Ultimately, I guess, it depends on what you're using FOP for.

I see XSL-FO as a means to print files. Or store them in a printable format.
XSL-FO is not an interchange format; it is not PDF, nor should it be looked
at as such. It exists to be converted into something based on the semantics
defined in the XSL-FO specification.

For all their promises of WYSIWYG, neither Word nor OpenOffice (nor the
formats they read) are WYSIWYG. Their formats cannot exactly conform to the
XSL-FO specification. They can come close, but they're not the same.

What a person wanting to put documents in a Word-readable format wants is to
be able to take his original XML file and turn it into something that can be
used by someone else. In that case, what is necessary is an XSLT to convert
the XML into ODF or whatever.

In any case, RTF lacks the full degree of expressiveness possible from
XSL-FO anyway, so some information is going to get lost. What's the point of
doing the XSL-FO transform to begin with if you didn't want your stuff to
have a particular look to it? It's like wanting to use a broken CSS renderer
for your HTML; you may as well not have bothered with the CSS to begin with.

Plus, there's the fact that ODF attempts to retain some basic semantic
information, which has already been lost by the time the XSL-FO comes
around. Since this information cannot be recovered, the ODF file produced
from an XSL-FO would be, at best, visually functional. But ideas on what
constitutes a title or whatever are completely lost. One would get much more
functional ODF files by writing a custom XSLT for the format. Plus, the
format isn't exactly that difficult to use.

Mark C. Allman wrote:
> I only meant that if we replace the RTF capability with something
> equivalent.  I use the RTF generation so that I can then convert a few
> things to MS Word (can't ignore the 800 pound gorilla in the room).  I
> think "ODF" was in the post I replied to, so I just used that as an
> example.
> What I do now is the following:
> doc.xml --------------|
> doc.pdf
>                       |--> XSLT ENGINE --> doc.fo  --> FOP  --> {
> doc.rtf }
> translator_doc.xsl
> --|                                             .....
> Putting something into FOP to generate ODF wouldn't make much sense,
> IMHO.  I think it'd just be another xslt script to translate the FO file
> to ODF.  Or write a plug-in for OpenOffice to read in FO files
> (obviously another project!).
> I think we'll lose users if we don't keep something that lets them
> generate docs that are interoperable with the 800 pound gorilla.  How we
> do that is the question.
> Just my 2 cents....
> -- Mark C. Allman,
> -- Allman
> Professional
> Consulting, Inc.
> -- 617-947-4263
> --
> www.allmanpc.com
> BusinessMsg -- the
> secure, managed,
> Enterprise IM/IC
> solution.
> On Wed, 2007-08-01 at 11:29 +0200, Vincent Hennebert wrote:
>> Hi Mark,
>> Mark C. Allman a écrit :
>> > Drop RTF with nothing to replace it, e.g., ODF?  I'd rather not.
>> >
>> > Swap out RTF with, say, ODF?  Sounds great to me.
>> >
>> > I'd even volunteer some time to help with development.  I seem to
>> > remember something about Java.... ;-]
>> Thanks for your offer to help!
>> However... would that make sense to produce ODF from XSL-FO? There is no
>> semantic construction at all in FO, whereas there is some in ODF. The
>> other way around looks much more useful to me; as style informations are
>> stored using FO, this should be very easy to convert an ODF document
>> into plain XSL-FO.
>> Typically the transformation chain:
>>     XML —> XSL-FO —+—> RTF
>>                    |
>>                    +—> PDF
>> would be replaced by:
>>     XML —+—> XSL-FO —> PDF
>>          |
>>          +—> ODF
>> WDYT?
>> Vincent

View this message in context: 
Sent from the FOP - Dev mailing list archive at Nabble.com.

Reply via email to