On 8/4/2015 4:47 PM, Kuperman, Aaron wrote:
> Why would you ever care about “We discover his sexual orientation, > old girl friends, women he slept with, he apparently had a couple of > sex tapes, he’s been an alcoholic, etc. There is lots of what I > think is scandalous information in there.” Sex life and preferences > in drink change throughout life, and would be worthless for “breaking > a conflict”. And you wouldn’t need “private data” anyways. If the > author is known as a “writer on gay rights” that would be a $c, but > would be a very poor choice for a living author since next year the > author may decide that writing on animal rights, or write epic > fiction, is more fun and profitable, and we would be stuck with a > very bad $c.
> The “personal” information we need are birth dates, death dates, the > resume information (schools, jobs), and information on career and > specialties. The gender is useful, especially for someone likely to > change their surname (as is the custom for American women when they > get married or divorced. If the information isn’t useful for > identifying an author or distinguishing an author, but we are library > catalogers and it has no value to us (exception: if you are using RDA > for a criminal organization or an intelligence agency to organize > internal files – though the people who might be using RDA for such > purposes tend to be prohibited from participating in public fora).
You miss my point. I don’t care about that stuff at all, but if we are importing information from other sites, which is what linked data is all about, we will have no control over what that information will be. We have to stop thinking that in a world of linking, we control this stuff.
It is completely different from a world limited to AACR2/RDA/LCC/Dewey/MARC/NAF/SAF.
The only way to try to control it will be to filter it in some rather specific ways, and that is not so easy.
In response to Sandra DeSio:
> Catalogers specifically may not be able to control what users see, > but our libraries as a whole can. Catalogs only display what > they’re programmed to display, and those displays are pretty heavily > customizable.
> As to the information, it may be considered scandalous, but it > didn’t exactly take a private investigator to uncover it. Rob > Lowe’s various foibles were well-chronicled by the entertainment > press and the tabloids as they occurred. That said, there’s no > reason to include any of that info in an authority record as none of > it is relevant to his identity as a creator. What other people > outside the library world choose to do is outside our scope.
If we decide to import from site A, and site A imports from site B, then we can be importing from site B without realizing it. This scenario can
keep going with sites C, D and so on. Yes, it may be possible to filter out the stuff we don’t want (as I mentioned in my previous post) but it may very easily come back through other routes.
I agree that *library catalogs* do not need this type of scandalous information because when I click on the heading “Lowe, Rob” I should get
a list of resources where he is a creator/contributor. But I realize that I think that way because I am thinking like a “traditional cataloger”, who is limiting his view to the traditional library catalog. This scenario is *disappearing* because the traditional library catalog is changing, for better or worse. Most people think it will be for the better.
From a *linked data* point of view, when I view the entity “Rob Lowe” (we won’t even have to click on it anymore!) it will be a different matter entirely. In a linked data tool, what will I expect to see? I don’t know, and I have never seen a discussion of anything so concrete, but it is clear that it will go far beyond the simple (and boring) list of records we see today. And it also have to go far beyond just adding the information from NAF/VIAF if we want the public to find it useful because (let’s admit it!) our authority information is even more boring than our bibliographic records. The library’s authority records will become only one data point among many, because we will be importing from lots and lots of places with all different kinds of information. In our user’s eyes, these other places will be much more interesting than our stuff. And from the linked data point of view, the information I mentioned that is so scandalous is relevant to the entity identified as Rob Lowe.
I am not saying anything against linked data, but I am just trying to show some of the new problems and difficulties catalogers and other librarians will have to contend with. A lot of it won’t be pretty.
My own opinion: with time people won’t care. I remember how shocked I was at some of the search results I would get in the early search engines such as Altavistas and Web Crawler. Sometimes I would even get up and show people!
Today, I am sure I get much the same shocking results, but the difference is: I have stopped being shocked. I just smile or frown, shrug it off and forget it. Probably the same will happen with our catalogs/linked data tools.
I don’t like that way of thinking because it amounts to giving up, but it will probably happen.
James Weinheimer email@example.com
First Thus https://blog.jweinheimer.net
First Thus Facebook Page https://www.facebook.com/FirstThus
Personal Facebook Page https://www.facebook.com/james.weinheimer.35 Google+ https://plus.google.com/u/0/+JamesWeinheimer
Cooperative Cataloging Rules
Cataloging Matters Podcasts
The Library Herald http://libnews.jweinheimer.net/