Re: What is PI data exactly?

Tim Bray (tbray@textuality.com)
Fri, 16 Oct 1998 07:45:51 -0700


At 09:56 AM 10/16/98 +0100, Michael Kay wrote:
> Actually I think the notion of "data" in a PI is a SAX notion, not an
>XML one.

Right, the spec just defines the "Target" part.

>Meta-comment: we're suddenly getting a spate of comments that point to
>little inconsistencies or ambiguities in the spec, reflecting the fact that
>it is written very informally.

Yes, we are finding little problems. No, this is not a result of
writing it informally, it's a result of the committee not having
infallible predictive powers. While W3C rules prevent me giving
details, I can say that the long-missing XML 1.0 errata document will
soon come to exist, and patch some of the typos and thinkos. The W3C
is also, in a separate effort, addressing the question of what exactly is
the data and what isn't.

>I think the problems are particularly acute
>when the spec tries to talk about the actions of "XML processors" or
>"applications" because these notions are very fuzzy.

That is not true at all. The processor and the application are cleanly
defined. The fuzzy spots are in the area of what exactly the processor
passes to the application.
-Tim