-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Andrew Turner wrote: > R E Sieber wrote: >> Hi Everyone, >> >> Can someone point me to the definitive critiques of wfs and wms? >> > Don't know of "Definitive", or what that would mean (OGC supplied?). > Here is a suggestive critique. > > - Documentation isn't approachable: click through license to long Word > document that lacks any examples. > - No concept of cached or 'offline' support. It's a *service* that > doesn't have a mechanism for archival > - Lack of good mime-type support (a GML mime-type is emerging I think) - > this prevents smart development of serving & handling requests. > - Don't handle very large datasets well, no pagination or compression > > WFS and WMS were designed not following successful best practice web > principles, and now unfortunately is harnessed with so much momentum > (vendor, industry, 'architecture enforcement') that it will be difficult > to resolve. > > Andrew
Renee, some more cents, not "definitive" but from personal experience. WMS has been designed prior to wide acceptance of REST principles (when would that be?) It would be fairly simple to make WMS more RESTful [1]. REST is on the agenda of the OGC, currently they are trying to separate bindings from actual interface standards. Painful work, especially in this broad environment of interests. One uptake of WMS is that it has been the lever to pry loads of public administration data loose from their respective silos. WMS is good for dynamic data, supports different coordinate reference systems for one data source and is not bound to a proprietary scale level system. This makes it much more interoperable than anything else that serves maps. Both cannot be accomplished as easily with cached tiles. Hybrids like GeoWebCache try to bridge that gap. WFS is a pita to implement. But regardless of this fact it does work in large scale organizations [2] and is supported by a fair number of software packages. One barrier to WFS (as was also less so the case with WMS) is that large proprietary vendors have little to no interest in opening up their carefully maintained vendor-lock-in strategy. This is also why it sometimes pains me to see highly respected and articulate folks bad-mouth standards in general. It shows a lack of wide sight. In general I think it is great thing to have the OGC maintain and push those standards out there. Some Open Source projects deem WMS, WFS, WKT, WKB (and to a lesser degree GML, OpenLS, etc.) so relevant that they design their architecture around these standards (deegree, GeoTools, GeoServer, GeoNetwork, Mapbender, PostGIS, etc.) or at least implement a fair number of these standards (OpenLayers, MapFish, MapGuide, OSSIM, GRASS, GDAL/OGR etc.). All larger proprietary software has to implement WMS if they want to get anywhere at all in the public administration [3]. Regards, Arnulf. [1] http://www.slideshare.net/arnulfchristl/ogc-capabilities-documents-in-the-roa [2] http://www.mapbender.org/FIONA_-_Online_Agricultural_Subsidy_Application_System [3] http://www.opengeospatial.org/standards/ >> Thanks, >> Renee >> >> _______________________________________________ >> Geowanking mailing list >> [email protected] >> http://geowanking.org/mailman/listinfo/geowanking_geowanking.org > > > > _______________________________________________ > Geowanking mailing list > [email protected] > http://geowanking.org/mailman/listinfo/geowanking_geowanking.org - -- Arnulf Christl Exploring Space, Time and Mind http://arnulf.us -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAktddFIACgkQXmFKW+BJ1b3MCgCdEg0HWHHKhyXvbKfhNNuYTegJ +wcAn3cIZpaaxH2gckU3apxRQHoVtQdR =FWc+ -----END PGP SIGNATURE----- _______________________________________________ Geowanking mailing list [email protected] http://geowanking.org/mailman/listinfo/geowanking_geowanking.org
