Re: [xsl] [XSL] XSL Browser Integration

Subject: Re: [xsl] [XSL] XSL Browser Integration
From: "M. David Peterson" <m.david@xxxxxxxxxxxxx>
Date: Sun, 16 Sep 2007 11:54:27 -0600
On Sun, 16 Sep 2007 03:15:11 -0600, Alain <alainb06@xxxxxxx> wrote:

(unreliable on client-side)

If anyone uses the "unreliable on client-side" argument, question their motives.


Some facts,

* Of *ALL* of the client-side standards, XSLT is by far and beyond the most reliable when it comes to cross-browser support.
* Cross-browser CSS and Javascript support is spotty.
* CSS and Javascript debugging tools are of poor quality or non-existent.
* XSLT debugging tools are of high quality and are plentiful (non browser-based tools)
* Using a single *cross-browser* supported command you can instantiate an entire transformation process.
* Of all four major browser vendors (MSFT, Mozilla, Apple, Opera) only one of those manafacturers has an incomplete XSLT 1.0 implementation.
* The single vendor w/o a complete implementation (lack of document function support) is Opera which represents less than 1% browser market share.
* PI-based client-side transformations encourages proper REST-based development practices.
* Using a coombination of PI-based transforms, the document function, data compression, and built-in browser-based caching that comes for free you can reduce the amount of data sent with each new page to less than 10% of your typical pre-rendered HTML page.
* Client-side XSLT encourages code re-use.
* XSLT in general encourages cross-organization best practices adherence.
* Client-side XSLT and server-side XSLT are not mutually exclusive, or in other words it's not a choice to use one or other but not both.
* Using them together creates an *extremely* powerful combination, as does adding XQuery to the mix. This becomes especially powerful when you use Saxon and/or Saxon on .NET as your server-side processor as it provide XSLT 2.0, XPath 2.0, and XQuery 1.0 support as part of the same package/assembly.
* There are more reasons. I'd be happy to share them w/ you if you'd like.


* There are some gotchas.
* Abel Braaksma has become the resident expert of what these gotchas are as they relate to each of the four major browsers.
* I'd bet if you asked nicely he would either point you to the various related threads from months and years past and/or summarize them for you in-line to this thread. ;-)


--
/M:D

M. David Peterson
http://mdavid.name | http://www.oreillynet.com/pub/au/2354 | http://dev.aol.com/blog/3155


Current Thread