Re: Nostradamus (was Re: FO. lists as tables)

[pandeng@xxxxxxxxxxxx (Steve Schafer)]

On Fri, 15 Oct 1999 15:45:40 +0100 (BST), you wrote:

>I think that one of the main points about the dsssl/xsl-fo kind of
>abstraction is that it allows a formatter neutral specification of
>style. If you were to enforce things down to the level of line breaking
>then essentially you would only ever have one formatter available.

Doesn't that go hand-in-hand with the idea of portability?

>(You could build another one, but why bother if it was forced to give
>the same results.)

Because you might be able to build one that works better (faster,
smaller, more portable, etc.). Also, conformance to a standard doesn't
mean that you can't include features that aren't addressed by that
standard. It's easy to envision two word processors that use exactly
the same file format, exactly the same line-breaking algorithms, etc.,
yet have completely different user interfaces.

>For example TeX isn't a bad formatter, but it isn't perfect, but if the
>algorithm specified for XSL-FO wasn't _exactly_ TeX's (for instance, if
>it had improvements like being able to limit runs of consecutive
>hyphens, or to hyphenate the first word of a paragraph, or to be able to
>dynamically change the shape of a paragraph depending on insertions or
>page breaks or...) then it would be impossible to use TeX as a back end
>for XSL.

Why is that a bad thing? If we're not allowed to improve upon existing
implementations, then why not just use those implementations as the
standard(s)?

>it seems the only way to ever get a standard that specifies things down
>to that level would be to not use any existing formatters and only allow
>(new, and a possibly unique) formatter that was written to the spec.

Again, I don't see why that's such a bad thing. Sure, it means that
existing code bases are going to have to be tweaked in order to
conform to the specification, but I don't care about the angst of
programmers (myself included). I care about the angst of all of the
people who actually have to _use_ this stuff.

-Steve


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