[Geowanking] Critiques of wfs, wms
crschmidt at metacarta.com
Sun Jan 24 06:55:45 PST 2010
On Sun, Jan 24, 2010 at 12:55:51PM +0100, SteveC wrote:
> On Jan 22, 2010, at 8:06 PM, R E Sieber wrote:
> > Hi Everyone,
> > Can someone point me to the definitive critiques of wfs and wms?
> The spec is 184 pages long for WFS. So rather than read it, OSM was born.
But you used HTTP, which has a 114 page spec? Why didn't you reinvent
that while you were at it?
To be more serious, WFS has some good points, but overall, it's overly
complicated for a lot of use cases. The primary (and required?!) transport
being GML is probably one of them. Complete lack of examples of how to
interact with it, the complexity of write support, etc. all add to it.
That said, for managing read-write of simple-features to a server,
there isn't anything else that I'm aware of, so whether it's good or
bad, it is what people use. (OSM doesn't use anything resembling
Simple Features, so that isn't a counter-point; it is solving an
interesting, but different, problem.)
More information about the Geowanking