From: Aron Roberts (aron@socrates.berkeley.edu)
Date: Fri May 31 2002 - 15:29:56 PDT
In the message "Re: [Webnet] Program guides online", dated
2002-05-31, Rusty Wright wrote:
>One of the possibilities I'd like to see investigated is using xml.
> ... You then need to write an xslt file, a style sheet for an xml file,
>and a program that applies the xslt file to the xml file and produces
>html.
Expanding a bit on Rusty's idea, above, you might use a page layout
program such as Adobe's InDesign to generate a brochure for printing,
then export the content of that brochure in XML. You might then
apply XSLT stylesheets to transform the XML content to generate HTML
Web pages.
Another potential route to explore would be to create or generate
(programmatically) your content initially in XML, then apply XSLT
stylesheets to your XML data to both create HTML Web pages and to
generate PDF or (potentially) PostScript output for printing.
(To create PDF or PostScript from XML content, you would later in
the process also be using a related technology known as XSL-FO, which
is described in an introductory article at
<http://www.xml.com/pub/a/2001/01/17/xsl-fo/index.html>.)
Neither of these approaches are trivial, but they nonetheless
represent some promising techniques in pursuit of the holy grail of
publishing content from a single source to multiple targets ... such
as a Web page, a PDF document, and a printed, color brochure.
Along these lines, there are a large and growing set of tools for
writing or generating XML. Among page layout and technical document
creation tools, InDesign 2 (as mentioned by Rusty), Quark XPress 5
<http://www.quark.com/products/xpress/new_in_5.html#xml>, and
FrameMaker
<http://www.adobe.com/products/framemaker/keyfeature7.html> all
support XML output.
The just-released Oracle 9i Release 2 offers extensive support for
XML, and is another example of how quickly and pervasively XML
support is beginning to appear in key products. If you currently
maintain content for your publications and/or Web sites in Oracle 9i
databases, Oracle's XML DB guide, describing Oracle 9i's new
XML-related features, may be of interest in helping you determine how
to generate XML from your database content, or even to natively
maintain some of your database content in XML:
http://otn.oracle.com/tech/xml/xmldb/content.html
>Saxon is one program that was recommended to me, XMLSpy is
>another. It's not clear to me if XMLSpy does the translation since
>it's marketed as an xml editor.
Under Windows, you can apply stylesheets to translate XML content
to HTML or any other markup format by using the XSLT processor which
comes with Windows's MSXML library. (More on this below.) You can
also do this by installing a third-party XSLT processor for Windows,
such as SAXON (or the easier-to-install "Instant SAXON"), both
available from <http://saxon.sourceforge.net/>.
Microsoft has bundled MSXML -- which includes an XML parser as well
as an XSLT processor -- with recent versions of Internet Explorer,
and hence MSXML has also become a standard component of Windows
itself. The following Microsoft document identifies the bundled
versions of MSXML:
<http://support.microsoft.com/default.aspx?scid=kb;en-us;Q269238>.
XMLSpy, a slick, commercial integrated XML development environment
for Windows, does not include its own XSLT processor. Instead, it
will use MXSML by default, although you can also configure it to use
SAXON and can easily switch between the two, as well, so that you can
make sure your stylesheets are fully compatible with both XSLT
processors.
If you're using Windows, you get MSXML 'for free' and may as well
use it for parsing and transforming XML.
However, if you're instead working in a Unix-based environment
(including Linux and Mac OS X), the most commonly-used tools for
parsing XML content, and transforming it to HTML and other formats
via XSLT stylesheets, are probably Xerces and Xalan, respectively.
These tools are available at no cost from the Apache XML Project:
<http://xml.apache.org/>. There are many other free XML parsers and
XSLT transformation tools available for Unix, as well.
>Using this method all of the information can be preserved. The uphill
>parts are learning xml and learning and writing an xslt file.
True. XSLT is not 'immediately intuitive to even the most casual
observer', at least if my own, often-halting, initial experiences
with it are any indication ;-). Out of those frustrations was born a
list of some selected resources for learning about XSLT:
http://seaotter.berkeley.edu/xml/xslt-resources.html
In addition to the resources already mentioned on that page, I also
strongly recommend the following (and relatively new) book as a
terrific beginner's introduction to writing XSLT stylesheets,
especially for transforming XML content to HTML for the Web:
Bob DuCharme
XSLT Quickly
http://www.snee.com/bob/xsltquickly/index.html
It's also worth spotlighting one of the best first introductions to
XSLT, which is also mentioned on the learning resources page, above.
Examining the examples in this article is a good initial way to see
what XSLT is all about:
David Jacobs
"Rescuing XSLT from Niche Status: A Gentle Introduction to
XSLT through HTML Templates"
http://www.xfront.com/rescuing-xslt.html
Finally, support for directly displaying XML content (with XSLT or
CSS stylesheets applied) in Web browsers is just starting to appear
in the most recent versions of the Microsoft and Netscape/Mozilla Web
browsers. But to support users of older browsers, it will likely be
necessary for some time to come to first transform such content into
HTML for viewing.
Aron Roberts
Workstation Software Support Group
-----------------------------------------------------------------------
The following was automatically added to this message by the list server:
Webnet information is available at <URL:http://webnet.berkeley.edu/>.
This archive was generated by hypermail 2b29 : Fri May 31 2002 - 15:30:18 PDT