[MEI-L] physLoc & provenance

[Johannes Kepper]

Hi Axel,

what you describe seems right and convincing, especially in the light of FRBR. I'm not sure if we can change it in the regular MEI model, though. The regular model is very open and has more or less the same content model for works and sources, which is obviously even without FRBR questionable. I wonder if we should really change this model, or instead tighten up our FRBR proposal. At some point, we will have to specify the models of works, expressions, sources (manifestations) and items anyway, so I think this could be another opportunity. If we can come up with a convincing model that works in this context, we can propose it for regular MEI as well (and who knows, maybe at some point 'regular' MEI will be a simplified FRBR-MEI?). Also, we have agreed that the current model won't be changed for the upcoming release anymore, so this couldn't be adopted before summer. At the same time, there is no such restriction for our FRBR-ODD, we may change it as often as necessary without worrying other's breaking software… So unless there is a consensus on this list that we should adopt your proposal in regular MEI as soon as possible, I would suggest to try it out in our own customization first. As soon as we have something debatable, we may re-present it here. 

But as always, that's just my first impression…
Best regards,
Johannes



Am 29.03.2012 um 13:28 schrieb Axel Teich Geertinger:

> Hi all,
>  
> working with source metadata encodings, my colleague Sigge and I are wondering why <physLoc> and <provenance> are children of <physDesc>, not its siblings. From our point of view, the physical description of an item focuses on describing the object itself, i.e. its dimensions, the medium etc. independently of its location or history. Where it is located or who has owned it does not change the physical object in this sense (except that the circumstances may have left some physical traces on the object itself, of course, but that is hardly the point...).
> It seems like <physDesc>, as it is, is to be understood somewhat like the FRBR “item” level, containing all item-specific data, but then the tag name “physDesc” does not seem very accurate. The issue becomes even more apparent in our current customization of the schema, which actually introduces the FRBR item level in <source> (as <itemList><item>, where <item> has almost the same content model as <source>).
> What would be the arguments against moving <physLoc> and <provenance> out of <physDesc>?
>  
> All the best,
> Axel
>  
>  
> _______________________________________________
> mei-l mailing list
> mei-l at lists.uni-paderborn.de
> https://lists.uni-paderborn.de/mailman/listinfo/mei-l