Fw: XML Information Set Requirements, W3C Note 18-February-1999

[Oren Ben-Kiki]

Michael.Orr at Design-Intelligence.com wrote:
>I'd like to understand this better...
>
>Naive question 1: Ignoring the realities of engaging with the
>recommendation process, just focusing on understanding the nature of
>your concerns, what types of requirements, modes of thought, etc. would
>become more prominent if the objects-over-streams bias were suddenly
>removed?

Butting into the conversation :-) I can give one example. XSL is defined so
that a reasonable implementation needs to store the input document in an
"object" - some random access structure, while the output can be emitted as
a stream. I don't know if the designers thought of it quite this way, of
course. A different set of design choices would have led to an opposite
implementation - handling the input document as a stream. This would require
that parts of the output document be built as "objects", but not necessary
all of it.

Given all the talk about the relationship between XSL and XQL, and the real
possibility of using an XSL stylesheet to extract a small amount of
information from a large (virtual?) input document, this might prove to be
an expensive design decision. Of course, it has other merits - XSL
stylesheets are somewhat clearer as a result. It would be interesting to see
how XQL would address this issue.

Share & Enjoy,

    Oren Ben-Kiki


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)