[
Lists Home |
Date Index |
Thread Index
]
4/26/2002 2:41:04 PM, "Simon St.Laurent" <simonstl@simonstl.com> wrote:
>I'm not expecting the entire world to use XML syntax - it's not ideal
>for everything - but when it's time to present a canonical and
>processable view of information, it's glorious stuff.
I presume that most of us agree in principle. There are a few
smokestacks spoiling the glorious view, however:
1- Not all syntax is canonical; you need something like the
InfoSet to say that single/double quotes don't matter, etc..
2 - Some syntax is sugar, but sugar is tasty: CDATA sections come to
mind; they're nice for some things (escaping scripts being the
canonical example) but they'll disappear from your syntax if you run
it through XSLT.
3 - There's a lot of inefficiency in that syntax, either for humans
to produce, or to stuff down narrow bandwidths to small devices, or
whatever.
4 - For better or worse, many recent specs are not based on the
syntax -- XPath/XQuery, SOAP, DOM (of course). This produces
an impedence mismatch.
So, how do you deal with all this and still enjoy the view?
|
