[
Lists Home |
Date Index |
Thread Index
]
> > As for the decision, it was all over in about 45 seconds on one
> > teleconference, I remember it clearly. Fairly early on in the XML
> > design process. Someone (maybe me? I forget) said "er,
> should we do an
> > API as well?" and James Clark said "isn't the idea that
> there's going
> > to be one API that's going to work for all the different things you
> > want to do kind of silly?" and everyone said "oh, right"
> and that was
> > that.
>
> But there's a big difference between a data model and an API
> onto a data model.
Yes. And they obviously had a data model in their heads: elements can
contain elements, elements can own attributes, elements and attributes
have names, IDREF attributes can reference ID attributes, etc. The
problem was that they didn't articulate it, so the rest of the world had
to guess what was in their heads, and decide, for example, that the
number of spaces before an equals sign was not a property in the data
model but an accident of the input format.
We've just had a discussion as to whether it is legitimate to use a
freestanding namespace declaration, that is never referred to, as an
information-bearing feature of a document. Everyone seems to think
that's not the way namespaces are intended to be used, but without a
data model, no-one can say for sure that they are wrong to use it, or
that a query language is right to ignore it.
Michael Kay
|
