The goal of the summit was to gather interested parties to take stock of the work done by the GeoRM Working Group launched in 2005 and move things along because the lack of rights management, per the webpage, "is a major barrier to broader adoption of Web based geospatial technologies." The summit was held in conjunction with the 69th OGC Technical Meetings and some other OGC summits (Sensor Web and 3D Fusion) that spanned the week. Despite the economy and other concerns, the technical meeting attendance was at 180+, matching or exceeding recent meetings before the economic downturn. (Update: Carl Reed reports that final attendance was 230+.)The GeoRM summit drew perhaps 50 for the daylong event, which included the Technical Meeting Plenary by the Sloan School's Michael Shrage.
The day started with a few perspectives on rights management. Note that the term is not digital rights management. OGC's working group dropped that term to separate itself from the negative connotations of that term, especially in the music arena. We got a great overview of the state of RM from Bill Rosenblatt of Giant Steps Media. He explained the different approaches (and successes and failures) of rights management in the consumer, enterprise, government and professional space. A quick summary: In the consumer space (music, e-books, videos) the goal is for vendors to "lock" users into their hardware. The technology is mediocre, but works in the sense that Apple achieved an 80% market share in MP3 devices, then offered DRM-free music! Enterprise implementations are far more effective and make money for the providers as part of larger enterprise systems. These solutions may include technologies like Microsoft SharePoint, Oracle and Documentum. (Rosenblatt skipped government as he's not very knowledgeable there.) The professional space refers to rights management for specialized e-periodicals, like those $5,000 reports you get from market research firms. That market is just emerging and the leaders are small companies of which I'd never heard: Vitrium, LockLizard, FileOpen.
Next we heard from a data provider. Kara John is vice president of intellectual property and privacy at Canada's DMTI Spatial. She highlighted the tricky aspects of some of the company's licensing, including "not making it look like you have access to personal information." How would you write that into an interoperable rights management system? She also referred to the 40-page license with complex "if then" types of rules that is not uncommon when DMTI Spatial provides data to clients. We referred to that document as the epitome of the complex situation we'd need to address with any complete solution.
Graham Vowles is the chair of the GeoRM working group and he outlined the history of its efforts. The group was formed in 2004 in Southampton, UK. The big achievements include the development of a reference model and GeoREL, geographic rights expression language, an XML language to describe access rights. GeoREL is now an ISO standard. The most visible efforts of the working group involved a pilot for ORCHESTRA, an effort to outline some possible licenses. The group extended the existing Creative Commons licenses with a few extra licenses to cover non-disclosure, commercial use and emergency use. I know I found the solution elegant and familiar. (I have spent a reasonable amount of time reading CC licenses for my work at Directions and at Penn State.)
Raj Singh, OGC's director of interoperability programs and compliance testing coordinator, introduced a case study proposed by his advisor a few years ago (Joseph Ferreira at MIT) which looked at MIT as a possible giant case study for everything from friend-finding to emergency situations.
We then had a few lightning talks. The first was the most technical talk of the day (which was not that technical, really). Hal Lockhart of Oracle introduced XACML, eXtensible Access Control Markup Language, a language that describes access control. To be clear, it describes a policy (who can and can't have access) but does not deal with enforcement or the quality or other matters related to the data. Lockhart explained that it can basically encode any possible access issues and, very importantly, add-on other XACML defined controls with ease.
Puneet Kishor, who has a fellowship with Science Commons, highlighted the special issues of scientific data: specifically, that they can't be copyrighted since they are factual. He discussed the Creative Commons "0" (CC0) license and Public Domain Certification (PDC). Like Kara John, he helped define the broad range of data types a complete solution would have to support. (He offers some thoughts on licensing geodata on his blog.)
We got a quick update from Matt Womer on the current work of the World Wide Web Consortium (W3C)'s Geolocation Working Group. Bottom line: There's a new draft of its geolocation API which, should all go well, will be a recommendation by the end of 2009. Some are not waiting that long and the API can be found in FireFox 3.5, a "labs" version of Opera and the Palm Pre browser.
The rest of the day was spent in open discussion (with the speakers as panelists) trying to tease out issues and how to define GeoRM licenses. As in all other OGC meetings I've attended the discussion was lively, passionate, opinionated and interesting. In addition to the speakers noted, many, many people - from Ed Parsons of Google, to Josh Lieberman of Traverse Technologies Inc., to an HP Intellectual Property (IP) lawyer - spoke their minds. And it seemed to work. Some very careful moderation by Jude Umeh of Capgemini kept the discussion (mostly) on track and by the end of the day Sam Bacharach of OGC was able to review a plan of attack for the future:
- Learn what we need to know - though we may know quite a bit from the last five years of the working group's efforts
- Define markets and use cases for GeoRM
- Divide the results into a nested hierarchy (I called it a "grid")
- Determine the easiest and most important (they may not be the same) cases, and build a "straw man" based on existent licenses (potentially from OGC member organizations)
- Spin up a test bed for input from the real world and see what happens
What OGC does is not easy or academic, nor is it beyond the comprehension of anyone who is truly interested. If you get a chance to participate in a summit like this, take it. You'll meet some interesting people. I had lunch with an IP lawyer and a geo-librarian from the University of North Carolina. I guarantee you will learn something and potentially help move the effort forward. You might even learn some "harmonization techniques" you can use in your own organization.