[Geowanking] ogc control? of kml

Tim Bowden tim.bowden at westnet.com.au
Sat Apr 26 03:54:41 PDT 2008


Died before it got to the list so trying again...

On Wed, 2008-04-23 at 17:07 -0700, Mike Liebhold wrote:
> Hi Mano,
> 
> Thanks for your reply. Your answer is  only a very literal
> interpretation of the first part of my question. What I was really
> asking about was the de-facto KML that google implements unilaterally
> in google earth way ahead formal OGC adoption. There are many KML
> elements  in google earth that  are -not- part of KML 2.2.  e.g.
> KML3.0++  Doesn't google still control the future of KML through these
> pre-standard implementations?
> 

Mike, I hope you're not implying that google (or anybody else) shouldn't
further develop *and implement* tech that has been embraced by a
standard.  IMHO good engineering requires that the standard comes
*last*.  Development is (comparatively) fast and cheap, whilst standards
are slow and expensive, particularly if they get it wrong.  Solve the
problem first, then the standard more or less becomes the documentation
as to how the solution works, with an appropriate (not excessive) level
of abstraction such that the solution (as specified by the standard) can
be reasonably applied to that class of problem.

A significant wrinkle of course is that extended implementations of
standards (which I've more or less just advocated) interfere with
interoperability (as we've all seen with embrace and extend from
elsewhere).  If you wait for the next version of a standard before
moving on with implementation, you're crippling both the standard and
tech development.

The solution is not to hamper the rapid development and implementation
of the underlying tech, or rush a standard to try and keep up.  Rather,
play to the strengths of both, but do extended dev in an open and
forthright manner offering to let others play in the sandpit with you on
the understanding that it's all up for grabs in the next round of
standardisation.

That of course requires a measure of goodwill that is often lacking, but
I don't think you're going to get a clean resolution to that problem.
The underlying cause is the disruptive tension between development and
standards, and I don't believe it's possible eliminate that tension
given the opposing fundamental natures of those two beasts.

