The tragedy of Edmontorcouver open data
Submitted by rw on Sun, 03/07/2010 - 17:33.
Edmontorcouver is a lovely place full of lovely, intelligent, generous people. That is not the tragedy of Edmontorcouver open data.
The mayor of Edmontorcouver realized that it would benefit the people of Edmontorcouver to have good data about their city widely available. That is not the tragedy of Edmontorcouver open data.
Edmontorcouver hosted a wonderful gala to announce the generous release of great quantities of data. That is not the tragedy of Edmontorcouver open data.
The tragedy of Edmontorcouver open data is that the data is not being widely used. The tragedy of Edmontorcouver open data is that it appears to a naive viewer to be a failed program.
Two minor mistakes
There are two minor mistakes that are holding Edmontorcouver open data back from wide adoption, two minor mistakes that are preventing the Edmontorcouver opendata initiative from being a tremendous success. Tiny mistakes. Easy to overlook. These are the O-rings on Edmontorcouver open data Success Boosters.
These mistakes are related. It is a commonly held myth that a community will be created spontaneously by the release of open data. Tellingly that myth is not held by anybody who has successfully created a community.
One can measure the likelihood of success in a community with the Fail Scorecard. While intended for open source software projects the application to open data projects is obvious. Application to the Edmontorcouver open data project gives an unfortunately high Fail score.
When you buy into the "the community will build itself" myth, it is no surprise that you don't assign the resources required to build the community successfully. "Why assign resources? This will be spontaneously successful!" And frankly, Edmontorcouver is not in the business of building open data communities. The taxpayers would, rightly, squeal if asked to pay an open data community levy on their property tax.
Here is a diagram of the Edmontorcouver open data community.
So what to do? The Edmontorcouver community is too small and has failed to reach critical mass and become self-sustaining. How does Edmontorcouver fix this?
Be the curator of your data, not curator of a community
Make your data available to the widest possible audience. Make it available to every open data community. Make the Edmontorcouver data the very best data that it can be. This is already your mission. This is your obligation to your citizens.
The OpenStreetMap community, for instance, already numbers more than 226,000 members around the globe. The OpenStreetMap community has been doubling in size every few months for years. It is a successful open data community.
Let the community experts curate the communities
You want OpenStreetMap and other successful geo data communities to use your data. Edmontorcouver data, as part of these global data sets will be seen by your citizens and will become useful to them. Hedge your bets, get every open data community to use your data.
So why hasn't the OpenStreetMap community or others started working with the Edmontorcouver open data?
License your data for the widest possible audience
It's the license. The Edmontorcouver license is incompatible with the OpenStreetMap community. And that has scuttled the dream of Edmontorcouver becoming a shining example of a successful open geo data project.
As Edmontorcouver shouldn't be curator to an open data community when an existing community could do a better job, also Edmontorcouver shouldn't be curator to an open data license when an existing license will do a better job.
Fortunately, the good folks at the Open Data Commons make it easy to select an appropriate license for your open data. Use the Public Domain Dedication and License if you want protect your interests by disclaiming liability and you want your data to have the widest possible audience. [Hint: you do!]
Summary
How do we improve the Edmontorcouver open data initiative and rescue it from the appearance of failure?
EdmonTorCouver References
Statue photo 
CCBYSA by fabbio.
Articles © 2010
looks like more of an OSM problem
It looks more likely that the OSM folks haven't looked at the licence. Seems perfectly compatible to me.
Incorrect, scruss. I'm an
Incorrect, scruss. I'm an OSM-folk and I'm drawing attention to the problems with the license. ;-)
Can you please link to the
Can you please link to the Edmontorcouver data and/or license and/or website?
See the references above.
See the references above.
Hi there, Could you please
Hi there,
Could you please clarify the incompatibility? The terms of usage describe a “worldwide, royalty-free, non-exclusive licence to use, modify, and distribute the datasets in all current and future media and formats for any lawful purpose” which sounds pretty good.
License and community
Hi ,
I've been advising the Mayor of Vancouver and working with the city on the open data project. Your post is interesting to read and I'm really trying to draw some helpful advice from it.
We worked very hard to make the License as open as possible... I hear that it isn't satisfying all your concerns though. I'm trying to draw lessons from your post and am finding it difficult - would love to hear more. I'll flag PDDL and see what traction it might get and concerns it might raise internally. I really want to help OSM so let me (and the gov community) so it would be helpful to hear what specifically is it that the present license doesn't do? What interests does PDDL satisfy that the present license does not? Would be great to get your feedback on this.
On the community part, I'll confess it is hard to read this section. Here in Vancouver we (especially with the leadership of Luke Closs) regularly hold OpenData Hackathons (often coordinated here), these events are kindly hosted by the City (at the City Archives - who provide Internet Access, coffee, and donuts), we open with circles so everyone can get to know one another and the projects they are working on, and often city staff - especially Jonathan Mark who knows the data best - shows up to answer questions and find out what other data sets people need. Indeed, on a regularly basis he has prioritized data sets that people have requested for applications they are imagining. This is an explicit effort to build a community around open data. My sense is that the use of open data is going to be enormous, but it is also going to take years to fully develop - we are doing our best to build community here, and if you live in Vancouver please come join us, and let me know what you think we could be doing better...
Cheers,
Dave Eaves
Also consider a related
Also consider a related article about a better, but still imperfect, municipal data license.
http://weait.com/content/better-approach-municipal-open-data
Consider reading this
Consider reading this detailed review of the weaknesses in the early version of the open data license used in Edmonton, Toronto and Vancouver.
http://weait.com/content/unintended-restrictions
community building
I am one of those people at the innermost circle -- interested, have the skills, wanting to do something, but as of yet, I haven't finished anything particularly interesting having to do with the City of Vancouver data. I have one half-way finished project and one barely-started project.
One reason is that the open data project has been quite good. In particular, with the second(?) revision, the city offered KMZ files with one-click to show them on Google or Bing maps. I think that I could apply some effort to make them even better (for example, when the KML is displayed in Google Maps, the title in the sidebar frequently (always?) is not meaningful). However, a) they are pretty good and b) I'm not interested in putting up a rival version which would draw away from the City version. I could probably work with the city to improve their KMZs, but it's not clear that they would particularly want the help -- it would take some City time and resources to work with me on that.
Another reason is that there is no obvious way for me to profit from finishing the projects, financially or otherwise. I'm not sure that anybody would care about either of my two projects, even if I were to finish them. :-(
I did spend an enormous amount of effort on Australian open data and NYC open data projects, even though I think I probably wasn't eligible for the NYC cash prize. Why? Because if I had won the contest, it would have looked awfully durn good on a resume, something that was important to me at the time. (I have since found a job, so am not as hungry, but a little web services company might be motivated.) I suspect that a prize of as little as $500 (plus bragging rights!) would get a lot of people active.
Post new comment