[
Lists Home |
Date Index |
Thread Index
]
jonathan@openhealth.org (Jonathan Borden) writes:
>I am encoding application specific information in XML documents. A
>bunch of people have gotten together and decided that such application
>specific information _ought not_ be encoded on the basis of attribute
>order (i.e. SAX doesn't tell the application about this). Why is this
>perverse?
It's perverse because they've taken an extra step to throw away
information that is useful in a wide variety of information contexts,
and have done it at the level of the XML spec - thereby ensuring that
this information would be lost pretty much across the board.
>The whole reason that I am interested in using an API is that this can
>help me extract the information in the document that I am interested
>in. Why should I be interested in whether an element is encoded as:
>
><foo bar="1"></foo> vs. <foo bar='1' </foo> vs. <foo bar="1" /> vs
>.... (and on and on)
>
>when I just want to represent an element "foo" with an attribute "bar"
>having a value "1".
You needn't be interested yourself, nor need you be interested in why
others would care. You just need to be aware that there are times when
throwing away this information causes real irritation, and be prepared
for that.
I'm hoping that future generations of XML APIs will at least preserve
such information if not use it, thereby reducing the amount of conflict
between those who just want the abstractions and those who actually
care what the markup looks like.
At present, those who care about the abstractions and the abstractions
only are far too greatly in the ascendant for me to be comfortable.
>Call this 'abstraction' but that's the essense of XML -- I'd dare say
>that XML wouldn't have become so popular if I'd _needed_ to worry
>about such details.
I don't find abstraction to be the essence of XML. I find XML to be a
text-based format with labels inside - and I do suspect that foundation
is what makes it popular, not the damn Infoset.
--
Simon St.Laurent
Ring around the content, a pocket full of brackets
Errors, errors, all fall down!
http://simonstl.com -- http://monasticxml.org
|