[
Lists Home |
Date Index |
Thread Index
]
This is, I guess, the "document-oriented culture", which has its peculiar
needs, as you describe. But there is also the data-oriented culture, with
its own peculiar needs. All what I was hypothesizing, was that the needs
of both could be covered by single syntax and a single data model, without
inconveniencing either of the cultures.
[Len: I am talking about possibilities of designing the thing anew,
however futile that could be. This wouldn't be XML, everyone knows well
that XML is cast in stone, to last for millenia ;-) ]
[Btw, for my education: what does "GI" stand for?]
VG
On Wed, 12 Oct 2005, W. E. Perry wrote:
> This is IMHO the fundamental misunderstanding of markup which brings us once
> again to the ur-permathread of XML-DEV. In markup the GI asserts a type, or
> general case, or Platonic Form, and asserts as well the connection of that type
> to the instance value which it marks up. What the semantics, or 'meaning' of
> that type might be requires that the marked-up instance be processed so as to
> elaborate a particular meaning appropriate to that instance in the context of
> that markup as handled by that process on that occasion for a particular
> audience. There is no requirement that any reader of a marked-up instance
> process that instance in the same way as the author of that instance might.
> Indeed, the fundamental extensibility of XML lies not merely in the privilege of
> an author to assert a novel GI, or to assert a GI in a novel connection to a
> particular instance, but in the equal privilege of every audience to that
> instance to process it in an idiosyncratic manner and thereby to elaborate from
> it idiosyncratic semantics.
>
> Respectfully,
>
> Walter Perry
>
>
> Vladimir Gapeyev wrote:
>
>> On Wed, 12 Oct 2005, Elliotte Harold wrote:
>>
>>> "XML is only semi-structured with no explicit type information." is a
>>> feature, not a bug. Removing it makes the data less useful, and the format
>>> fairly uninteresting. XML would not be where it is today if it featured
>>> strong data typing.
>>
>> Why wouldn't it? In a sensibly well-designed XML-with-data-types, the
>> untyped (textual, PCDATA) values would be just one of the datatypes. So,
>> any document in XML-as-we-know-it would be a well-formed document in
>> XML-with-datatypes. Anyone not interested in types is free to ignore
>> them. (Maybe it is even still possible to define XML-with-datatypes as
>> an extension of XML 1.0, maintaining backward compatibility?)
>>
>> VG
>
>
> -----------------------------------------------------------------
> The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
> initiative of OASIS <http://www.oasis-open.org>
>
> The list archives are at http://lists.xml.org/archives/xml-dev/
>
> To subscribe or unsubscribe from this list use the subscription
> manager: <http://www.oasis-open.org/mlmanage/index.php>
>
|
