stephen white wrote:
On 25/06/2007, at 7:20 AM, Frank Warmerdam wrote:
I will concede that for folks coming from a GIS background, your wish
seems
bizarre and not something I would ever think to explicitly support in
a format.
But I don't see this as a standards problem, so much as as a viewpoint
difference based on different approaches and backgrounds.
If I may be incredibly arrogant, I would say that the GIS crowd have the
real problem and they are the ones that need to update their approach.
Stephen,
Well, it might qualify as arrogant to think I'll just take your word
on this. I think you would have to provide a more detailed explanation
of what it is the GIS folks have wrong and why it is our problem to
resolve it.
The GIS crowd have been hammering at this problem for decades, and
they're not really getting anywhere. The formats and standards used to
represent data must accommodate the limitations of computers, so that
the ideas of representing reality can be realised despite the
imperfections of our tools.
If the problem you are referring to is data standards, then I think the
GIS community is doing adequately well at "the problem" for our purposes.
GIS data formats might not be very suitable for solving CAD problems,
or 3D visualization problems, but then for the most part I think those
are somewhat different domains and may, in some cases, need their own
tailored formats instead of assuming there can be one overall GIS+CAD+3DVIS
geo format that brings all the data together, and in the darkness binds
it.
I note that you're the president of "OSGeo", which is a name I vaguely
recognise.
It is the "Open Source Geospatial Foundation" - basically an organization
by which various open source geospatial projects, users and other
interested parties work to promote open source solutions. It is not
a standards development organization though we have taken a position
in favor of supporting and promoting appropriate open standards.
> Therefore I'm going to ask you... how are you going to update
GML so that it has the features required to be more useful to a larger
range of implementers, not just arranged in ways that GIS people would
like?
Frankly, I think GML is already too general for practical implementation.
I have also never taken it upon myself to make GML or other formats
sufficiently general to serve all possible uses. I try to pick a few
goals in life, that I think I can have a meaningful impact on, and avoid
getting dragged into too many where I'm unlikely to have a positive
impact.
So, my short answer is I do not choose to pursue this particular battle.
That said, I am active in some OGC activities, currently the "Web Coverage
Service Revision Working Group". Within the areas I select, I try to
encourage specifications that useful, not too complicated and practically
implementable.
The default option is what is happening now... people just don't bother
with the standards and hack up code that just does it... and then the
new standards come from the working code, and GML is history. VRML was
very big in its day, but it also failed to provide relevant access paths.
Lots of people do work around the standards and get get real value out of
them. I have lots of specific concerns about OGC standards, ISO standards
and the whole standards development process. But I think it is already
deliverying meaningful value.
In 1998, when I started work on the GDAL/OGR project (geospatial raster/vector
format translation and data access library), I made a deliberate decision to
try and align myself with OGC standards where practical. In fact this most
important aspects of this were explicit use of the OGC Simple Features geometry
model and use of OGC "Well Known Text" for coordinate system descriptions.
Both of these have paid substantial dividends in terms of improved
interoperability with other systems, and by virtue of GIS folks being able to
apply their understanding of the standards to my package after learning
them on others.
On the other hand, I don't define my project in terms of standards, and I
free go beyond them, or develop my own approaches where there is nothing
appropriate.
Best regards,
--
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up | Frank Warmerdam, [EMAIL PROTECTED]
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush | President OSGeo, http://osgeo.org
_______________________________________________
Geowanking mailing list
[email protected]
http://lists.burri.to/mailman/listinfo/geowanking
