Ratification of ESRI restful web service as an open standard

Posted on: May 9, 2013 - 19:26 By: pcreso

In the global GIS, spatial data & web mapping arena ESRI plays a similar role as the dominant global player to Microsoft, but as with all analogies, don't treat this as 100% gospel. There are many players out there, in the FOSS arena and others.

The Open Geospatial Consortium is a standards developing body in the spatial software domain, generally supporting/facilitating interoperability.

ESRI has a proposal to have a slightly modified version of its proprietary standard endorsed as an OGC standard. There is a relationship between Open Source and Open Standards, which is well described here: https://portal.opengeospatial.org/files/?artifact_id=45126

The proposed "open" standard has to be backwards compatible with ESRI's proprietary RESTful API, so there is little about it that is genuinely open. ESRI is free to release the standard in the public domain, without OGC ratification and without dividing the global GIS community. The global industry could vote with their feet, and if global uptake warrants ratification as an open standard, then would be the time to apply.

ESRI has been emailing voting members of the OGC with a list of reasons to vote for the proposed standard. They have now stated that the contents of such emails will NOT be made public.

My comment from a mailing list:

Personal opinion in a nutshell:

From an OGC perspective: (yes I'm a member via NIWA's affiliation - once I pay the current bill :-)

This is not a genuine attempt to improve interoperability & support open standards, it is an attempt to undermine open standards & replace existing open standards which are widely used & supported in the community by ratifying standards currently used by one commercial vendor.

I'm currently responsible for implementing federated (interoperable) systems between research agencies & central/regional govt based on OGC standards. If these standards provide for disparate ways of doing the same thing, then "OGC standards compliant" will mean one of two things:
- may or may not work together depending on which OGC standards are supported
- everyone has to support both standards to be fully OGC compliant.

My organisation has just been quoted a few 10's of thousands of $ to develop a CSW service for ESRI datstores because, despite their claimed support for these standards - it is pretty minimal & of limited functionality & use.

At present we can build federated, interoperable systems including CSW, SOS, WMS, WCS, WFS, etc, and if an agency fails to interoperate, that is their problem. This change would fundamentally reduce, if not destroy, the value of OGC standards to the wider community, as OGC compliance will not indicate any real measure of interoperability.


From an OSGEO perspective (I'm in the Australia/NZ chapter):

This weakens the FOSS community & strengthens ESRI's place in the global GIS community. OSGEO is there (IMHO) to support FOSS GIS. Agreeing to a change which gives a commercial competitor a strategic advantage
- giving them a mandate to let the FOSS community play catchup - is NOT in the best interests of the FOSS community.


From a FOSS perspective (I'm a council member of the NZOSS):

This is pretty much a repeat of Microsoft's refusal to support an existing, community based XML document/file standard & their forcing of a competing standard on the community, which has been of no value to the user community & created problems for the FOSS community.

We should learn from that fiasco & not make the same mistake again, as much as it in our power to prevent it.

Some members of the Open Geospatial Foundation have started a common letter opposing the ratification. This foundation is the global voice politik of FOSS in the geospatial arena (plus more). The letter, including comments and explanations about the concerns, is available here:

http://wiki.osgeo.org/wiki/Geoservices_REST_API

 

Brent Wood