Government Operations Committee on May 8th, 2014

[David Hume]

We are almost always in consultation. One of the strategies we have always used is to be wide open with our users. We're constantly engaged with them through events as well as through social media and other tools.

I think the best thing I can say about the response we have received is that we have served over 300,000 users since we launched three years ago. That number continues to grow steadily. Whenever we are out explaining the opportunity to people, there is always a lot of interest. Especially when you're working with people who are real data users, academics, teachers, students, business people, the response is always very positive.

One of the things about open data is there are often challenging issues that are profiled through the data, but it's also important to understand we're not sharing personal information, so a lot of the challenges that people have with data on that front are not an issue for us on open data at this time.

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

Thank you.

Ms. Nadeau, you seem skeptical about open data. You talked about “open by default”, which is a caveat for this concept. You also talked about international taxonomies, meaning international classification.

Could you tell us what the ideal situation would be in this case?

[Diane Nadeau]

Yes. Right now, all governments, whether at the federal, provincial or municipal levels, are adopting various taxonomy systems to categorize the services they offer. The categorization of data is completely different from one website to another. As a result, when people consult a site, they have to kind of figure out how the data and services are categorized. The same is true internally, within governments. Everyone has their own way of categorizing data.

In New Brunswick, we started setting up a proper classification system. We adopted the Australian system because the data were more open there. In fact, all the information was on their website. We have made some changes since we are a province, not a country.

I agree with the idea of having municipal taxonomies across the country and I think we should apply this principle in the provinces and across the country so that we are all aligned.

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

Which is not the case right now.

[The Chair, NDP, Pierre-Luc Dusseault]

Thank you, Mrs. Day. Your time is up.

Mr. O'Connor, you have five minutes.

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

Good morning.

I just want to talk about the different levels of government. There are the federal, the provincial, and the city governments and each of them have their responsibilities. I'm wondering if there are any jurisdictional issues to be resolved or any impediments in information? For example the federal government is also involved in environment, health, resources, and other things, as the provinces are. Is there any possibility that one is stepping over the bounds of the other?

I'll start with Ms. Nadeau.

[Diane Nadeau]

Yes there could be some instances where we perhaps do some duplication of work. And I think there are a lot more opportunities to share information between the three levels of government, especially in geographic data.

Like I mentioned the Government of New Brunswick shares data with Natural Resources Canada. Geographic data can be used anywhere with just about anything.

I remember in 1998, I did some work with Elections Canada in trying to share information between all of the election offices across the country for the electoral districts. At that time, I don't think it was possible and I think today, it is much more possible.

So yes there are a lot of opportunities for sharing.

[Gordon O'Connor]

Mr. Hume.

[David Hume]

Yes I would echo Ms. Nadeau's comments.

I actually don't think the issue is us stepping on each other's toes, it's more an issue of us not sharing enough. The opportunity to actively share data and in an open way means that we can be consuming each other's data much more seamlessly and repurposing those things for the things we need within our own respective context.

The opportunity for us to be able to take federal data, bring it together with provincial data and municipal data to run different kinds of analyses, and create policy options or service improvements as a result of that kind of measurement is really important. And it gets so much easier when the process doesn't involve the creation of a data sharing license agreement or having to negotiate terms between different levels of government and create that kind of red tape around just the sheer access to data that exists in a lot of circumstances right now.

This is why the point about open data by default is an important one. It's because if we assume and we define and we design our data in a way that we anticipate it will be shared, it means it will be consumable by everyone and it will essentially just reduce red tape across the federation.

[Gordon O'Connor]

Okay.

Ms. Nadeau, my understanding is that New Brunswick does not have an open data site, per se. Are you planning to do such a thing in the future?

[Diane Nadeau]

Yes, I believe we are. Right now I think the only one that is there is GeoNB. That is listed on the site. We actually have a lot of open data. It's just not organized into an open data site.

[Gordon O'Connor]

Are all the various departments within British Columbia feeding information into your data site? Or are there some that are not?

[David Hume]

Well, some are doing more than others, but yes, all of them are. Probably the largest contributor in volume is the Ministry of Health. I think the smallest contributor is the Ministry of Aboriginal Relations and Reconciliation. It just has to do with how much data they're holding and what they are able to make open.

