DMA: Two roles are needed

[VladimirAlexiev]

Attn @caknoblock @workergnome @azaroth42: DMA Roles correctly maps Formerly attributed to -> http://vocab.getty.edu/aat/300404270. However, that's not a person role but a qualifier. It could be applied to any role, though most of the time is applied to Artist/Author.

So in this case I think that 2 AAT concepts should be applied: 300404270 and 300025492 author.

A similar case is present in Autry Classification. They've mapped

• bentwood box -> 300045643 boxes (containers); with a note 300172505 bentwood

It would be nice to map both:

• object type = 300045643 boxes (containers)
• material = 300172505 bentwood

But in the case of Formerly attributed to, I think it's necessary

[azaroth42]

Okay? If DMA chooses to change their classifications, that would be fine :)

[workergnome]

@VladimirAlexiev: I don't disagree with you that that sounds like a better mapping than what might be there, however, I believe that the AAC's role is to correctly convey what the museums have expressed.

To me, this is sounds very similar to discussing attributions for paintings—I think that if you are willing, and the the museums are interested in modifying their data, these sorts of improvements could be very useful for the field, but it's in my interest to get them to link to AAT, not to police how they're using it.

[VladimirAlexiev]

This issue is posted to DMA.
You guys are in cc to say whether it's technically possible to map to 2 values.
"bentwood box" is a more complex case because it maps to 2 different fields.

[bmacelhose]

"So in this case I think that 2 AAT concepts should be applied: 300404270 and 300025492 author."

This makes sense.

However, I do not think that the DMA sent over any data that falls into this "role" of formerly attributed to. I included that role in the AAT crosswalk i submitted so that the mapping would be in place should we choose to supply that information in the future. Hope this comment is of some service.

[caknoblock]

Currently, the software assumes there is a single mapping. It could be extended to support this case, but we will have to put it on the todo list for now.

…

On Mar 8, 2017, at 4:10 AM, Vladimir Alexiev ***@***.***> wrote: This issue is posted to DMA. You guys are in cc to say whether it's technically possible to map to 2 values. "bentwood box" is a more complex case because it maps to 2 different fields. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#10 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABB-qb3LYVij973R-sOTUnlWI1Io-W7aks5rjpq_gaJpZM4MUkkJ>. {"api_version":"1.0","publisher":{"api_key":"05dde50f1d1a384dd78767c55493e4bb","name":"GitHub"},"entity":{"external_key":"github/american-art/AAT-Term-Mappings","title":"american-art/AAT-Term-Mappings","subtitle":"GitHub repository","main_image_url":"https://cloud.githubusercontent.com/assets/143418/17495839/a5054eac-5d88-11e6-95fc-7290892c7bb5.png","avatar_image_url":"https://cloud.githubusercontent.com/assets/143418/15842166/7c72db34-2c0b-11e6-9aed-b52498112777.png","action":{"name":"Open in ***@***.*** in #10: This issue is posted to DMA.\r\nYou guys are in cc to say whether it's technically possible to map to 2 values.\r\n\"bentwood box\" is a more complex case because it maps to 2 different fields."}],"action":{"name":"View Issue","url":"#10 (comment)"}}}

[kateblanch]

These lists were compiled as reference tables / thesauri / look up lists of terms that were not specifically engineered or designed to match up well to the AAT.

We are literally forcing AAT associations to descriptive text values that support an often idiosyncratic museum classification system, not a finely tuned machine like the AAT. It's destined to be imperfect. Reclassifying objects isn't always possible by partners.

Watching these mapping issues emerge has given us a new category of LOD best practices to learn and document, they're all valuable though not all resolveable (?). HTH!