- From: Michiel de Jong <michiel@unhosted.org>
- Date: Mon, 14 May 2012 12:20:04 +0200
- To: Michael Brunnbauer <brunni@netestate.de>
- Cc: Danny Ayers <danny.ayers@gmail.com>, public-rww@w3.org
i think Jeni's proposal is valid, and 303s and hash uri rule are not. On Mon, May 14, 2012 at 10:53 AM, Michael Brunnbauer <brunni@netestate.de> wrote: > In the case of properties like dcterms:subject, we need four different > properties. that was also my second thought. > If yes I would be surprised that the pain from httpRange-14 is > so big that people are considering something like this which also seems to > involve a rethinking of all semantic web standards and ontologies. no. if a vocabulary has not already thought about which one of the 4 options a certain property means, then it was broken. we should just indicate what the correct interpretation is for the properties we already use, and add new once if needed. IF for any given vocabulary this process involves "rethinking" as you say, then that vocabulary was broken! Not rethinking a broken vocabulary is an option, and probably many vocabularies will choose that option for a long time, but the ones that are really being used will /want/ to rethink "when i say 'license', what do i mean?" if they have not already done so. you could make a common practice to prefix properties, so you get "license" to mean subject-sense, object-contents, and e.g. "doc-license" to mean subject-contents, object-contents. my unfinished proposal for Hungarian rel names (see other thread on this list) is very much related to that. Let's just name our variable right, and then there is no problem. For me, Jeni's blogpost closes the http range 14 discussion, and moves it into each vocabulary, where naming is probably easier to get right.
Received on Monday, 14 May 2012 10:20:38 UTC