OGC On Google KML Submission
Mike Liebhold asked about “the scope and implications of Google’s efforts re: OGC adoption of KML” on the GeoRSS mailing list. OGC’s Carl Reed replied:
A bit on the submission by Google of KML into the OGC process.
At the December San Diego meetings, Michael Jones, John Hanke, and Brian McClendon collectively spoke to the OGC Technical Committee in a Plenary session. One of the topics they discussed was a proposal to submit KML into the OGC standardization process. The next day at the OGC Planning Committee meeting, the PC members in attendance had a very open and frank discussion regarding Google’s proposal. We covered such topics as how to best (and to what extent) KML should be harmonized with other OGC standards, the standardization timeline, intellectual property and copyright, how to make sure that the current (and future) KML developer community can remain engaged in the process without being OGC members, backwards compatibility issues, and so forth.
The motion as approved by the OGC membership with endorsement by Google:
a.. KML will be submitted to the OGC by the 3 week rule for the April meetings for consideration as an OGC Best Practices paper
b.. The new Mass-Market Geo Working Group will be the home for discussions related to KML.
c.. That a new OGC public discussion list (.dev) will be started for KML to allow coordination and engagement with the KML developer community.
d.. That the OGC members will begin work on an initial, but limited, harmonization of KML with existing OGC and ISO standards. Stated work items include coordinate reference systems and geometry. The results of this work will be a candidate specification for consideration by the OGC membership for approval as an adopted OpenGIS specification. (Target date: end of 2007 early 2008)
e.. Staff will work with Google and Mass Market Geo WG to facilitate this process.
f.. There needs to be a position paper that clearly defines the problem domain that GML solves and the problem domain that KML solves.
I am currently in the process of putting the KML reference guide into the OGC document format (including maintaining all links). This document will be posted to the OGC pending documents archive for discussion at the April meetings sometime next week.
The key short term item beyond document formatting is developing the position paper that clearly defines the problem domain that GML solves and the problem domain that KML solves. I believe that there is a fair amount of confusion in the community as to what KML is best suited for and what GML is best suited for. The issue is doubly interesting given that the geometry elements in KML are identical to GML 2.1.2. We will be working on this position paper over the next month or so.
Borrowing from Ron Lake and from discussions with GE staff, we think KML and GML are targeted at solving different problems. This has nothing to do with complexity vs simplicity - but rather just different objectives and requirements. KML is fundamentally focused on Geographic Visualization - meaning visualization of places on the earth - and annotating or describing places. It is not intended to model geographic objects. KML could even contain additional GML elements. KML, because it is connected to the description of place is also (KML Search) a means of providing spatial indexing - and this is being done through the Google robot.
And for additional reflections on the legal aspects of this topic, I would suggest visiting Raj Singh’s blog http://www.rajsingh.org/blog/?p=18 . If anyone on this list has any thoughts, suggestions, or concerns regarding the Google submission of KML into the OGC process, please let me know.
The thread includes questions on process and public participation.