On May 9, 2014 4:46 AM, "martin.hepp@ebusiness-unibw.org" < martin.hepp@ebusiness-unibw.org> wrote: > Second: For conceptual modeling geeks and the group of On[t]oClean fans: As far as I understand, the "roles" proposal in schema.org is mainly for > > - keeping meta-data about type membership for anti-rigid properties (~R) in OntoClean terminology That would be the school of thought I alluded to in my earlier message... Since some of the founding members do not fully subscribe to the original tenets, I didn't want to summon any demons. [If you're dealing with time, you need Tennants, not tenets]. The name of the Role class seems to be confusing, and parts of the proposal Eco or overlap with the generic property proposal. There may be some underlying unity that could simplify both. Maybe embedding property definitions via the script mechanism, scoped to the rest of the webpage, could simplify the instance data (for the Bill Murray as lead actor case, and possibly some the generic properties)? These properties could be defined in terms of Roles with some attributes already specified. The associating of periods of time with a property seems to be too narrowly scoped; there might be many related statements that have related time bounds; forcing each property to be re-roled builds character, but is a little unsightly. If it were easy to wrap (no-@) context around statements in all formats this could make things simpler for some cases. A separate issue has to do with relations that intrinsically involve more than two things (e.g. Chigwell _is closer to_ Grange Hill than it is to Snaresbrook.) This parallels the "Bill Murray played the character King Lear in National Lampoon's Royal Family Vacation". It might be possible for schema.org to specify such relations (and their mapping to serializations) that makes them simpler to define and understand; however because RDF only allows binary relations, the serialized forms may end up looking the same.
Received on Friday, 9 May 2014 17:48:30 UTC