Re: Ending DTD proliferation at the W3C

On Tue, Jan 14, 2014 at 12:39 AM, David Sheets <kosmo.zb@gmail.com> wrote:
> On Mon, Jan 13, 2014 at 7:22 PM, Michael[tm] Smith <mike@w3.org> wrote:
>> Specifically, I'd like for there to be a document from the TAG somewhere
>> stating that:

I support Mike's proposal (though see below for an exception).

> How should vocabulary specifications exchange named character references?

They shouldn't.

As a special exception, the spec 2nd edition of XML Entity Definitions
for Characters (http://www.w3.org/2003/entities/2007doc/Overview.html)
could be allowed to proceed, since browsers are already required to
special-case some public ids so that they map to a (non-validation)
DTD that defines those entities. Note that XML Entity Definitions for
Characters doesn't use DTDs for validation and, therefore, is
orthogonal to the anti-DTD rationale that Mike stated. There
potentially exists a different anti-DTD rationale against entity
definitions, but, as noted, this particular list of entity definitions
is, for better or worse, already a requirement in practice. (Maybe in
the future we'll get XML5/XML-ER that makes these character names
built-in.)

-- 
Henri Sivonen
hsivonen@hsivonen.fi
https://hsivonen.fi/

Received on Tuesday, 14 January 2014 07:14:52 UTC