In the run up to the State of the Map 2008 we’re going to be talking to some of the speakers to see what they’re up to. This week I caught up with Nestoria’s founder and CEO, Ed Freyfogle to see what he thinks of OSM:
You’ve been involved with OpenStreetMap for a while now. What first
drew your attention to the project?
When we started Nestoria, our property search engine, in 2006 we
needed geodata to do things like geocode properties and show relevant
content on maps. I was blown away to discover how much this type of
data (which, like most people I assumed to be in the public domain)
typically costs, especially in the UK, so we started looking for
alternatives. I’ve used open source software for many years, so I
knew the benefits of the opensource model, and it’s something we
actively try to encourage – at Nestoria we do our best to support the
perl community for example.
So we discovered OSM and immediately saw the potential, even if it was
a bit rough around the edges at that time. It’s clear that OSM
attracts an amazing international community of talented people doing
very innovative work with geodata.
What are the biggest changes you’ve seen in OpenStreetMap over the
last few years?
Two major issues: usability and comprehensiveness. As the project has
gained momentum, comprehensiveness is becoming less and less of an
issue. I now see usability as the major bottle neck. Even for tech
savvy folks it can be unclear how to put data into OSM and more
importantly how to take it out and use it.
Tell us a little bit about your talk at SOTM 2008.
I’ll be going over some of the examples of ways we’ve experimented
with OSM data at Nestoria, what has worked well, what could be
better. If commercial use of OSM is a goal (and possibly for many volunteers
it isn’t) there are still a few barriers.
How have Nestoria been using OSM’s data? What are your plans for
using it in the future?
So far much of the work we’ve done with OSM has been more experimental
or proof of concept rather than mission critical. We’d like that to
change in the future.
What in your opinion, are the biggest barriers to wider commercial
usage of OSM data?
Beyond a doubt the biggest barrier is usability (not
comprehensiveness). OSM is an amazing dataset, that gets better every
day, but the learning curve to play with the data is still too steep
in my opinion.