who work in government. The problem was that before you had an open data portal, in order to make use of a dataset I would have to go and talk to you, and then your manager, and then maybe your ministry's lawyers had to get involved before they decided who was allowed to use this data or not.

You had nine meetings, eating up 10 public servants, 40 hours for a week, just so I could get access to data, and most of the time we're like, forget it; I don't want to even bother anymore.

Now all of a sudden, not all, but a significant amount of government data is available in a place where public servants can very quickly access it. All of that time spent negotiating over whether or not I should have access to something that's actually already a public asset has just disappeared. So the productivity opportunities within government, I think, are quite significant.

[The Chair, NDP, Pierre-Luc Dusseault]

I would ask you to wrap up your presentation, Mr. Eaves.

[David Eaves]

Yes.

If I were looking at this committee and I was trying to think about how we were going to assess the value of Canada's open data policies, I'd be looking at three things.

The first thing I'd be looking at is whether we are thinking strategically about the policy in economic areas that we want to be driving into and what the data is that we're releasing that might support those places.

The second is whether we are thinking hard about how government itself is using the open data that it releases, so it does what we call “dogfooding”, which is that it uses its own information rather than sharing with others and expecting them to use it, but using something completely different itself.

The third is whether we are actually sharing information about government itself. Where is the budget? Where are the things that make government transparent so that citizens themselves can better understand and make government more legible, so they can become more engaged in the political process and contribute in interesting ways in the policy debates?

Thank you.

[The Chair, NDP, Pierre-Luc Dusseault]

Thank you for your presentation.

I now yield the floor to Ms. Miller, from the University of Toronto.

[Dr. Renée Miller, Professor, Department of Computer Science, University of Toronto]

I would like to thank the members of the committee for inviting me to participate on this important panel.

I am a computer science researcher, and I study the problems and the opportunities that open data presents to the science of computers. In particular I study the problem of data curation, which can briefly be defined as ensuring that data maintains its value over its lifetime, ensuring there is still value in that data and that value can be used by humans.

I would like to present three points. I'm going to try to reduce the geek level. I realize I am used to talking to computer science students so please let me know if you don't understand anything I'm saying here.

I have three points that I think can help Canada become a leader in the open data revolution. My first point is that I think the open data portal should adapt the principles of open link data. So when we put a file up on the web we are using technology that's been around for even more than 20 years. Since the beginning of the web we've been able to share data in files over the web. The state-of-the-art data sharing is not just sharing these static, inanimate files. When I say “we” I mean scientists like myself, academics but also industry leaders. When we share data we share data that's linked and that means the objects we're referring to in the datasets are dereferenceable, it's a fancy geek term meaning I can click on it. When I click on it I get important and interesting information about that object and among that important and interesting information I get relationships to other objects and important information about them.

So let me give you a concrete example of that. The most downloaded file from the open data portal in February 2014 was a file about charities. It's a static file. It just has strings and it has numbers in it. So it has strings naming different charities and it has facts about those charities, but it's just a dead file. What I would like is when I download that file, that when I see, say, the Rideau Street Soup Kitchen, I would like to be able to click on that link and get important information about the soup kitchen. For example, I'd like to get important information of where it's located, what community it serves, how many people it serves, some of the facts, some of the data. How much federal money it gets is in that file but other information like whether it gets provincial funding, private funding, who those private funding agencies are and information about them, that's not there.

But it's very easy to provide using today's technology to make the data linkable and to use the principles of open link data to enrich this data. So that's my first point, to embrace the principles of open link data.

My second point, which is highly related, is that open data is about information flow and that information flow can't be unidirectional. If the flow of information is solely from the government to the public then there's no incentive for people to do interesting and creative things with that data. So if we just make the data available, take a data file and throw it over the transom, close our eyes and hope somebody is going to do something interesting with it, we're not creating the incentives to get people involved, to change lives with the data, to solve problems, improve government, just for the economy.

