Re: another WebCGM spec glitch

[Lofton Henderson <lofton@rockynet.com>]

Thanks for the feedback, Dieter (and others, of course).  Taking your 
points a little bit out of order:

At 07:02 PM 5/25/01 +0200, Dieter Weidenbrueck wrote:
>Most of this is common sense behavior, or in other words, the behavior
>expected by the user. So far no customer has asked for any other behavior.
>Especially nobody has asked for isotropic scaling to "fill" the viewer
>rectangle completely.

Right.  Common sense works fine, until you read the exact words and try to 
write Test Cases that are defensible under the wording of the standard!

>Still the wording should be consistent throughout the spec.

Exactly.

>[...]
>XHTML requires lowercase for all tags these days, so I think it would make
>sense to go this way with our "tags". If you think about an external XML
>companion file with linkuri tags etc it makes even more sense.

Agreed.  We are still all unanimous, then.

>[...]
>With regard to the scaling question, I can say that IsoView's behavior is to
>"shrink-to-fit", as is defined in 3.4 as the default behavior for the OBJECT
>tag. We have an "resizeToFit" property that allows for switching to metric
>mode. However, this will not be used in the future since it caused more
>confusion than anything else.

Good, everyone agrees on this as well.  As you point out, it is specified 
for OBJECT for in-line WebCGM content, but is never clearly specified for 
hyperlinked content in frameset environment.  We'll clarify wording for 
Second Release.

>All target rects a.k.a. view_context or object hulls are scaled to fit into
>the viewer rectangle. This is how we understand the spec: show the area
>covered by the view_context in the viewer rectangle.

Right, this part is pretty unambiguous as it stands.

Thanks,
-Lofton.

*******************
Lofton Henderson
1919 Fourteenth St., #604
Boulder, CO   80302

Phone:  303-449-8728
Email:  lofton@rockynet.com
*******************