Talking open data in Edmonton with Minister Tony Clement

Tony Clement, President of the Treasury Board of Canada, was in Edmonton today as part of a cross-country tour to gather feedback from the Open Data community. The federal government is preparing to launch a revamped Open Data Portal, and Minister Clement has been given the mandate to make it happen. After stops in the morning at Startup Edmonton and TEC Edmonton, Minister Clement was at City Hall for an Open Data Roundtable, hosted by the one and only David Eaves.

"Open Data is a global movement that is really gaining momentum across the country. Our Government wants to ensure we are making it as accessible as possible so that innovators and enthusiasts can harness this rich resource," said Minister Clement. "We are getting ready to unveil the next generation Open Data Portal and the input we received from Edmonton’s vibrant Open Data community will help us build a user-friendly site that will allow users to capitalize on this opportunity."

In addition to holding face-to-face meetings, Minister Clement also hosted a Google Hangout on Open Data last month. You can watch the whole thing here:

I was fortunate enough to be invited to the roundtable today, along with roughly two-dozen other Edmontonians interested in open data. We had a very limited amount of time to chat, but I think we still discussed a wide range of topics. I hope the information gathered was indeed valuable for the team in charge of the new portal.

As host, David organized our time around a series of questions. The first was to suggest ideas for what the next generation open data portal should be. We broke into small groups and then shared ideas back with the larger group.

The first thing I suggested was that it should not look like it was designed in 1995. I find all of the Government of Canada websites lacking in the aesthetics department! Certainly there’s something to be said for consistency, and I understand there’s an initiative underway to reimagine the entire GoC web presence. On the flip side of consistency though are the preconceptions that you may not want to be carried forward. If I look at the Open Data Portal today, it looks like every other government site, which makes me think it’ll be a mess of weird hierarchies and PDFs buried away. It’s not very welcoming or inviting!

Another theme was based around the idea that we can’t build a data portal that serves all possible audiences. But, we can do more than we are currently. So my group discussed the idea of intent-based profiles. The idea is you’d login, set some criteria like whether you’re a developer or not, and maybe your location, and the portal would then give you a personalized view. Of course, anonymous access should be preserved, so it shouldn’t be a requirement that you need to login.

Three other themes that emerged included: historical data and the realization that any data we create now will at some point become historical, articulated well by Heather and Maureen; the notion that the portal should facilitate the two-way movement of data, so that citizens can publish data into the catalogue as well as get data out of it; and the fact that documentation about data is important, and it doesn’t necessarily have to be created top-down.

The second question was related to datasets, both specific datasets that we’d like, but also criteria that make datasets valuable and/or interesting. The first thing that came to mind for me was geography. I’d love to be able to see all of the datasets related to Edmonton, or to municipalities, or to Alberta, or to provinces. Right now you have to really hunt to find datasets that compare cities, for instance.

I think Matt‘s two suggestions in response this question were spot on. The first was that the data he finds interesting is the data that makes the government uncomfortable. Minister Clement jumped in to assure us that there’s no conspiracy preventing certain datasets from being released. That would suggest a level of organization that most governments just don’t have, he joked. The second suggestion was that geographical data should be a key foundational dataset. Let’s see a base map of the country, zoomable to the neighbourhood level. Or to whatever smaller regions exist, whether it’s postal code, census district, garbage collection zone, or something else. I love this idea, and my only add-on suggestion was that geographical data doesn’t necessarily have to mean maps. Knowing the list of neighbourhoods or postal codes can be incredibly valuable outside of a map as well.

A few other themes that emerged about datasets were trends and historical data (I personally love the idea of a revision history for any datasets), some sort of metadata (the first dataset any portal should have is the list of datasets it contains, David suggested), and the notion of a data management plan.

We finished up the roundtable with a brief discussion on data standards, followed by a few minutes of open time. I spoke up on data standards along with Ben and Eugene, and suggested that data standards fall squarely under the "nice to have" category. It would be great if different datasets shared a common format, but we’d rather just have the data and worry about the differences with an abstraction layer.

Final thoughts mentioned by the group included dogfooding (the government should actively use its own data portal and datasets), the idea that everyone carries a phone and could be contributing data back into the catalogue, and the future world described by Devin that I think can be summarized as the Internet of Things. "The idea that we searched a catalogue of datasets will seem just as quaint as when we searched the web using Lycos," he said.

I really enjoyed the roundtable today, and I appreciate Minister Clement and David Eaves taking the time to listen to what we had to say. Thanks also to Chris and Ashley from the City for providing the venue and helping to facilitate.

Reflecting on it now, I think what I enjoyed most about the roundtable was the opportunity to chat with people in the local open data community. I haven’t given open data as much public attention lately as I should, and when you shift your gaze elsewhere it’s easy to miss all of the incredible people doing great things.

2 thoughts on “Talking open data in Edmonton with Minister Tony Clement

  1. Since I was not there, I can’t make any direct comments related to the meeting with Minister Clements but will add the following:

    – definitely the current Canadian government Open Data portal needs to be more intuitive to use. To paraphrase others “it is easy to make something hard but hard to make something easy”. The portal does not have to be all things to all people – the swiss army knife approach needs to be avoided. Of related interest, the UK government has made a big effort to streamline their government website. Two links which may be of of interest are:

    https://www.gov.uk/designprinciples

    and

    http://digital.cabinetoffice.gov.uk/2011/09/19/introducing-the-needotron-working-out-the-shape-of-the-product/

    – more emphasize needs to go into the contents of the portal – the catalogue, so to speak. One should be able to access the catalogue as a dataset or via an api (both really). An implication of this is – if users can access the catalogue of datasets directly then how elaborate does the government portal really need to be?

    – how the contents of the catalogue are organized (knowledge organization) needs to be available and clear.

    – if the government wants to add functionality to the catalogue for data visualization, is that something the government should do or should the government create incentives for the not-for-profit and for-profit ecosystems.

    – while there is a place for PDF’s (some people do want them and they are better than nothing) it is no substitute for actual data that can be consumed programmatically.

    – the concept of having profiles to suggest datasets – I suspect this is more likely to miss something than find something.

    – historical data is important – previous versions of data should be preserved.

    – citizen publishing of data – crowdsourcing of data – there is merit to this idea but I am not sure if this should be part of a government open data catalog.

    … gerry tychon

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s