xml-dev - Re: Future of Formatting Objects (XSL/FO)

[Peter Murray-Rust <peter@ursus.demon.co.uk>]

I would like to support the XSL-FO and  FOP efforts - encourage the current
coders to hang in there - and to do what little I can to help. I'll say why
below.

At 11:04 AM 6/9/00 +0100, Sebastian Rahtz wrote:
>Thierry Hanser writes:
> > As I am not expert in page-media publishing, I am
> > not able to evaluate the current power of FO and its
> > fitness with the wide field of paper-like publishing
> > It seems to me that the FO specifications are pretty
> > well designed (at last a humanly readable rendering model)
> > and quite comprehensive but I might be wrong.
>
>You express my feelings exactly. "pretty well designed" and "quite
>comprehensive". It is a promising start, but until
>
> - the spec is finished
> - we see some chance of ongoing commitment to updating it
> - we have a reasonably complete implementation
>
>I personally would not advise my pension fund to invest in it...

I agree with both views. However we have been through these iterations
before on "XML" and XSL-FO and FOP are following previously successful
models. They deserve to succeed.

Potted history:
	In late 1996 the "XML" effort consisted of three parts:
	- XML syntax
	- XML link
	- XML style
The aggressive timescale had an XML-LINK spec due for (I think) mid 1997
and XML style for late that year. The Link did arrive, but then "matured"
for several years until recently being revitalized. Originally XML Style
was seen as a monolith (cf DSSSL) but then was released based on:
	- splitting the operation into two: transformations and formatting
	- transformations based on a language and syntax completely different from
either DSSSL [or the current XSLT!]


[I needed XSLT so much I actually implemented a lot of this release before
it changed :-)]

I have always assumed that the "FO" activity (whatever it might be called)
was an essential part of the mainstream vision of XML. IOW XML has to have
a mechanism of rendering documents for high-quality presentation -
primarily print, but possibly also screen and other media.
>
>What worries me is that XSL FO's parents are DSSSL and CSS. Not a
>very healthy start. DSSSL was never accepted by any major player, or
>implemented to the extent one could do top-quality typesetting with it;

I take the reverse view. I am *glad* that there has been previous
experience! Fred Brooks:

	"plan to throw the first one away - you will anyway" (The MMM).

>and CSS, well what can one say, it's like one's more embarassing
>relatives, those you hope keep quiet when the boss comes to dinner. Not
>exactly unimplemented, but implemented so badly and so incompletely it
>leaves a nasty taste in the mouth.

There is a distinction between bad design and bad implementation. I
personally welcome CSS design - there may well be flaws, but I don't think
they are responsible for the poor implementations. If the implementations
are poor it is a cultural thing. Customers don't care enough to insist that
the spec id properly implemented and manufacturers don't see this as a
selling point (I assume). This simply highlights the very low public
understanding of the value and place of style. In fact I give
demonstrations of  CSS as a good example of how content and style *can* be
separated - most users don't even know that CSS exists and  they *can*
style pages

>
In support of FOP
=================

Axioms:
	1. Print is and will remain a critical technology for the foreseeable
future. [XML will probably increase the demand for high quality print!]

	2. Typesetting is intrinsically complex, and in some cases difficult. 

	3. There is no likely communal solution for XML other than XSL-FO 

The gold-standard for many of us is TeX. Like many scientists I have used
TeX/LaTeX extensively and with some success, but do not find it trivial
even after practice. TeX was written by a genius and it is one of the very
few examples of single-author top quality robust software that has survived
for decades. These features have allowed it to accrete a virtual community
which continues to support and develop it and to generate industrial
applications.

If we want high-quality print I can only think of the following:
	- Develop XSL-FO and one or more FOP-like solutions
	- Use TeX as an intermediate
	- Tear up the current XSL-FO and try again

Any of these are potentially acceptable. I prefer the first, but would
settle for the second. [This would require an abstract model of TeX as a
DTD/Schema to which any XML document could be transformed by XSLT. It is
conceivable, but not trivial.] The third is only acceptable if there are
real problems with XSL-FO that can be rectified. I am not a typographer,
but I would be surprised if the current WG hadn't learnt a lot from DSSSL,
etc.

If we do not do this we are faced with:
	- fudge, kludge, and botch. This is the most likely future. I regard it
with dread. It will consist of O(2) glue packages converting between
different bits of different manufacturers offerings.
	- "XML Word". Everyone  will be encouraged to use Word as not only the
authoring, but also typesetting tool. For example, I will have to write
plugins that allow CML to be dragon-dropped into Word. Arghhh! 
	- "Hit the print button on the browser". High-quality?...

So. I cannot see a responsible alternatively to XSL-FO, and we *have* to
have activities like FOP to drive it.

I am happy with an evolutionary approach. At present I use of would like to
use FOP for:
	- rendering chemistry into reports, etc. This does not need every XSL-FO
option implemented. At present I am happy with proof-of-concept
	- rendering HTML to paper. The "Print" button has limited value - it is
not page-aware and on some browsers it mangles some things. When I do
presentations I do everything in XML. [Most others use Powerpoint :-(, but
I am usually promoting XML so I make a virtue of it.] So I have to be able
to print. I now have a system which prepares all my XML slides both for
screen display and will output something sufficiently good for printed
output for those who have to have paper to take away. Because I use SVG for
the graphics in my talk, I need to be able to print SVG *in an XML
context*. FOP is starting to provide this.

To give heart to those involved - SVG made it spectacularly. Look back
through the archives  - even two years ago and you will see gloom about
every coming out with a communal spec for vector graphics. We now have
*many* high quality implementations. 2-D graphics is also non-trivial - I
have spent many years in that area - and I think SVG is a top class spec. I
can see no reason why XSL-FO cannot follow in this tradition - the
principles are known and the value should be clear.

	P.


***************************************************************************
This is xml-dev, the mailing list for XML developers.
To unsubscribe, mailto:majordomo@xml.org&BODY=unsubscribe%20xml-dev
List archives are available at http://xml.org/archives/xml-dev/
***************************************************************************