The stuff on ‘same as’ worries me. “In a Linked Data context this [making a sameas statement] is problematic, as you are then asserting that all of the statements made by the data source you are linking to through this relationship are true”

While I understand that this is the definition of OWL Sameas, I would argue here that common use has quickly overridden the strict definition, and we (and the semantic web community) will need to accept this, and move on. Even if we wished to, and we were willing to put a lot of manual effort into it, how would we ever establish that the statements someone else had made about an entity were ‘true’ – truth being a slippery concept anyway.

This paper on the use of sameas http://ceur-ws.org/Vol-628/ldow2010_paper09.pdf I think summarises the issues quite well. It introduces (among a number of contexts for asserting ‘sameness’) the concept of “Same Thing As But Referentially Opaque” which it defines as “when the two URIs do refer to the same thing, but all the properties ascribed to one URI are not necessarily accepted by the other.”

The paper suggests the community might “continue using owl:sameAs with a kind of ‘importer beware’ principle”. The paper concludes ” The (ab)use of owl:sameAs in Linked Data is not a threat, it’s an opportunity.”

This feels like the right attitude to me, and one that is essential to us exploiting linked data in a meaningful way. I worry that if the archive community (or any other community) don’t take this opportunity, the use of linked data in the community is a nice experiment, but ultimately ends up being a way of reconfiguring an isolated silo.