Re: [xsl] Using xsl:output in browsers, was: Re [xsl] XHTML html validation

Subject: Re: [xsl] Using xsl:output in browsers, was: Re [xsl] XHTML html validation
From: "Manfred Staudinger" <manfred.staudinger@xxxxxxxxx>
Date: Tue, 20 Feb 2007 15:36:21 +0100
On 20/02/07, Robert Koberg <rob@xxxxxxxxxx> wrote:
On Tue, 20 Feb 2007 08:25:15 -0500, Abel Braaksma <>

> Perhaps it is worthwhile to create a little list with pros and cons for
> people that are new to the subject so that anyone can decide for his/her
> own situation. I understand that you know pretty much of this type of
> transformation invocation, maybe you can help improving a Wiki article
> on this subject (not yet there).

Not complete, I'm sure, but...


- no caching of the processor object
- when using XHR (XMLHttpRequest), the transform will not be invoked. You
will have to parse the DOM to find the PI and apply the transform from JS.


- simple
- works in almost every browser (just not in XHR)

You got the wrong starting point: using PI is _not_ an exclusive to JS initiated
transformations. In my view using PI can provide you with a rich html structure
for the starting point and should use css for the presentation. This part does
not even depend on JS. After that, you can use JS with or without XSLT to
provide the truly dynamic parts of your webpage.
XSLT when used via PI it provides an impotant _additional_ layer for
the webpage.
Its available in most modern browsers, it has no known security issues, it
is working independent of browser settings, it runs before any JS, it provides
screen-reader compatibility - this in _addition_ to the JS triggered
that are available afterwards for dynamic effects.
What I want to compare is:     PI + JS transformations vs. JS transformations

Best regards,

Current Thread