Re: [Json] Nudging the English-language vs. formalisms discussion forward

[John Cowan <[e-mail]>]

Phillip Hallam-Baker scripsit:

> That is because XML markup requires the schema to be specified in the
> document

Not so, unless you mean DTDs.

> because the extensibility scheme depends on it (i.e. the mechanism
> is bust)

Not so.

> The problem with XML is that it was designed to be a less awful version of
> SGML while allowing the HTML document markup that already existed to be
> supported

Not so.

> which meant support for most existing DTDs.

Not so.

> So XML Schema has to
> have a lot of features that are unnecessary and complex so that all the
> horrors of SGML could be supported.

Not so.

> Both are cases of someone or rather a committee trying to define a
> specification for mapping bits on the wire to the abstract data model
> rather than mapping the abstract data model to bits on the wire.

Not so.

> I can't think of an intrinsic type that isn't a boolean, a number
> or a string. And as far as bits on the wire goes, there are no sets or bags
> either, there are only structures and lists.

All I can say is, your imagination appears deficient.

-- 
Verbogeny is one of the pleasurettes    John Cowan <[e-mail]>
of a creatific thinkerizer.             http://www.ccil.org/~cowan
   --Peter da Silva