Nestoria Interview: Steve Chilton - OpenStreetMap expert/author

This month the Nestoria UK blog spoke to Steve Chilton, co-editor of the recently published book:  "OpenStreetMap - Using and Enhancing the Free Map of the World".   Australia's OSM community might not be quite as big as the UK's yet, but I'm hoping that what he had to say may inspire a few more people to get involved.  So here it is re-edited for an Australian audience, but you can read the whole interview on the UK blog.

Steve manages the Educational Development Team within the Centre for Learning and Teaching Enhancement (CLTE) at Middlesex University, and is also one of the University's Teaching Fellows. By training Steve is a cartographer, having originally studied at Oxford Brookes, and is a research cartographer at the University, and Chair of the Society of Cartographers.  Steve has mapped large sections of Enfield/North London for OpenStreetMap, as well as many other places in the UK and abroad in his travels. He is the main designer of the look of (and additions to) the mapnik layer of the default slippy map.  Steve co-edited the English version of this definitive OSM book with Frederik Ramm and Jochen Topf.

Steve, thanks for talking with us (well, the UK blog readers and us by default). On to the questions!
The global OSM community has grown rapidly over the last few years. Nevertheless, newcomers often complain of the high barrier to entry. Does your book attempt to address this?

The book doesn't specifically set out with that in mind. To quote from the jacket blurb "The book explains the community, the data model and the software behind the endeavour. It enables you to use OpenStreetMap's ever growing body of geodata in your own projects." Expanding on this, there are sections within the book that cover how to use the two main map 'renderers' (Mapnik and Osmarender), with simple worked examples that explain everything in a stepwise manner. There is also a chapter that takes you through the basics of mapping practice. Some examples from within the book are also available on the accompanying website http://www.openstreetmap.info/ , where you can also see the full contents, plus a sample chapter - it is actually the 'Mapping practice' one just referred to. It is my hope that people within the project will soon apply their abilities to addressing the perceived high barrier to entry that you mention.

As a long time OSM contributor, what are some of the things that have surprised you the most about the project? What do you see as the biggest challenges ahead?

Well, when I first heard the other Steve C presenting his vision for OpenStreetMap at a Society of Cartographers conference in 2005 I was very sceptical, and thought to myself "well that will never happen". Well it did happen and it is now a big player in a seriously changing geodata landscape. All sorts of things surprise me about the project. Maybe I shouldn't be (knowing my own situation), but I am constantly surprised by the commitment some particular individuals make to the project. The mapping hours that folk put in are just amazing, as are the software development pathways that we have seen - all in people's 'spare' time remember. There have been a couple of supernovas that flared up and then disappeared, but many of the major players have been prominent from quite early days.

There are several challenges as I see it. Obviously there is the one of getting more comprehensive global data coverage. Then there is the maintenance of the existing data. Will the currently active mappers want to do the less 'exciting' mapping and routine checking that needs to be done to keep the data with its present currency of information advantage? Conversely, will new project members come along who love that aspect and develop new ways of ensuring it happens? Developing a really basic entry level portal for people to add or edit what they see in a non-techy environs is a challenge we must face too I feel, and also more simple data export/conversion capabilities. The planned change of licence from Creative Commons to ODBL has also created some serious potential divisiveness. Managing this process and any possible forks in the project are another big challenge that has to be faced.

In the past few months there have been more and more announcements of companies embracing OSM, be it start-ups or megacorps like Microsoft. How do you predict this will impact OSM, as a data repository, as a web service, and as a community?

I have said elsewhere that my ability to predict the future is about as good as Thomas Watson's of IBM was way back when, but I'll have a go. The interest in OSM shown by the likes of Microsoft and Mapquest recently is only to the good in my view. At the time of these two events I did predict the possibility of some head hunting of major players in the project with either software or project management skills, and to some extent we have seen a little of that already. But I firmly believe that there is such a large pool of fantastically skilled people in the project that I am confidant others will step up to the plate in any situation where this might happen. Also if I may cite the example of Mapquest, they do seem to have taken a pretty altruistic approach, and there seems to be areas of mutual interest between themselves and OSM. We have already seen that in the UK the freeing of some OS data last April has NOT had a dramatic effect on mappers or on data importing in OSM.

Having said that the project will have to make decisions as it goes along about what data is appropriate for the OSM database and what might not be. There is already discussion about keeping data in the DB for things that are no longer there (demolished buildings, dismantled railway lines, etc) - as this has real value for people studying historical geographic data. The availability of OSM data though Mapquest and Bing gives another possible source for people to access the data in mapped form. I think there will increasingly be an opportunity for people/companies to provide customisable map output using the data. Cloudmade already have this, up to a point - but the styles are not portable, the Mapquest ones are already. I would also like to see a major commitment from the members of the OSM community to engage with both the major map and data suppliers, and also governmental and other organisations - in both cases with a view to future possible collaborations.

