[XSL-LIST Mailing List Archive Home] [By Thread] [By Date]

Re: Transformation + FOs makes abuse easy

Subject: Re: Transformation + FOs makes abuse easy
From: "Chuck White" <chuck@xxxxxxxxxxxxxxxxxxxxxx>
Date: Wed, 28 Apr 1999 12:20:10 -0700

Simon St.Laurent wrote:

>Extensible Style Language similarly focuses on external style sheets, but
>these style sheets tell programs how to transform documents from one
>vocabulary to another.  This transformation process provides powerful tools
>for converting and rearranging documents.  In addition to defining the
>transformation vocabulary, XSLT, XSL also defines a presentation
>vocabulary, called formatting objects (FOs).  Formatting objects contain no
>semantic information, apart from the conventions designers have
>traditionally used to convey information to human readers.
>While the end result of both of these style tools may in fact look
>identical within a browser frame or on paper, the underlying information is
>quite different.  In the CSS version, the original semantics are always
>available to the recipient, and the information can be easily reused in
>other processors.  In the XSL version, the final product is useful only to
>a human reader after the transformation has taken place.

I'd like to thank Simon for summing up the argument of those who are opposed
to FOs. I'm wondering if the anti-FO forces could clarify a couple more
parts of the argument. I'm not a programmer, so I really don't understand a
couple of issues. I am a graphic designer, so these issues are important to
me, as is the outcome of the final spec.

1) How are XML semantics lost if the original XML source document is
ultimately delivered? Maybe I don't understand what is even meant by
semantics, but my impression is that there is concern that the data
represented by the original XML element markup will be lost using FOs. If
that's the argument, I just would like to know how.

It seems to me that the formatting vocabulary could be just about anything,
as long as the source document is always available.

If I am not understanding what is meant by semantics, perhaps someone could
define it for me.If this sounds like an argument, it's not, it's a real
question that I'm hoping someone can explain.

2) How are the current FO's structured differently than the CSS objects
described in Håkon's W3 note (http://www.w3.org/TR/NOTE-XSL-and-CSS)? To a
non-programmer like me, the following (from Håkon's note) looks similar to
the FOs that he and Simon oppose:

<template match="/">
  <css:page size="landscape"
            margin="1.5in 1in"
  <css:page name="left"
  <css:page name="rotated"

To a non-programmer like myself, the architecture looks similar.

For reference, Håkon's paper, which launched this whole debate, is located
at: http://www.operasoftware.com/people/howcome/1999/foch.html.

Thanks for any enlightenment.
Chuck White
Creative Director
Advance Recruitment Advertising, Inc.
visit our online job site:

 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list

Current Thread