Update: Neo vs. Paleo Geographers - Either way you look at it, it’s not your “father’s GIS” [#gita]
Video now available on Peter Batty’s blog.
—- original post 4/28/10——
Neo? Paleo? It seems to be almost irrelevant to characterize the younger vs. the older technocrats of geospatial technology as associated with either camp. But it sure does make for a good panel discussion.
Lay the blame on Peter Batty of Ubisense who assembled a group of geospatial technorati to debate the current state of data sharing, data quality and the fate of standards in the new world of volunteered geographic information (VGI). Steve Coast, of OpenStreetMap (OSM), Andrew Turner of FortiusOne, Ron Lake of Galdos Systems, and James Fee of WeoGeo comprised the roundtable that discussed three primary aspects of the neogeo movement: Crowd sourcing; data quality and standardization.
How are conflicts resolved such as when there are varying accuracies if more than one person submits or adds data to the same area and what does this mean for data quality? Some in the audience questioned the moderation model for OSM and Coast stated that it is similar to that for Wikipedia now.
Ron Lake commented that "Wikipedia succeeded because the articles were not just written by "the average joe"; they were written by experts. Similarly, crowd sourced data is sometimes collected by regular citizens but also involves quality checks by GIS experts…and that’s why OSM can succeed as a wiki-type model.
Dr. Bob Austin of the City of Tampa, an audience member, asked "If I’m using OSM and complying with the license, how do I know someone hasn’t violated the license and using proprietary data?" This was a question that seemed to hang in the air. Certainly issues related to violation of licensing agreements and liability are beginning to surface with using OSM.
Another key topic in this discussion was the use of standards for data sharing and interoperability. Andrew Turner took Lake to task on GML indicating it was too complex of a specification. Lake said Turner’s description was "a massive misrepresentation of reality."
Lake indicated that GML is quite thick but the complexity of the specification is not due to complexity of GML. "The basic model is simple (object-property-value) but it contains a lot of different geometric types so people can build 3D geometries and time-dependent features. GML is a language to write schemas; you don’t visualize with it; you write those things in GML; spec size has to do with a dictionary of lots of types to use," said Lake.
Coast asked Turner, "Why is there all this format wanking," to which Turner replied, "I want all this data to be interchangeable…Standards are great and we all agree that we want interchange; I want everyone to be a part of the same conversation."
James Fee said, "You’re going to share your data by using the path of least resistance." Fee said he was still amazed that people are still using dBase; still using shapefiles. "How do we break out to start using some of these new tools… What makes a standard successful?
The discussion wrapped up by Batty asking the panel what will happen in the future. Coast said, "OSM works; it’s just going to be like Wikipedia; encyclopedias are gone. The stuff we are discussing are really ‘edge cases.’"
Turner said, "To be honest, there is an over emphasis on fragmentation; it’s great we are discussing it. We’ve hashed these out and now it’s time to collaborate… the problems we are thinking about ... only .1% of the people are thinking about them or care about them."
Lake said, "there is more common ground here than we would like to admit. But at the same time ... a popularity contest on how to decide things is rather dangerous," referring to more popular arguments surrounding OSM.
Fee said, "I don’t know if I agree with Ron. If nobody likes a format; no one will use it. It just works its way out; no one used to know about OSM; now everyone uses it. This next couple of years will be lots of fun."
Batty finished the discussion by commenting that "we’ve made huge progress on all this data sharing. I think the whole crowd sourcing business is very interesting to geo."
The discussion was captured by video and I will update this blog post once the link to it is released.
