Web APIs explained on CBC Spark - and Open Data Under Construction

(I almost wrote Spark CBC, since that's their Twitter name.)

Spark Episode 76 (audio link available directly in the post, as well as various podcast options)

At 22:02 or so in, they take on the challenge of explaining web APIs, or more specificially, they ask Jer Thorp to help walk them through the concept.  It's always interesting to hear the descriptions people use.  For example, I would generally say "machine-to-machine", which is probably way too abstract.  I also tend(ed) to describe APIs in the context of Service-Oriented Architecture, which probably confused the issue (and the audience).  I don't generally talk about computer programs communicating with other computer programs.

I think in general what's presented on the show is a pretty good explanation: websites are opening up their information using APIs, so they can leverage open innovation - outside developers.  We are a long long way from a completely interoperable web of standard APIs though.

Here's the Twitter-sized explanation I had proposed (taking quite a lot of my space to talk about how there wasn't enough space):

scilib: @sparkcbc I don't think that will fit in a tweet. Basically standard interfaces (APIs) allow data to flow between sites, which = mashups.

I would argue as well that web development has gotten sophisticated enough that, while APIs are ideal (at least if well constructed), you can actually get a lot by opening your data, which is the key first step.  Open data enables mashups, APIs just make mashups easier.  Open data means sharing the information your organisation has, out on the web - ideally your default becomes to share.

We're still in early days of open data.  The Guardian calls their approach "Data Store - Facts You Can Use".  I've written previously about the US Data.gov initiative, which currently has the world's simplest website (a giant box reading "coming soon"), but I think is supposed to launch this month.  It's similarly challenging to point to open data cities, because while the Twitter-enabled Toronto @MayorMiller announced toronto.ca/open at Mesh, it also reads simply "Under Construction".

What will be possible is mashups, visualizations, APIs, analysis and much more.

I believe the long term success of projects like StimulusWatch Canada and ChangeCamp Ottawa will depend on open data, and (eventually) on all levels of government having open APIs as well.

Which circles me around to the opening topic of the podcast, about whether online activism ("slacktivism") can actually translate into meaningful real-world activity.  The answer, I think, is tied in with the segment about lurking... the web is mostly lurk, only maybe 10% participate.  Some tiny fraction of those online participants might translate into offline actions.  Maybe one in a thousand?  But nevertheless, it does happen.

While I generally refuse to join these "click your support" Facebook groups (in part because I don't like FB much anyway), they can be low barrier entry points, in particular since so many Canadians (who may otherwise not be very social-web enabled) are in FB.

The kind of canonical Canadian example is the Fair Copyright for Canada group, with its (at time of posting) 90,071 members.  It was brought up in the House of Commons.  It did translate into some offline activism.  And the sheer numbers did, I think, get both attention and generate concern for the party proposing the bill.  There are still lots of issues with that number.  Lots of people around the world care about copyright.  For all I know, that's 81,000 copyright-concerned Americans, and 9000 Canadians.  Such is the global web.

I do think "feel-good clicks" are a bit dangerous, they give you the perception of action without actually doing anything.  I've long been concerned by this kinda of almost mystical power ascribed to online organising.  In my review of Al Gore's The Assault on Reason, I said

Don't get me wrong, I think the Internet has a role to play in reasoned discourse. A small role. A useful tool for pointing attention to falsehoods and referencing inconvenient truths. But electronic communications have a fatal allure of virtual action.

Concerned about the environment? No need to go outside and walk in the woods, or clean up a polluted lot in your neighbourhood, or knock on your representative's door and explain the urgency of your position.

No, instead you can just fire off an email, write a blog posting, and then turn up the air conditioning and the lights and stretch out on the couch and read a good book.

That being said, I have myself translating the online into real world action on a number of occasions.  As I wrote in the StimulusWatch blog, it was an online posting that led me to an event that started a chain leading to the creation of the project.

That same event, and online chatter about a local conference, also led me (as partially outlined in my posting Making government data visible - and is Change coming to Ottawa?) to ChangeCamp Ottawa, a very real event happening at City Hall on May 16, which I have been helping to organise, an event which of course has a substantial online presence including a social network for the specific event, as well as being part of the larger ChangeCamp group on Facebook.

Similarly, a local news article in a free neighbourhood paper (yes, in print, with ink and everything) about a small garden/park space led me to a Facebook group which led me to an offline meeting which led me to create http://www.savethegarden.ca/

And of course, on a much more spectacular scale, the Obama campaign used (and continues to use) online organising as a tool, but they were very clear that the purpose of online was to drive a very extensive (and successful) ground game, people talking and knocking on doors, calling on phones, out in the real world.

So I think when it works best, the online world leads you offline, and offline leads you back online.  It's an ongoing discussion that flows across place and time.

Discussions enable meetings, data enables websites, websites enable more meetings, meetings come to consensus on APIs, APIs enable mashups... round and round it goes.