Relationship with GeoAPI

[Martin Desruisseaux <ma...@geomatys.fr>]

Hello all

On the work organisation topic, I would like to know SIS opinion on the 
relationship with GeoAPI (http://www.geoapi.org). GeoAPI is an attempt 
to standardize interfaces derived from OGC specifications. If SIS wishes 
to be a GeoAPI implementation, this would have a strong influence on its 
API. GeoAPI development actually happen in two places:

* GeoAPI is hosted on SourceForge and have a public mailing list (quite 
inactive for a while)
* GeoAPI is also an OGC Standard Working Group and has a second mailing 
list accessible only to OGC members.

The politic up to date has been to use the public mailing list for all 
developments, and the OGC mailing list only for the standardisation 
process inside OGC (vote for submission as a standard, etc.). The public 
group can deploy as many milestones as they wish, but official releases 
(I mean with "OGC standard" label) can be performed only by OGC. To date 
there is only one official release (GeoAPI 3.0.0) and many milestones.

GeoAPI development tends to be very slow, because GeoAPI mission is to 
create interfaces from OGC standards. The main debates are about how to 
interpret a specific element of a standard. Some additions do happen, 
but are expected to be rare. API not covered by standard stay 
project-specific, so SIS would have its own API as a complement of the 
GeoAPI one.

What are the opinions about SIS-GeoAPI relationship?

     Martin

Re: Relationship with GeoAPI

["Mattmann, Chris A (388J)" <ch...@jpl.nasa.gov>]

Hey Martin,

Apache had a long history of helping to participate and in many ways, *drive* the JCP 
process (Java Community Process) at Oracle. We went so far as an organization to establish 
a VP of JCP, and then have direct interaction including even having some members of our
organization participate as individuals in those closed lists and to have our projects implement
JCP standards and APIs.

The board of the ASF leaves it to the Project Management Committees to decide how to 
run their projects, so if the SIS PPMC (we're not a full project management committee yet,
since we're still Incubating, so we are still under the umbrella of the Apache Incubator and
its IPMC) desires to participate in the GeoAPI standards organization, our best bet would be
to get the relevant Apache organizational parties involved to make sure things are coordinated
as we engage in the process.

For example, the first thing I see is:

http://www.geoapi.org/license.html

That "sort of" looks BSD-y to me, but I am not sure. So we'd probably have to bind to those
Java APIs led by the GeoAPI spec and then *implement* them here, which means we'd have
to consume that upstream API jar file from GeoAPI. As long as the license is kosher, we're 
probably OK as a first pass to start implementing that API inside of SIS. Other than that, I 
can't thing of anything else we need at this point. As we graduate and become a Top Level
Project (which I'd say we're pretty close to in the next few months), we may want to coordinate
this with the e.g. Board, Legal Committee, etc., just to make sure we're good and consider having
a representative from the ASF participate in the OGC, the same way we have ASF representation
inside of the W3C, and other standards bodies (e.g., we have a VP, W3C, which currently is
Andy Seaborne and used to be Sam Ruby).

My 2c.

Cheers,
Chris


On Aug 1, 2012, at 7:38 AM, Martin Desruisseaux wrote:

> Hello all
> 
> On the work organisation topic, I would like to know SIS opinion on the relationship with GeoAPI (http://www.geoapi.org). GeoAPI is an attempt to standardize interfaces derived from OGC specifications. If SIS wishes to be a GeoAPI implementation, this would have a strong influence on its API. GeoAPI development actually happen in two places:
> 
> * GeoAPI is hosted on SourceForge and have a public mailing list (quite inactive for a while)
> * GeoAPI is also an OGC Standard Working Group and has a second mailing list accessible only to OGC members.
> 
> The politic up to date has been to use the public mailing list for all developments, and the OGC mailing list only for the standardisation process inside OGC (vote for submission as a standard, etc.). The public group can deploy as many milestones as they wish, but official releases (I mean with "OGC standard" label) can be performed only by OGC. To date there is only one official release (GeoAPI 3.0.0) and many milestones.
> 
> GeoAPI development tends to be very slow, because GeoAPI mission is to create interfaces from OGC standards. The main debates are about how to interpret a specific element of a standard. Some additions do happen, but are expected to be rare. API not covered by standard stay project-specific, so SIS would have its own API as a complement of the GeoAPI one.
> 
> What are the opinions about SIS-GeoAPI relationship?
> 
>    Martin
> 


++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Chris Mattmann, Ph.D.
Senior Computer Scientist
NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
Office: 171-266B, Mailstop: 171-246
Email: chris.a.mattmann@nasa.gov
WWW:   http://sunset.usc.edu/~mattmann/
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Adjunct Assistant Professor, Computer Science Department
University of Southern California, Los Angeles, CA 90089 USA
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++