Posting to RDA-L
On 8/17/2015 9:30 PM, Heidrun Wiesenmüller wrote:
I’m not saying this [i.e. place of origin of the expression/work] should be recorded, by default. And I’m definitely not saying catalogers should try and actively find out information like this. But I think RDA should give us the possibility to record it if one happens to know it. There are probably a lot of elements in RDA already which are only rarely recorded. But when something is a legitimate attribute of an entity, my feeling is there should be a “slot” in RDA to > cover it.
The idea that RDA should allow the possibility to record some bit of information that some may find legitimate is interesting. Obviously, such a bit of information is useless for *access* if it is recorded only sporadically. By this I mean that a searcher who wants to search works or expressions by their “places of origin” (e.g. works created in Rome, Italy) can never get a result that would be even semi-reliable and such a search should not be implemented. Therefore, place of origin can only be–to follow Bernhard’s criteria–related to description and never related to access–not if we want to give access that is in any way meaningful. Description would also be only on an ad-hoc basis.
But I guess the idea is that it cannot hurt to have it in there because although it doesn’t help the searcher and it doesn’t help the librarians, it doesn’t hurt to be there. Seymour Lubetzky came up with other criteria: a record should include only the information that is necessary. (His paper, “Is this rule necessary?”)
But when we consider: “what is a “legitimate” attribute of an entity?” we discover there is almost no end. In IMDB, there are lots of “attributes” of actors. In the IMDB entry for Arnold Schwarzenegger (http://www.imdb.com/name/nm0000216/bio?ref_=nm_dyk_aka) we find his height, “Trade marks” and so on. On yet another page, we find his Star sign (http://www.imdb.com/name/nm0000216/?ref_=nv_sr_1)
In the Notable Names database (http://www.nndb.com/people/685/000022619/), we find his Ethnicity, Sexual orientation, Religion, Girlfriends and Mistresses, and “Risk factors” among other tidbits. Many of these are links, so I assume that this information is for access. For instance, I can click “Steroids” under “Risk factors” and get a list of other who have also taken/take steroids.
These represent a lot of attributes that somebody, somewhere considers to be “legitimate”. I am sure there are lots more attributes for people out there. I hope catalogers will not be expected to add them because some out there consider them legitimate.
I am not criticizing any of this because people are obviously showing interest, but what is the role that library catalogs should play in such an environment that the library world is aiming for? The idea of linked data is that all of this will be mashed together in some way, shape, or form and when that happens, what role will library catalog data play in it?
One concern that I have is that some other linked data project will see the library community’s “place of origin of the work/expression,” not understand the issues and actually implement it as an access point. When a searcher inevitably discovers it doesn’t work, who will he/she blame?
Perhaps we should plan for the world of linked data, so that other projects that want to add in information can be included. Why can’t we assume that some other project will include our records, and if they want something like “place of origin of the work/expression,” it will be implemented or crowdsourced by people who might want it and are willing to spend the time and resources to make it at least semi-useful?