Good morning, everyone, and thank you for having me.
If I could just ask you to talk to Environment Canada to change the weather out there, I'd appreciate that. That's my first ask.
I'd like to break down my comments into two main sections. First is a little bit about the history of open data and open government at the City of Toronto and some lessons learned that might help you. Second, I'd like to address the key questions that you've asked of me in terms of our response and input into the federal data portal.
Just briefly, my name is Harvey Low. I'm the research manager of the social research unit at the City of Toronto. In 2012, I was the OGP representative down in Brazil on behalf of Canadian civil society and Canadian municipalities. I am currently the chair of the city's research committee, and the human services cluster representative on our open government committee. The city has an open government committee, actually.
We have learned some lessons at the city. We've been in the open data game since about 2010. As Robert mentioned, I'm one of the original four. We have an open data site, and now the city has actually moved to multiple portals for access to free data. Essentially we have not just the open data site, which has about 110 data sets up there, if my count is correct, but also an application called Wellbeing Toronto. It's an online mapping application that provides an additional 200 to 300 variables, many of which are from federal and provincial data sets that have been long in coming. So I agree with Robert that if there is a way the federal government can help us stickhandle and manage access to both provincial and federal data, that would be wonderful.
I am also on the steering committee of the Canadian Council on Social Development's community data program. For those of you who are unaware of that program, that was formed about eight years ago, with 22 Canadian municipalities, to get easier and low-cost access to federal data. That's been going on, as I said, for eight to ten years.
One thing we've learned at the city is that it's not always wise to simply release data sets without context. I'll speak to that in a little bit more detail.
We want to make sure that a website that is open to the public also has a context explaining what all those data sets are about. Typically users out there are of a common profile for open data sites. They are usually more technically savvy, younger in age, and very competent in the area of social media.
The first question you asked us was how we compare to other municipalities and how we meet the needs of Canadians. You asked about how we become accountable and transparent while maintaining privacy. I think another question you asked was how we use the data for very specific issues such as job stimulation.
I'll try to address those now through a couple of key theme areas. The first one is to engage the unengaged. We need to decrease the digital divide. As I mentioned before, typically the users out there are people who are hackers and developers—and that is wonderful—and they're very good at data manipulation. Ron mentioned that a lot of them know how to manoeuvre around data. A lot of them know how to analyze data, and they know how to change formats in data. The common public don't have those skill sets.
I think, in line with the provincial and local municipalities, the federal government also needs to begin to recognize that there are other users of your data out there. You need to engage the unengaged.
Who are those? Well, outside of the developers and hackers, there are cities and municipalities. We need data for very strong, place-based evidence planning. Everything that we do in municipalities is done by neighbourhoods, by place. I'll speak to the issue of geography in a moment. You need to reach out to non-government agencies, community agencies, and those that also use data and analyze data on behalf of the public and the clients they serve. The other area of engagement is of non-traditional users of open data sites: professional organizations. They are sitting on a landmine of information from professional geographers and urban planners. Those associations would recognize the value of consolidated federal data. Of course the last group is the people we serve. These are the at-risk population groups—seniors, new immigrants, youth. Those are the groups that also need the data.
So you need to begin thinking about it not being just about hackathons. While I think hackathons are great, you asked whether we are serving the needs of Canadians. What we've heard in the city is that they need an intermediary. They need an intermediary to take that data, such as that from municipalities or the province, and analyze it on their behalf.
The second thing I'd like to talk about is releasing the relevant. What do I mean by that? You asked how we compare to other jurisdictions. The real question is how we can be better than other jurisdictions.
I did a quick scan of the international federal data sites from the Netherlands to Australia. There is one thing in common with us here in Canada. I went in there pretending to be a common user from the public, and I wanted to type in an issue area. We talked about job stimulus. I typed in “unemployment” and I came up with—and I hope I'm not putting out anyone from Australia—zero on the Australian site, zero on the Netherlands' site, zero on the U.K. site. People search by issue areas. The search engines that we have out there...and I'm glad, Ron, that you mentioned we need a national search engine, because that's exactly what we need. We need a better way to tag data. Right now, you need to know what that data set is called before you search on it, which is a bit of a problem. The developers will figure it out, but the general public won't.
We operate right now in a local municipality in what we call a shared-service delivery model. We as municipalities work with communities, non-government agencies, in interpreting data on behalf of the public. Why do we need the data? We need data on vital socio-economic data right now about people's economic circumstances, education levels, those types of things. Those types of data sets, while they may be available at federal ministries like CIC and StatsCan, are not connected right now with your data portal. Raw data's in one place, and that's fine. That serves a specific audience, but we need to pay attention to that other audience.
The other thing we need to do is to link and leverage. You asked the question about how data can stimulate job growth, as an example. One way is to begin linking your data portal with the service provider areas in the various ministries across the federal government, and down through the provinces and municipalities. So if somebody goes in there looking for the economic situation of their municipality, their neighbourhood, or Canada, it would be good to have the data in your data portal linked with, say, economic development or the ministries, even if it's a link to their website. Then you're connecting and networking that raw data to a very specific mandate of a particular ministry. If I understand your standing committee, one of your mandates is cross-departmental connection of services for the public. If you just provide data on a raw data site, that will satisfy the hackers and developers, but it may not truly get at providing the services to the public you want as a federal government. That's what I think we're all here for, for all three levels of government.
The other thing we want to do is to make sure we tag the data by thematic areas. To get back to my question and issue about not having sufficient tags when I type in something as simple as “seniors” and “youth”, I'll get summary papers, which is great. We need to provide better tagging and search capabilities on all our websites.
We need to begin thinking about geography. We need to give the gift of geography. We do everything at municipal levels by place. It's wonderful to have statistics at the city level, what StatsCan calls the CSD level, but that level of data we need to be more granular. We need it down at the neighbourhood level. There should be search engines out there that allow users to search by geography, as well as theme. When we look at the data we need, we're going to need unemployment rates, we're going to need health care statistics. Those have to be at a level of granularity that makes it useful for cities in their place-based evidence planning.
The question on privacy can also be solved through geography. We do not have to look at individual data, we aggregate data to a level of geography that allows us to do that aggregation. Right now, you could say we have the Stats Canada site, but as I said before, begin to link the data portal at the federal site with other ministries and their service mandates so when people look at a table they know where to go if they have questions about that data, and if there are interconnections.
One of the ideas is that you look at service delivery from a multiple factors analysis. We call it in social planning, social determinants of health model. That means if a person has an issue with unemployment, and it has to do with health, housing, transportation, all of these things need to be linked. I think there's an opportunity to link your open data site with service providers in the federal ministries, down through the provinces and the City of Toronto.
Robert mentioned that—not the horizontal connection, which is one issue, but the vertical connections between all levels of government. This is because people understand geography. They don't understand jurisdictions. They don't care where the service comes from. They don't care where the data comes from.
Where do we go from here, as a conclusion? We need to be thinking about intergovernmental data sets, not just data out in Victoria or Halifax, or one ministry over in Alberta or a federal ministry. We need to begin thinking about what the public wants and creating data sets that represent intergovernmental service delivery to particular groups.
Thank you for the time, and I look forward to your questions.