Worse, I think it has the potential of creating this adversarial relationship. It gives the perception that the government's in control of the data, and is just handing it out. There is no ownership or investment in the data itself. So I think open data is fundamentally about creating participatory opportunities where people can become invested in that data and are incentivized to contribute to the data itself and incentivized to improve the data and to create new innovative ways of using the data. I think this investment creates trust and people will trust the data if they can contribute to the data. It also provides an information flowback into government—David was speaking to this as well—where the information itself is flowing back into the government improving government decision-making based on better data.

My third point is that opening up data is important, but it's equally important to create and curate participatory opportunities with this data. These are not just appathons. I think they are other ways in which the community can get involved in doing analysis over this data and improving this data.

I know the open data portal is already deeply involved in this. They have the Great Canadian Appathon, which is in its fourth year. It was just at the University of Toronto, so we're already doing quite a bit along these lines.

But I think there are two important outcomes of this. These activities are not just educational; you're not just teaching students how to use data. Rather, you're looking to find those visionary students—I call them students; everybody younger than me is a student—those visionary people who want to create new entrepreneurship opportunities with that data. I think that using open government data is an absolutely terrific way of getting those folks to stay in Canada. Too many of my entrepreneurial students go to Silicon Valley because that's where the air is filled with start-up culture. You can walk into a start-up and somebody will be there to help you with your start-up.

That kind of start-up culture just doesn't exist here in Canada, but if we have more people using community data, open government data, in their start-ups and building things around that, there is an incentive for them to stay within Canada. But, also, in addition to these visionaries who are going to spur our economy, I would encourage you to look to activate the power over the crowd; and by that I mean creating datasets where the community itself can contribute to those datasets. These are things like allowing the community to comment on the number of open beds in homeless shelters and activities like this where the community gets involved in improving the data and gets invested in the data itself. The power of the crowd can be really important in leveraging government open data.

In conclusion, I'm absolutely thrilled to see this initiative in Canada. I think it is a tremendous opportunity here, and Canada has the potential to become a leader in the open data revolution, and I look forward to seeing much more.

[The Chair, NDP, Pierre-Luc Dusseault]

Thank you for your presentation.

We are now moving on to Mr. Gayler, who is testifying by videoconference live from Vancouver.

You have 10 minutes, Mr. Gayler.

[Mark Gayler, Technology Strategist, Western Canada Public Sector, Microsoft Canada Inc.]

Hello, and thank you to the committee for inviting me to participate this morning. It's bright and early in Vancouver.

My name is Mark Gayler. I work for Microsoft Canada. I've been working with Microsoft for more than 10 years. I'm a technology strategist for Microsoft Canada. I work primarily with municipalities. As part of that role, I'm a subject matter expert on open data and open source technologies.

I'd like to comment on a few things. First of all, I very much appreciate the comments by my colleagues David and Ms. Miller just previously.

One of the things I have experience with is working with different governments around the world, and so I've been engaged with open data projects in Canada, but also in the U.S.A.., Colombia, Japan, central and eastern Europe, and the U.K. I'd like to make some comparisons, even though I totally and fully agree with David's comment earlier on that it's dangerous to make comparisons in terms of a league table. But I think there are some insights we can gain from what other countries are doing compared with how open data has evolved in Canada today.

I'd like to start there, and then I'd like to pick up on a couple of other points that my colleagues have raised already.

What is interesting about the way open data is evolving around the world is that it's evolving in different ways based on the way that government agencies have chosen to engage it.

For example, in the U.K. and the U.S., we see a very top-down approach whereby the U.K. and U.S. governments at the very top levels of government have sponsored open data initiatives. They are driving adoption of open data throughout government departments and agencies, and we see this top-down approach as it flows downwards through the government infrastructure.

I would say that in Canada what we have seen is more of a bottom-up approach to open data. In early days it was adopted primarily by the cities, and then the provinces caught up. I think Vancouver started in April 2009, and we have seen other cities adopt open data initiatives. Then the provinces have come in, and I think the federal government has come in after some of these cities and smaller agencies had already adopted open data initiatives.

