xml-dev - RE: NOTATION FPIs (was: <XML:SCRIPT>)

["Rick Jelliffe" <ricko@allette.com.au>]

> From:  Michael Kay

> ->Anyway there is a general expectation in XML that system
> identifiers should
> >be URIs  and that public identifiers should be SGML FPIs.
>
> I will be pedantic again. There may be a general expectation
> among the XML congnoscenti, but there is no general
> expectation "in XML", or in the wider community who assume
> that XML is what the published standard says it is, nothing
> more and nothing less.

While I certainly agree that this should be spelled out in the XML spec, I
disagree that about the distinction between "XML cogniscenti" and "XML" to
an extent.

It seems to me the way of the world that a technology and its human hosts
cannot be divided. Unless a standard is about something trivial or unless
the writers of the standard have perfect knowledge of the presuppositions of
its readership, then a specification will always be incomplete (especially
initially).

It would be nice if a spec was presented complete like Moses' tablets, but
actually when there is incompleness it will be subject to
* iterative revisions, or
* an "eldership/judges" system of dispute resolution, or
* a Maoist/NRA "all power comes from from the barrel of a gun" system where
the implementation from the largest players determines practise, or
* some additional regulation system (e.g. extra TRs at w3.org), or
* some communal agreement system like XML-DEV voting, or
* anarchy.

It has long been the bane of international standards that people treat
incompleteness in standards as surprising flaws rather than expected
incompleteness which must be constructively dealt with -- in the case of
international standards, national bodies can request the international
committees to clarify matters.  Maybe the attitude is a sign of a text-based
society rather than a human-interaction-based society in the West.  Of
course, specifications should be as complete as possible; but the
expectation that they will be complete and therefore you dont need
"congniscenti" is unrealistic, IMHO. (This is nothing more than the
traditional criticism of non-iterative waterfall models where analysis and
design and implementation cannot feedback, so I don't think it is a radical
view.)

Rick Jelliffe





xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo@ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo@ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@ic.ac.uk)