You're not just a man of letters, but also a man of action - a few years back you headed into the "field" to map Antigua. Tell us a bit about your adventure. What opportunities and challenges are there for OSM in the developing world?

Antigua? I was SO naive and under-prepared for that. Knew it would be hot, but just took my un-reconstructed European mindset with me. Assumed I would just hire a bike, ride that around all day, mapping all the streets/names and POIs/addresses as I went, and cover the whole island in a week. Well - no bike hire to be found, too hot to map for more that about 3 hours in any one day, hardly any streets had nameplates and virtually no house numbers, or even names for POIs such as roadside cafes, etc. Also, I had some good contacts who seemed to promise geo-data from various local govt department sources. All saying yes/soon and stuff, but in the end not coming away with any. But lovely friendly people everywhere I went, including an amazingly laidback TV studio when I managed to swing a breakfast time slot for being a newsworthy novelty visitor - announced as "world renowned cartographer Professor Steve Chilton". Still living off that story! But seriously, there are immense opportunities for OSM to make a real difference to people's lives in the developing world. You only need to look at the effect that the OSM project furnishing up-to-date geodata in both Haiti and Kibera recently has had to see that.

Many thanks Steve for the interview and for your contribution to the OSM community. We highly recommend the new book for anyone looking to quickly grasp the key concepts and technologies behind OpenStreetMap.

Those who are interested should follow Steve on twitter (and of course you can follow all Nestoria interviewees via our twitter list.)

Have GPS and a bit of free time?

What are you doing this weekend?  If you live in Brisbane, this blog may just have become your entertainments officer, because there's an Open Street Map Mapping Party on this Saturday (the 17th of April).  It's on at the Brisbane City Library, so it's not a roaming the streets mapping event, it's a sit down one designed to get people together to talk about the mapping that's going on in Brisbane with an eye to sharing the knowledge, fixing errors and improving the quality of the mapping alignments.

David Dean, one of Brisbane's OSM heros, says there will be people there who know about Potlatch, JOSM, Merkaartor and Mapzen, and who can give you hints and tips – so this is your chance to become involved in mapping your part of Australia for posterities sake.

If you're inspired, or if Saturday's not enough notice for you, there's also a Mapping Party going on in Coomera, one of the largest suburbs on the Gold Coast, on Saturday the 15th of May.  This suburb not only has varied terrain – from mangroves to hills and even its own island, it's also got changing traffic conditions – so it needs some OSM love.

David spoke to us in his interview about how easy it is to get involved and help out mapping your local locale.

How the web has helped in Haiti

On September 11 2001 when the twin towers fell people were glued to their television screens or radios following events as they unfolded, but 8 years on when the earthquake struck Haiti people found there were there other ways of keeping abreast of the news.  It didn't take long for Google to release a KML of their latest Haitian imagery which could be viewed on Google Earth, giving people a birds eye view of the destruction.

Here are before-and-after screenshots of the Presidential Palace and an area of Port-au-Prince:      

  

OpenStreet Map sprang into action as well, with mappers on the ground and remote mappers working off the latest CIA maps and GeoEye imagery showing the latest building collapses toiling towards creating an up to date GPS layout.  This was of huge assistance to emergency workers who would have taken a lot longer to reach victims using their pre quake GPS downloads.

The rest of the world could then monitor emergency situations on the Crisis Map of Haiti, a mash up of reported emergencies that people could report by text, phone call, twitter, Facebook, live streams and information reports – these emergencies have been anything from missing people reports, to fires, contaminated water supplies, aftershocks or collapsing buildings, to personal calls for help from families in their homes running low on food and clean water.  Most of the mapping is happening in America and Europe, but it's been happening in real time none the less.

Almost a month on from the quake the internet is still coming up with fresh ways for people to stay tuned with what's happening in Haiti, but also to lend a hand.  It may not be the most obvious mash up, but Gruvr has added a Haitian Benefit Concert Finder to their concert listing site, which maps your nearest fund raising concerts – they say that there are about 30 events being added a day.

Nestoria Interview: David Dean on OpenStreetMap Australia

As well as viewing the Nestoria listings on the GoogleMaps interface you can also look at them on OpenStreetMap, which is all well and good if you live in the UK where there have been big OpenStreetMap meets designed to map whole cities all at once, but how's the lay of the OpenStreetMap land in Australia?  We'd better chat to someone who knows and find out...  That someone being David Dean, Speech, Audio, Image and Video Technologies PhD, and the member of the Brisbane OSM community most likely to be organising the next meet up or party.

