MS-RFC-35 - Solving the WMS STYLES parameter issue

[Kralidis,Tom [Burlington]]

> Bart van den Eijnden wrote:
> > Tom,
> > 
> > I was thinking the same thing last week.
> > 
> > Why introduce a mechanism which can be misused in the future, if it 
> > can be as simple as making styles optional again.
> > 
> > Best regards,
> > Bart
> > 
> > On Nov 14, 2007 1:26 PM, Kralidis,Tom [Burlington] 
> > <Tom.Kralidis at ec.gc.ca> wrote:
> >> Hi,
> >>
> >> One final thought here :) would it be easier to revert to making 
> >> STYLES an optional parameters rather than introduce a new user 
> >> directive, given that STYLES= will be optional in the future?
> 
> Tom,
> 
> Well, we only wrote this whole RFC because we were under the 
> impression you want to be able to enforce the standard to the 
> letter.  If you are happy to just make it optional again, 
> then I for one would be happy to set aside the RFC and just 
> do that.  But I think Daniel made the motion so it would be 
> up to him to retract it.
> 

This is correct.  However I did not anticipate such a backlash based on
one or two WMS clients.

Note that this only applies to STYLES.  Ticket 1088 additionally made
FORMAT, SRS, BBOX, WIDTH and HEIGHT required.  I do not think we should
revert those.  I also think that our WMS Client support (which now sends
STYLES=) should not be reverted.

> But I do feel like I was put on a treadmill somewhat needlessly.

I think this is a very valuable lesson in interoperability issues and
balancing between what a standard says and what is implemented/used.

I still feel the logical solution is for the standard-offending
component to update their software to spec.

..Tom