That explains why we see different countries and different initiatives at different stages of evolution, to a certain degree.

In the U.K. and U.S., I would say that open data initiatives across government are fairly mature and fairly consistent in the way open data is thought of. I would say that in Canada we see open data being adopted in different ways at different levels of government jurisdiction.

The second point I'd like to make around this is that as we look around the world, it's important to understand that open data itself is not an end point. Open data is a transition to something else. It's an enabler for other things to happen. It's an enabler for such things as economic stimulus, as we have discussed, and I'm sure we'll discuss more on that during the session. It's an enabler particularly for citizen engagement, getting citizens actively involved and participating in the business of government.

I think it also represents a cultural change internally for government and government agencies. When I've been around the world talking to national and provincial and state governments about their open data initiatives and the way we can use open data to engage citizens, particularly those parts of citizenry we may not already be engaged with, a big comment that I get at the end of my engagement with that particular government is: this is great, but now that we have this capability to share data and to collaborate, we want to do it internally as much as we want to do it externally. I think that point was made very well by my colleagues previously.

The opportunity for the Canadian government here is to provide guidance, to provide a framework to take the open data initiatives that already exist, to create opportunities to share more open data, to engage citizens and third parties and encourage them to share this data and use this data, and to enable the sharing of the data in such a way that it can easily be consumed by any of the actors in the ecosystem, be it a data scientist, a researcher, a citizen, an application developer, or a student.

But it's very important that we understand that this is a cultural change that will lead to other positive benefits; this is not just about sharing data itself. And so it's important that the government provide a framework to encourage parties to collaborate around the sharing and reuse of open data—private-public partnerships, for example—and particularly engage those parts of the citizenry with whom perhaps we are not already engaged and get them actively involved in the business of government.

Let me give you a very simple example. Two weeks ago we ran a teen hackathon in the city of Surrey. The City of Surrey is sharing its open data; they have an open data portal. They invited teens, young people from the ages of 13 to 19, to participate in this hackathon. For half a day we worked with them with technology and showed them how to produce applications. What was interesting is that at the end of it we asked for feedback and ideas, and it was amazing to see these teenagers come up with ideas about how to use transit data to better navigate through the city, how to use weather data to better understand when weather might affect particular tourist spots or landmarks.

You could look at that initially and just say that these are interesting ideas but ask whether they would ever come to any kind of fruition. But what was really interesting about the whole thing was that the city was stimulating students and young people to think about engaging the city in ways that had not previously been possible. These were young people who were thinking about actively working with the city—visitors to the city, citizens of the city. Getting them excited and engaged in looking at ways to improve city services both for visitors and for folks who already live in the city is quite transformational. This is a very simple example of transformational cultural change that can be brought about by sharing open data.

Another example I will give you, from a cultural aspect, comes from when I was engaged with the Government of Colombia. I was invited down there to provide some guidance to them about the way they would share data with their citizens. When I went down there I said I was surprised that the Government of Colombia was thinking about sharing open data, because they're not known, to an external person, for their openness or the way they might engage a citizen in a transparent way; that it might be considered to be a threat to the government.

They said that this was their entire reason for doing it. Whereas other governments say they're doing this for economic stimulus or doing it for better engagement with certain parts of society, in Colombia they are doing it deliberately to show that they're being open and transparent. This is part of their cultural change with their citizens.

The last point I would like to make is that I think the opportunity is huge for Canada to be a leader in this area. Even though we look around the world and see open data initiatives evolving in different ways, we have a long way to go with open data, to speak to David's point earlier on. There is much more that can be done and there is much more transformational benefit that can arise out of open data.

But I think the government can help. It can stimulate this by providing, for frameworks for working particularly in public-private partnerships, guidance in the sharing and openness of data, and also by providing ideas and guidance about the sustainability of open data and how it can be part of the ongoing business of government and citizen engagement, rather than just being seen as an end in itself.

