W3C home > Mailing lists > Public > www-tag@w3.org > January 2014

Re: Ending DTD proliferation at the W3C

From: Paul Libbrecht <paul@hoplahup.net>
Date: Thu, 16 Jan 2014 15:01:34 +0100
Cc: "Michael[tm] Smith" <mike@w3.org>, www-tag <www-tag@w3.org>
Message-Id: <CA13CBD2-BCCD-405A-B18A-744CDE214010@hoplahup.net>
To: Henri Sivonen <hsivonen@hsivonen.fi>

>> I did not talk about *speccing* DTDs (oh boy!). Neither did Michael I believe.
> Mike talked about publishing new DTDs under TR. Seems like speccing to me.

Publishing a new DTD that is part of new specs is not "specifying a DTD" I believe.
For example, including a DTD in a sample set of the introductory material part.

>> So… are you telling us you want spec-makers to not produce XML collections anymore?
> I'm not sure what "XML collections" means, so I can't answer.

Think of any introductory material.

>> Or you prefer that we happily ignore the namespace features of XML the way HTML5 has done it (meaning… you push the art of mixing to the spec writers and readers)?
> I prefer that you make an editor (or system input method; see Typinator for Mac) macro for inputting those terrible URLs into your
> XML source explicitly.

That's for inputting. And it means that the namespace introduction is actually defined by the application (for example using a DTD), i.e. it is not visible. Sure that works but making this available in a standardized way within the delivery of a document is a much more interoperable way than expecting users to open the right application. Also, this only addresses the input and not the presentation within a spec and the inclusion in a web-accessible set of example XML documents.

Did you expect to include all namespace attributes in the beginning of all examples in a spec that presents XML fragments?
Even prefixes?

Oh, and is the editor proposal the way to approach the ever-linguishing attempts to approach compound-document formats?

I seem to be hearing from a parser writer.

Received on Thursday, 16 January 2014 14:02:10 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:33:24 UTC