[The Chair, NDP, Pierre-Luc Dusseault]

Thank you, Mr. O'Connor.

Mr. Byrne, you have five minutes.

[Gerry Byrne, Liberal, Humber—St. Barbe—Baie Verte, NL]

Madame Nadeau, New Brunswick is Canada's only officially bilingual province, and presumably you'll be developing your open data sites in both official languages. Obviously the data is numeric, but the descriptors to the data and the capacity to be able to search would depend on which particular language you use.

My question is framed from this point of view. If there were inter-jurisdictional sharing of data, would you perceive there being any problems in the capacity to do searches? If a Canadian citizen were to attempt a search, using both federal and provincial data, and there wasn't a bilingual capacity to be able to do so, would that be any inhibition to being able to have a successful search?

[Diane Nadeau]

I think there were two questions, so I will answer the first one. Yes, we are bilingual, and the information that would be put out would satisfy the English and French requirements. When we publish information, we have to put it out in both languages.

Second, I would refer you to the search engine Google. If I search even in English for French information, it usually brings back some results. It doesn't just say, “Sorry, I don't speak that language.” So, I think it is possible and we have to work towards that goal.

[Gerry Byrne, Liberal, Humber—St. Barbe—Baie Verte, NL]

Okay, Mr. Hume, to follow up on that, one of the presenters we've had at committee did suggest that, if the Canadian government and the provincial and territorial governments were to attempt to harmonize, wherever possible, in an open portal, and to develop a single search engine to be able to do this, it would be very valuable. Could you see any constraints if we were to.... At the ground level now, which is what open data is—we're really at ground zero to a certain degree—are we losing capacity and opportunity if we're not moving towards a goal of harmonization, the capacity for jurisdictions to share and to provide data in a mutually acceptable and agreeable format?

Put it in context, as well, with the official languages aspect, if you could.

[David Hume]

I think, through the work of the open data and open information working group, which I referenced in my remarks, we're trying to create opportunity. We're beginning some work that will be looking specifically at this question of search, and what we would call it is “federated search”. The idea would be that, if you did a search at data.gov.bc.ca, you'd be able to find data referencing any jurisdiction that's in the province of British Columbia, as well as any data that is relevant from the federal government; and that capability would be across the country for data portals. That would be the ideal state.

Regarding the question about English and French, I think Ms. Nadeau's comments about the way Google works are right on. It's mainly just a question of how you organize the information. If you're smart about how you tag, organize, and label that information, the question on bilingualism will be less of a constraint.

There are challenges that come when you need to work to standardization in ensuring translation is effective, well-recognized, and standard. But most international standards, which is what you want to be looking towards, are able to accommodate for that, because they're international. They're in multiple languages, not just English and French, so they're intelligible globally.

[Gerry Byrne, Liberal, Humber—St. Barbe—Baie Verte, NL]

Thank you.

I am taking from what you said that we need to be smart in doing that. So are we being smart in doing that?

[David Hume]

Well, we're attempting to be as smart as we can.

[Gerry Byrne, Liberal, Humber—St. Barbe—Baie Verte, NL]

Thanks, Mr. Chair.

[The Chair, NDP, Pierre-Luc Dusseault]

Thank you, Mr. Byrne.

Now to Ms. Ablonczy for five minutes.

[Diane Ablonczy, Conservative, Calgary Nose Hill, AB]

Thank you, Mr. Chairman.

I thank the two of you for being here.

I'll start with Mr. Hume. You impress me because you addressed the questions very logically and that was very helpful.

I want to follow up on Mr. Byrne's question. We've been told that the RDF format is the one that should be used for open data posting. Are you using that or do you intend to use it?

[David Hume]

We have some RDF data, not a lot of it. It's a new-ish format that a lot of our.... I think the thing to remember about where we're at is that a lot of the systems where we will be pulling data to publish aren't designed to create data in the RDF way. RDF is a really interesting and powerful data format because it can create those interconnections among different datasets, but we haven't done a lot of that so far, no.

[Diane Ablonczy, Conservative, Calgary Nose Hill, AB]

Ms. Nadeau, have you grappled with that question at all?