Thank you very much.

[The Chair, NDP, Pierre-Luc Dusseault]

Thank you for your presentation.

We are now moving on to the last, but not the least, witness, Ms. Dybenko, from the University of Waterloo, who is appearing by videoconference from Kitchener, Ontario.

[Ginny Dybenko, Executive Director, Stratford Campus, University of Waterloo]

Thanks very much.

There have been great remarks already that have taken a lot of the points that I was going to make.

I would start off by saying that over the past 20 years, we've seen an awful lot of innovation, creativity, and disruption. Today we don't know exactly where open data will lead, but we do know that it will be very transformative. Some ways of doing business will start and some will evolve, and learning how to navigate them will be the challenge that lies ahead for us. But the potential that certainly we saw in the very early days of the web—and I lived through all of that—is what I see now with open data.

I believe open data is data with a mission. It will create jobs; it will fuel start-ups and launch new industries with revenue purportedly in the billions. However, every day untold numbers of people try one more time to figure out Facebook's privacy settings and wonder exactly what Facebook knows about them anyway, and most people have only one concern about their personal data, and that is that they want to keep as much of it as they can as private as possible.

So there is a central paradox here. Releasing personal data as open data can definitely benefit society and ultimately help the individual, but if the data is not controlled carefully, having it out in the open will damage individual privacy and may outweigh the benefit and slow the process down.

I was introduced as the executive director of University of Waterloo's new Stratford Campus focusing on digital media. Additionally, I have had an extensive background in technology in corporate Canada with 30 years at Bell Canada in IT and digital communications and I am currently a member of the board of governors for SSHRC, the Social Sciences and Humanities Research Council of Canada, so you can understand that I have a number of viewpoints on the open data opportunity.

I'd like to begin with the viewpoint of our wonderful young digital natives enrolled in our undergraduate and graduate programs at the University of Waterloo. Digital natives are young people who were born in 1997—ouch—who have literally never not had a device in their hands. And as Mark mentioned, we have engaged them on a number of occasions with the municipality.

We recently ran a project for Stratford on garbage, of all things. It was a hackathon run over a weekend. Essentially, when do you put your garbage out and how do we communicate with our citizens? We didn't think that our young students would be particularly interested in this. They dove in and produced some remarkable methods of connectivity that the city is now looking to continue to develop.

But that pales in comparison with Code 2014, run by Tony Clement. That was a hackathon run across Canada about a month ago engaging 900 young people that challenged them to develop apps around the open data that the Canadian government already has laid out. I am delighted to tell you that of the 900 applicants or participants, a team from our school actually won. Their application essentially delved into StatsCan employment and social development, Canada data, the Canada Revenue Agency, and the CMHC in helping immigrants choose the right place to come when coming into Canada. It was featured on the CBC this morning.

There are a number of other kinds of applications that were developed as well. Fifteen were finalized.

But, Mark, I loved what you said. From my point of view, what we're really doing here is engaging young people in the affairs of the government. That has been a huge challenge, I think, certainly at municipal and provincial government levels.

With regard to the expectations of these digital natives, as a consumer, they definitely want the personalization of their experience which comes from open data, but they also want to ensure that their data is very private, or that they have control over that. As an entrepreneur, they want ready access to the data, but they also want assurances of ownership once they have developed their idea.

At a corporate level, I think a storm is brewing. Corporations want access to unattributed personal data to examine trends by demographic group, for example, but they want attributed data to do specialized or specific targeted marketing, which is scaring a lot of people. The opportunity examples that McKinsey has pointed to look to billions, $300 billion annually in health care, in the U.S. alone. They go throughout the world and the opportunities are limitless, as David referred to earlier.