Regards,
Tim Bowden

> My other question  is what is left out of the OGC KML2.2 spec ? I
> understand that a fair amount of 3d grammar is not part of KML 2.2.
> What else?
> 
> In addition to a formal  Google response (yours), I was hoping that
> someone like Raj Singh, Carl Reed, Ron Lake, or Andrew Turner might
> also comment.
> 
> Thanks for any insights you can share
> 
> Mike
> 
> 
> 
> Mano Marks wrote: 
> > Hi y'all,
> > 
> > I just wanted to address a couple of the points raised on the thread.
> > To be completely out there, in case you missed my email address, I
> > work for Google supporting Geo APIs.
> > 
> > Mike asked if the OGC owns the future of KML. It does. Google no
> > longer owns any KML IP.. The OGC owns all of it. Google is a member of
> > the OGC, and we hope to be one of the drivers of innovation, but we
> > don't own it or have any special privileges other OGC members don't
> > have.
> > 
> > Allan, you asked for a non license-agreement copy of something. Did
> > you mean of the spec or just the documentation?
> > This link:
> > http://schemas.opengis.net/kml/2.2.0/
> > doesn't require signing a license
> > 
> > Also, the KML 2.2 SWG at OGC is also open:
> > http://www.opengeospatial.org/projects/groups/kml2.2swg
> > 
> > Also, the KML docs on Google's site are pretty up to date.
> > http://code.google.com/apis/kml/documentation/
> > 
> > If it is a different doc, I can see if I can get it for you.
> > 
> > Mano
> > 
> > On Fri, Apr 18, 2008 at 6:18 PM, Allan Doyle <afdoyle at mit.edu> wrote:
> >   
> > > It's a little hard to tell, they have a click-through license at the OGC
> > > site. Does Google have a non-DRM'ed copy anywhere?
> > > 
> > >         Allan
> > > 
> > > 
> > > 
> > >  On Apr 18, 2008, at 2:37 PM, Mike Liebhold wrote:
> > > 
> > >     
> > > > More (perhaps old) news ( 4/13)   " OGC(R) Approves KML as Open Standard"
> > > >       
> > > The formal press release from OGC appended below, and this on a google blog
> > >     
> > > http://google-latlong.blogspot.com/2008/04/kml-new-standard-for-sharing-maps.html
> > >     
> > > > "Starting today, Google no longer controls KML. The Open Geospatial
> > > >       
> > > Consortium (OGC), an international standards body, has announced the
> > > completion of KML's standardization process. KML has become an OGC Standard,
> > > and the OGC will take responsibility for maintaining and extending it. This
> > > transfer of ownership is a strong reflection of Google's commitment to open
> > > standards. Fundamentally, our interest is not to control information, but
> > > rather to encourage its spread."
> > >     
> > > > Despite this  announcement , does the OGC community  actually now 'own'
> > > >       
> > > the future of kml?  I'm not sure.
> > >     
> > > > Perhaps someone more knowledgeable can comment  on whether not Google
> > > >       
> > > actually still controls KML destiny, with all of their queued up kml3.0 and
> > > beyond features and extensions  they can unilaterally adopt and  implement
> > > in google earth, before submission or adoption by OGC committees.
> > >     
> > > > I'm also wondering about highlights of alignment with GML, and what major
> > > >       
> > > elements are left out  or postponed in this 'standard' version.  Some 3d
> > > elements perhaps?
> > >     
> > > > Raj or Carl or  Ron anyone in the OGC  KML groups care to share any
> > > >       
> > > comments?  (Andrew?)
> > >     
> > > -------------------------------------------------------------------------------------
> > >     
> > > > http://www.opengeospatial.org/pressroom/pressreleases/857
> > > > 
> > > > OGC(R) Approves KML as Open Standard
> > > > 
> > > > Wayland, Mass., April 14, 2008 - The members of the Open Geospatial
> > > >       
> > > Consortium, Inc. (OGC) today announced the approval of the OpenGIS(R) KML
> > > Encoding Standard (OGC KML), marking KML's transition into an open standard
> > > which will be maintained by the OGC. Developers will now have a standard
> > > approach for using KML to code and share visual geographic content in
> > > existing or future web-based online maps and 3D geospatial browsers like
> > > Google EarthTM.
> > >     
> > > > "We are pleased to see the adoption of KML as an OGC standard," said Ron
> > > >       
> > > Lake, chairman and chief executive officer of Galdos Systems Inc. "We
> > > believe that this is a major step forward for the OGC and for the entire
> > > geographic information community, as it provides the first broadly accepted
> > > standard for the visualization of geographic information."
> > >     
> > > > "Geographic data adds tremendous value to the online experience. More and
> > > >       
> > > more people are looking for ways to incorporate location information into
> > > their online content," said Michael Weiss-Malik, KML product manager for
> > > Google. "The standardization of KML makes it possible for both novice and
> > > expert users alike to publish and share geographical information in an open
> > > format. It's not unlike web browsers' standardized support for HTML, which
> > > allows any web browser to read any web page."
> > >     
> > > > KML version 2.2 was brought into the OGC consensus process by a submission
> > > >       
> > > team led by Google and Galdos Systems Inc.
> > >     
> > > > KML is an XML-based programming language, originally developed to manage
> > > >       
> > > the display of geospatial data in Google Earth. It's still used heavily in
> > > Google Earth but is also supported by a variety of vendors' tools and
> > > mapping websites.
> > >     
> > > > The OpenGIS KML 2.2 Encoding Standard formalizes the KML 2.2 model and
> > > >       
> > > language while remaining backwards compatible with existing KML 2.2 files
> > > and tools. In comparison with the GoogleTM KML 2.2 Reference, the standard
> > > defines:
> > >     
> > > > * the KML 2.2 geometry encoding and interpolation model
> > > >  * an extension model in support of application profiles
> > > >  * conformance requirements and test cases
> > > > 
> > > > The adopted OpenGIS KML 2.2 Encoding Standard (OGC KML) is available at
> > > > http://www.opengeospatial.org/standards/kml/.
> > > > 
> > > > About the OGC
> > > > 
> > > > The OGC(R) is an international consortium of more than 345 companies,
> > > >       
> > > government agencies, research organizations, and universities participating
> > > in a consensus process to develop publicly available geospatial standards.
> > > OpenGIS(R) Standards support interoperable solutions that "geo-enable" the
> > > Web, wireless and location-based services, and mainstream IT. OGC Standards
> > > empower technology developers to make geospatial information and services
> > > accessible and useful with any application that needs to be geospatially
> > > enabled. Visit the OGC website at http://www.opengeospatial.org/.
> > >     
> > > > Google and Google Earth are trademarks of Google Inc.





More information about the Geowanking mailing list