First up how did you get into mapping and OpenStreetMap?

I first found out about the project soon after it began through a link on Boing Boing, but then forgot about it for a few years until a geocaching friend mentioned it around 2007. I had a look around Brisbane then and coverage was pretty poor so I got started.

What are the OSM projects you're interested in or engaged with at the
moment?  Or are you just mapping as you go?

I've done a little bit of bug fixing and minor coding for the gosmore navigation project, but for the most part I just get out there and map. I'm generally more interested in walking or riding out in the field to collect street names and other map features that are hard to get from satellite imagery. I'm happy to leave the armchair mapping to others.

What would you say is your mapping 'claim to fame'?

I seem to have become the main organiser of OpenStreetMap community events in Brisbane. We have organised mapping events for more than a year now, and they have been monthly since July 2009, with growing attendances, which is encouraging.

How would you say the Australian OSM community is doing in terms of the rest of the globe?  And are there regions of Australia where you think OSM is better than professional maps?  Where would you place your home town, Brisbane, on regards other capital cities?

I can only really talk from personal experience in Brisbane, but I'm fairly certain that OpenStreetMap has good coverage in most of the larger cities of Australia, but probably nowhere near the detail that OpenStreetMap has in the UK or Europe. Brisbane's coverage is pretty good these days, but I think Melbourne, Sydney and Canberra probably have better coverage at the moment.

And what about in terms of other paid for maps?

In comparison to commercial maps, I would say that wherever someone has done a detailed ground survey they are probably the best maps you can get. I'm particular happy with the use of OpenStreetMap for hiking/cycling trails and paths, because most of the commercial mapping providers are not even interested in collecting this data.

There are a few problems that people always site about open sourced mapping, things like how reliable is volunteered information?  How should data be licensed?  How do you expand the number of users to improve the quality? Do you consider any of these problems to be major challenges in Australia?  And if not what do you personally feel are  the big challenges for OSM Australia?

I think the 'problem' with volunteered information has already proven to be minimal by Wikipedia, which I doubt many internet users can go a week (or even a day) without using to learn something new.

Of course, I think that OpenStreetMap needs an easier way to quickly revert vandalism edits, but I believe there are some technical problems to making this a one-click process like it is for Wikipedia. However, despite the exponential user growth, I don't believe that vandalism is a bug problem at the moment. In Brisbane, I've only had to revert one case of vandalism, and that user gave up relatively quickly.

I don't think OpenStreetMap is having any problems expanding its user base, but I'm sure there are still a large volume of people who would be interested in OpenStreetMap, if only they knew that it existed.

One of the major issue for OpenStreetMap in Australia will be how are we going to go about getting all of the roads. As Australia is a large country with a very low population density there are a lot of roads to map, and it is clearly not something that will be completed if people only map their local area (unless we can convince everyone in Australia to help).

While there are some datasets that we could potentially convince the government to release, it is unclear how reliable this data is, with many rural areas reported as having many incorrect details, particular with gazetted roads that don't actually exist. Some of the commercial mapping providers who care about accuracy pay workers who drive around Australia full-time to collect these details, but it is not clear how OpenStreetMap can collect similar information (and keep it up-to-date) purely through volunteer methods.

What kind of things to you see on the OSM horizon?

I think the best thing to improve OpenStreetMap is to get more people using it. If we can get people using OpenStreetMap maps in their car navigation devices or on their mobile phones they will have the best maps possible in many areas, and when they spot mistakes they can fix them easily, or just pop up a flag to let more dedicated mappers have a look at the problem. However, there still isn't really any software using OpenStreetMaps that approaches the ease of use of modern car navigation software, but I know people are working on it...

You said you'd heard of Nestoria - do you have any advice for us entering the Australian market?  Or any suggestions about how to improve what we've got at the moment?

Make the OpenStreetMaps default instead of Google :) .

I don't have a lot of experience in real estate websites, but it'll be interesting to see how Nestoria manages to crack the near-monopoly that the existing main real estate has in Australia. I've tried using other real estate websites in Brisbane before, but found that there was nowhere near as many places listed as the main real estate website that nearly everyone uses.

Is there anything else you'd like to talk about now you've hopefully got flowing?

If the reader is interested in OpenStreetMap, go and attend a mapping party in your local area (check wiki.osm.org). If your local area doesn't have any mapping parties, start one!

Thanks again for taking the time to share your Aussie take on OpenStreetMap.