Finally, I do believe that the Canadian government is at the leading edge of governments around the world. Certainly we are seen as leaders, and we are regularly referenced, particularly in U.S. documents on the topic. I think we are well positioned for significant savings. The granting agencies, the tri-council, conducted extensive consultations around big data's role in the development of digital scholarship in Canada, which was conducted in the fall of last year, coining the term “open research”.

The conclusions were threefold: first, that there is a culture of stewardship that asks for an establishment of clear policy for data sharing; second, that there is a coordination of stakeholder engagement, in other words, long-term planning—and remember that this isn't about data on colliding particles but mostly data on people—and therefore the involvement of SSHRC is very important; and third, they raised as an issue the developing capacity, so that engages funding and roles and responsibility among national, provincial, and institutional stakeholders.

In conclusion, I'd just like to say that I believe open data is our next natural resource. Canada has the digital infrastructure. We have the reputation for collaborative management. We have the respect of many in the world in this arena, and we have a hugely developing knowledge worker population, through programs such as ours in Stratford. Canada should make open data a priority, establishing policies, engaging in long-term planning, and developing capacity.

Thank you.

[The Chair, NDP, Pierre-Luc Dusseault]

I want to thank all four of you for your presentations.

As planned, we will now begin the question and answer period between the committee members and the witnesses. I would like the committee members to specify which witness they are putting their questions to. That will make things easier, since two of our witnesses are appearing by videoconference.

Mr. Blanchette, go ahead. You have five minutes.

[Denis Blanchette, NDP, Louis-Hébert, QC]

Thank you, Mr. Chair.

Thank you very much to our guests.

The issue we are discussing this morning is very thought-provoking. Everyone approaches it in their own way, from a unique perspective. There may not have been enough time for this, but I would have liked us to put open data in its true context—in other words, the transformation of our society to a digital society. Whether we like it or not, the Internet now plays a concrete role. It's practically a right nowadays. People have access to the Internet almost as they do to running water.

The use of open data is part of that context. Some elements that were mentioned are already becoming a reality. Websites such as Facebook and Amazon are already partially personalizing users' and consumers' preferences.

Mr. Gayler, I thought your approach was quite noteworthy. You talked a lot about the approaches and the context. According to my understanding of your presentation, the Americans started with the federal government, while our approach was a bit more heterogeneous.

Have you looked into what approaches have been used outside North America? What other trends are out there? Do you have an idea of what is going on elsewhere? As representatives of the Canadian government, we would like to have a good idea of what is happening on the international stage.

[Mark Gayler]

First of all, I can only obviously comment on the jurisdictions where I have personal experience.

I would say a couple of things. If I back up a little bit and re-clarify what I said earlier on, I think where we see a top-down approach such as in the U.K. and the U.S.... And what we mean by a top-down approach in open data terms is where guidance is given by the national government in ways that data can be shared, how departments can share that data. They provide guidance and frameworks to enable that to happen as part of the business of government. That's what we mean by that top-down approach.

What we tend to see is that governmental departments, then, become more encouraged to share data because they have been given a mandate by the national government, if you like, and it becomes more baked into the process of government rather than being seen as, “Well, we do government, and oh, we also do open data.”

I think there's some good learning there for Canada, certainly.

Canada is in the position where it can certainly exploit some of this learning that we see in the U.K. and the U.S., but that's not exclusive to other countries. If we look at Germanic countries—for example, Austria, Germany, Switzerland—again what we see there is that this is very much city-based. The national governments are looking at open data initiatives, they are looking at open data policy, but by and large, to date, the way that the citizens have engaged on open data is through city and provincial open data initiatives.

I would say the same for Italy, for example. If we look at some work that's been done with the Italian Ministry of Health to share data, if we look at the initiatives that are going on throughout Italy, they are largely city- and provincial-based. And there is a reason for this.

If you think about data—the value of data and its relevance to citizens—national data, of course, is interesting; statistical information, of course, is interesting. It's particularly interesting to data researchers and data scientists.

