> Interoperability considerations:
> The ability of implementations to handle all the defined
> applications (or profiles within applications) of TIFF may
> not be ubiquitous. As a result, implementations may decode
> and attempt to display the encoded TIFF image data only to
> determine that the image cannot be rendered. The presence
> of the application parameter may aid in allowing this
> determination before dispatching for rendering. However, it
> should be noted that the parameter value is not intended to
> convey levels of capabilities for a particular application.
I think this, too, alludes to the role of the TIFF 'application' parameter
that I raised in a previous message.
One of the topics we tried to address fully in
<draft-ietf-fax-feature-schema-xx.txt> was to be able to capture precise
TIFF capabilities. Specifically, the "variable" capabilities within each
mode were separated from the 'image-file-structure'; the former are
identified explcitly as resolution, compression, colour capabilities, etc.
while the latter is identified by reference to TIFF modes per RFC2301.