[OSGeo-Discuss] make software; Let the OGC do the standards side of it

Jody Garnett jgarnett at refractions.net
Wed Jul 18 08:53:27 PDT 2007


I am rather tempted by this topic - I would love to join a committee on 
this topic if it is created.

However I don't think it should be created - the problem of how the 
standards process (ISO and OGC) can benefit from free and open source 
geospatial software is something for the standards bodies to work out.  
We can facilitate this as a point of contact; but that is probably where 
we should stop.

I repeat let them come to us - in many cases this already happens.

The OGC has included some open source software in its Open Web Services 
initiatives - GeoSever was used as a reference implementation against 
the CITE WFS 1.1 tests for example. This is a sensible approach to 
making the WFS 1.1 specification actually useful; in addition to a 
document, conformance test engine, they now have some source code people 
with questions can download to see how it is done.  Good on them; their 
standard is all the better for it (and frankly more believable as a 
standard).

We should stick with our mandate; making great open source software; if 
you want to join the standards police there are a couple of options. Pay 
to play is the one causing grief here, you can also review their 
documents as part of their public review period, and finally you can 
hunt an open source project that is implementing that standard and 
contribute in code. The last option is a bit weak; I contribute code to 
the GeoAPI project, and was able to contribute to a working group report 
last month, but as I am not an OGC member it fell upon others to attend 
the meeting in Paris and make the presentation.

The borderline case here is "Tile Map Service". When the occasion merits 
we do like to work together. We can define shared assumptions, 
conventions or in the case of "Tiled Map Service" additional metadata in 
order to collaborate more effectively. If some of our members are also 
OGC members they can trot these ideas out to a working group; having 
several implementations to point to for working examples should make 
that a fairly efficient process.

In the past I was able to get into the discussion on "Open Web Context" 
document (think Web Map Context document but for WFS, WMS and WCS). It 
sounds like the OGC has had a change of policy and I will now need to 
pay $400 for this ability?  For most of my feedback they will get it 
anyways - OGC members take part in the same open source projects I do - 
this will be more of an example of both organizations leaning on these 
individuals. At worst the OGC will just be inefficient - they release 
documents for public review (and if they are silly or unimplementable we 
can always just laugh - think of GML3 where the schema was not valid).

I understand that the OGC has an arrangement with ISO where they have a 
well defined mechanism for sharing ideas (cross publication or some 
such). If we do feel strongly that our involvement is needed we could 
ask our board to pursue talks with the OGC to set up something similar. 
For projects that interact with the standards process already they will 
have members in both organizations already.

Cheers,
Jody
Lorenzo Becchi wrote:
> I guess this thread is pretty hot, there are many sub-threads and 
> maybe a good solution could be to set up a mailing list, for the 
> beginning, as: ogc AT osgeo.org
> Trying to be practical I can offer myself to administer the list as 
> I'm doing with the Spanish Chapter and the Spanish GIS Book.
>
> There  will be the possibility to define possible actions, 
> participation to OGC meeting, support to OSGeo new standards (as for 
> Tile Map Service), creating or not a Committee ecc ecc.
> I guess we can, as minimum target, set up a lobby of OSGeo softwares 
> to promote effective interoperability of OGC standards.
>
> ciao
> Lorenzo
>
> _______________________________________________
> Discuss mailing list
> Discuss at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/discuss




More information about the Discuss mailing list