[
Lists Home |
Date Index |
Thread Index
]
- From: "Michael Champion" <Mike.Champion@softwareag-usa.com>
- To: <xml-dev@xml.org>
- Date: Fri, 7 Apr 2000 17:03:10 -0400
----- Original Message -----
From: "THOMAS PASSIN" <tpassin@idsonline.com>
To: <xml-dev@xml.org>
Sent: Friday, April 07, 2000 8:57 AM
Subject: Re: Feature Manifest (Was:RE: Parser Behaviour (serious))
> Michael Champion observed -
> > That's right, PIs show up in legacy browsers. Seems like a
theoretically sound reason to forbid them to me ;~)
> >
> I don't see it. The only legacy browser so far that deals with xml is
> Internet Explorer 5
The ";~)" denoted a feeble attempt at irony. I should just say what I mean,
I guess ... So since I put my foot in it, I guess I should try to explain
...
My *understanding* of why PI's are non grata at the W3C is because XML
metadata (or data that a script will operate on, or whatever) *is* useful
inside HTML even if it is not displayed. In the use case where
a) XML is embedded in HTML or XHTML on a Web page and
b) the page is viewed in a pre-Level 4 NS or MS browser ...
then the element content shows up as plain text, attributes are invisible,
and PIs show up as PIs in all their ugliness. I haven't tried the
experiment of actually checking this out for old browsers, but in IE and NS
4 the element text, but not the tags, show up but the comments and PIs are
invisible. But if this is more or less correct for "legacy" browsers, it
could explain the tendency to encode out-of-band information such as
namespace declarations in invisible attributes rather than intrusively
visible PIs.
This was all a snotty aside ... what I'd really like to see is a discussion
of the relative merits of a Feature Manifest coded in a PI in an instance,
in elements/attributes in a wrapper/package, and in the schema.
***************************************************************************
This is xml-dev, the mailing list for XML developers.
To unsubscribe, mailto:majordomo@xml.org&BODY=unsubscribe%20xml-dev
List archives are available at http://xml.org/archives/xml-dev/
***************************************************************************
|
