> -----Original Message-----
> From: WISSENBACH,DAVID (HP-Boise,ex1)
> ...
> I'd change the paragraph in B.2 to the following:
>> Producers and consumers of Application/Vnd.pwg-multiplexed
> entities, as defined in [MIMEMPX], should consider each
> component image message of the compound document as having
> one and only one reference. The producer of the compound
> document must assume that the consumer of the
> Application/Vnd.pwg-multiplexed entity has limited memory and
> therefore include a unique image message for each image
> reference found in the root document. If a ContentID is
> present in the header of an image message, that ContentID
> must be unique. If a Content-Location is present in the
> header of an image message, that Content-Location is required
> to be unique except for the special case where a repeated
> reference to the same image URL causes several messages
> containing of the exact same image data to be present in the
> compound document. Consumers may release the message data
> associated with an image reference as the image is rendered,
> because the Consumer can be confident that another reference
> to the same image will be accompanied by another message
> containing an copy of that same image data. Consumers may
> also substitute image data for a message with a given
> Content-Location header value with image data from other
> messages with the same Content-Location header value because
> Consumers can be confident that messages with identical
> Content-Location values do in fact contain identical data.
>> I suggest a further clarification by adding a following paragraph.
>> URL references in the root document of the multiplexed
> document must be matched to Content-Location and/or
> Content-ID fields of the referenced message object according
> to the rules given by RFC2557. An exception to the rules
> given by RFC2557 occurs when a reference is made to a message
> object named with a Content-Location. In that special case,
> multiple instances of that message are required in the
> compound document.
>
I like this wording and if there are no objects I will amend the XHTML-Print
spec to contain this text. I will wait before amending the document to give
participants a chance to comment.
Jim Bigelow,
Editor: XHTML-Print & CSS Print Profile
IEEE-ISTO, Printer Working Group
http://www.pwg.org/xhtml-print
Hewlett-Packard
208-396-2068
jim.bigelow at hp.com