However, if you look at the average citizen, they're interested to know when their garbage is available, what the health situation is in their local school area, for example. Local data has a lot more relevance to the average citizen in many cases than, say, national trending data. That's why we see these initiatives evolving in different ways and citizens engaging and taking up that data in different ways.

[The Chair, NDP, Pierre-Luc Dusseault]

Thank you.

[Mark Gayler]

I hope that was....

[The Chair, NDP, Pierre-Luc Dusseault]

Thank you, Mr. Gayler. Mr. Blanchette's time is up.

We now go to Mr. Trottier, who has five minutes.

[Bernard Trottier, Conservative, Etobicoke—Lakeshore, ON]

Thank you, Mr. Chair. I'm very impressed with our panel of witnesses here this morning.

We started this initiative a few weeks ago, and it was recognizing that, as a federal government, we've signed a G-8 open data charter. There's a mandate now to develop a road map for the federal government, and I appreciate the different perspectives.

I'd like to get some input from all four of our witnesses this morning on this notion of the government as a publisher of data, very much a one-way flow of information from government to citizens versus the notion of the federal government being more of a facilitator or creating the public square where people can publish certain data, a way to engage them. Many examples come to mind where the government can't create the data. If you think of species at risk, for example, where there are eyes and ears all over the country, and people might be able to spot a rare bird and they can provide that information.

One of the challenges with providing that public square is how do you confirm whether the data is good or do you need to confirm? Some people could be there, not so much to publish data, but they have a certain point of view they try to advocate and they could hijack that public square.

Can each of you in turn, in the order that everybody spoke, talk about that 1.0 version of open data versus a 2.0 of more of an engaged version of public data?

[David Eaves]

Absolutely. I agree there's opportunity there, but for the reasons you mentioned, I'd be fairly conservative about how I would try to engage in that opportunity. One of the things that the government has done, and I think has been quite effective, has been a canonical source for data that is highly trusted. Statistics Canada creates data that is highly trusted by people in the non-profit, for profit, and government space. Having something whereby people can all point to a dataset and say they believe that and they use that as the foundation for their conversation is enormously useful and cannot be underestimated.

To talk about how you crowdsource the creation of data creates an enormous number of methodological problems that I would be wary of rushing into, especially when we have so much data that is canonical and is verifiable that we already are not sharing and I would argue are not leveraging as effectively as we could. I'd much rather solve that initial problem first before thinking too much about that second problem.

[Bernard Trottier, Conservative, Etobicoke—Lakeshore, ON]

Ms. Miller.

[Dr. Renée Miller]

Let me give you an example of where this has worked. In the U.S. They have a portal, I forget the exact name, but it's something like peer to patent database, where they have opened up the patent process to input from experts, recognizing that the expert on a particular patent topic is often not in the government itself. They invite scientists and inventors to comment on patents that are under review. They have found that they get much higher quality information from that process than they could just adjudicating the patents themselves using their own experts. They didn't abdicate the responsibility for making the final decision, for having somebody making sure there wasn't somebody with a bias inputting data, but they were able to get much richer information on which to make their decisions.

I think you can walk that line. I think you do have to be careful with it, and still have somebody adjudicating the information itself.

[Bernard Trottier, Conservative, Etobicoke—Lakeshore, ON]

Mr. Gayler and Ms. Dybenko, could you comment briefly on that? One of our objectives of this study is to give our own government, the Treasury Board specifically, that direction on how to create that road map for the federal government.

[Mark Gayler]

I think one thing that's very important to understand with this topic is that all data is inaccurate to a certain level, so you can't wait until the data is 100% accurate to share it. That's something that some government departments feel very concerned about. Again I agree with David's point. I think the emphasis here is on sharing data that's not being shared today and setting an expectation for the integrity and accuracy of that data as it goes out either to public or to commercial entities for that matter. I think as long as you're clear about what that might be, that would be where I would place the emphasis first.

[Bernard Trottier, Conservative, Etobicoke—Lakeshore, ON]

Okay, thank you.