Last week a group of volunteer programs from across Canada announced the launch of Represent – a website that tries to map all of Canada’s boundaries. Confused? Don’t be. It’s simple. This is a nifty piece of digital infrastructure – try visiting the website yourself! After identifying where you are located it will then tell you which MP riding, MLA/MPP district and census subdivision you are located in.

So why does this matter?

What’s important about a site like Represent (much like its cousin site Mapit, which offers a similar service in the UK) is that other websites and applications can use it to offer important services, like letting a user know who their MP is, and thus who their complaint email should be sent to, or identify what by-laws are applicable in the place where they are standing. Have you ever visited the site of a radical group non-profit which urged you to write your MP? With Represent that organization can now easily and cheaply create a widget that would figure out where you are, who you MP is, and ensure you had the right address or email address for your letter. This significantly lowers the barrier to advocacy and political mobilization.

This is why I consider sites like Represent to be core digital infrastructure for a 21st century democracy. Critical because the number of useful services that can educate and engage citizens on politics and government is virtually limitless.

But if we accept that Represent is critical, the site’s limits tell us a lot about the state of our democratic institutions in general, and our open data policy infrastructure in particular. In this regard, there are three insights that come to mind.

1) The information limits of Represent

While Represent can locate any of the federal and provincial ridings (along with the elected official in them) there are remarkably few cities for which the service works. Calgary, Charlottetown, Edmonton, Mississauga, Montreal, Ottawa, Stratford, Summerside, Toronto and Windsor are all that are identified. (The absence of Vancouver – my home town – is less alarming as the city does not have wards or boroughs, we elect 10 councillors in an at large system). The main reason you won’t find more cities available is simply because many cities choose not to share their ward boundary data with the public. And of course, things don’t need to stop with just city wards, there is no reason what Represent couldn’t also tell you which school district you are in, or even which specific school catchment area you are in, in say Vancouver, or North Vancouver.

The paucity of data is an indication of how hard it is to get data from most cities and provinces about the communities in which we live in. There has been great success in getting open data portals launched in several cities – and we should celebrate the successes we’ve had – but the reality is, only a tiny fraction of Canadian cities share data about themselves. In the overwhelming majority, useful data about electoral boundaries, elected officials, schools, etc… exists and are sued internally by governments (paid for by our tax dollars) but they are never shared publicly and so cannot help drive democratic engagement.

So here’s a new rule. If your city boundary data isn’t in Represnt – your city is screwing up. It’s a pretty simple metric.

Oh, and Canada Post, you’re the biggest offender of them all. Your data is the default location specific data set in the country – the easiest way to locate where someone is. Being able to map all this data to postal codes is maybe the most important piece of the puzzle, but sadly, Canada Post clings to data our tax dollars subsidize the creation and maintenance of. Of course, in the UK, they made Postal Code data completely open.

2) Lack of Standards

And of course, even when the data does exist, it isn’t standardized. Previously non-profits, think tanks and even companies would have to manage data in various forms from innumerable sources, (or pay people lots of money to organize the data for them). It shouldn’t be this way. While it is great the Represent helps standardize the data, standard data schemas should already exist for things like MPP/MLA/MNA ridings and descriptions. Instead we have to rely on a group of volunteer hackers to solve a problem the countries leading governments are unable, or unwilling to address.

3) Licenses & Legality

However, the real place where Represent shows the short comings in Canada’s open data infrastructure is the way the site struggles to deal with the variety of licenses under which it is allowed to use data from various sources.

The simple fact is, in Canada, most “open data” is in fact not open. Rather that have serious restrictions placed upon them that limit the ability of sites like Rperesent.ca to be useful.

For example, many, many cities still have “share alike” clauses in their licenses, clauses that mean any product created using their data may not have  “further restrictions of any kind.” But of course, each city with a “share alike” clause has slightly different restrictions in their license meaning that none of them can be combined. In the end it means that data from Vancouver cannot be used with data from Edmonton or from Montreal. It’s a complete mess.

Other jurisdictions have no license on their data. For example electoral boundary data for British Columbia, Prince Edward Island, Newfoundland and Nova Scotia is unlicensed, leaving users very unclear about their rights. Hint to these and other jurisdictions: just make it open.

What Represent really demonstrates is that there is a need for a single, standard open data license across Canada. It’s something I’m working on. More to report soon I hope.

Despite these hurdles, Represent is a fantastic project and site – and they are looking for others to help them gather more data. If you want to support them (and I strongly encourage you to do so) check out the bottom of their home page. Big congratulations to everyone involved.