- From: Liddy Nevile <liddy@sunriseresearch.org>
- Date: Tue, 19 Nov 2013 08:00:45 +1100
- To: Dan Brickley <danbri@google.com>
- Cc: Charles McCathie Nevile <chaals@yandex-team.ru>, "Peter F. Patel-Schneider" <pfpschneider@gmail.com>, "<public-vocabs@w3.org>" <public-vocabs@w3.org>, "a11y-metadata-project@googlegroups.com" <a11y-metadata-project@googlegroups.com>
perhaps bear in mind that the ISO standard, for example, will want to allow for extensions, in the refinement way to maintain structure, so I agree that lists of literal values could be tricky ... Liddy On 19/11/2013, at 7:47 AM, Dan Brickley wrote: > On 18 November 2013 13:36, Charles McCathie Nevile > <chaals@yandex-team.ru> wrote: >> On Tue, 19 Nov 2013 02:44:39 +0800, Peter F. Patel-Schneider >> <pfpschneider@gmail.com> wrote: >> >>> schema.org has enumerated types, which might be better to use than >>> text >>> with a list of expected strings. >> >> >> Yes, that was what I was thinking... We should make that change. > > I'm not so convinced yet. There are quite a lot of values, and given > schema.org's flat namespace we would have to consider each term as > _the_ schema.org use of that word. > > e.g. MathML; sound; captions; latex; timing etc. would become > http://schema.org/sound ... > > My inclination (especially having seen the variety of views earlier in > these discussions) is that allowing Text and also allowing values > represented by URL might be the right combination. Schema.org's > enumerations work best for short, rigid, fixed lists that won't evolve > or get extended... > > Dan >
Received on Monday, 18 November 2013 23:44:52 UTC