[
Lists Home |
Date Index |
Thread Index
]
Bryan writes:
> I would normally agree about this but I've recently been working on
> something that will be released in the next month or so, that while
> it has the possibility of having processing completely separate from
> the document also has some parts of the processing definable inline.
> I implemented the inline processing because of a definite lack in the
> system, other solutions would no doubt have been possible because I
> believe there is always another solution, but the environmental
> conditions expected for the project, the technical level of users,
> and demands on the project's capacity meant that some sort
> user-defined inline processing had to be allowed. That said although
> the project does allow for inline processing there is certainly
> nothing "intrinsic to the xml which binds it to particular
> processing".
>
> To me, any statement that something should not be done(even though it
> is well within technical capabilities to do it) smacks of
> religiosity. I am of course devout, but not above sneaking behind the
> pews for a quick bit of inline processing if I feel it absolutely
> necessary to the goal at hand.
If it works in your own process, and you don't plan to make the rest of
the world use it, it's fine by me. I mostly get upset when those
controlling the "XML Core" inject such understandings into the
specifications at the heart of XML work today.
> As a philosophical thing, what kinds of embedded markup do you often
> see in documents that YOU would consider better separated from the
> document?
Any place where there's the potential of multiple overlapping
hierarchies applied to the same information is an excellent candidate
for such separation. There are also situations where the annotations
provided by markup aren't all that well-suited to support for complex
(or worse, changing) metadata.
--
Simon St.Laurent
Ring around the content, a pocket full of brackets
Errors, errors, all fall down!
http://simonstl.com
|
