Semantic Web (Was: Re: Index in the Toolkit)

Posting to RDA-L

On 5/11/2015 10:50 AM, Bernhard Eversberg wrote:
I figure the latter statement may be understood as if you state it as a matter of fact. It may also be understood as if you point it out as a paradox of sorts, or even something bordering on scandal.

I meant it as a paradox.

Well, but is it not the Semantic Web the library community have committed themselves to? Is it not true that RDA has been developed under the impression and with the agreement that libraries must find and embrace their position in the networked world as aworldwide consortium of entities whose mission it is to make the world’s knowledge accessible and useful?

So, I’m sorry but the inflationary use of “The new international standard” smacks more like a commercial slogan more than necessary. Whereas a new, really commercially motivated standard, schema.org, comes along for free.
What has befallen this profession?

Of course, I agree with this, but I think one of the main problems is something you mentioned only in passing. What does it mean to join the Semantic Web? Libraries want to get into the Semantic Web, but there is a huge difference in the understanding of the “Semantic Web” among different constituencies. I have already discussed how library bibliographic data differs from other types of data. (http://blog.jweinheimer.net/2014/06/presentation-to-la-sapienza.html)

The Semantic Web is a huge conglomeration of “data” consisting of billions or trillions of statements (RDF triples) that can be manipulated however someone may want. Google seems to be working very hard with this. Specifically, I am referring to their Knowledge Graph and their emphasis on creating a “single search result”, or considering Google to be an “Answer Machine” (See the excellent articles “The never-ending search” specifically the section “Single answer for a query”: https://medium.com/backchannel/how-google-search-dealt-with-mobile-33bc09852dc9, and “10 Reasons Why Search Is In Vogue: Hot Trends In Semantic Search”, especially section 1 http://searchengineland.com/10-reasons-why-search-is-in-vogue-hot-trends-in-semantic-search-171284). These articles talk about how Google wants to provide, not so much a list of links, but “the answer” itself. It works to some extent. I have played with the search “When was [name of book] written?” (Here is Hamlet http://bit.ly/1E1PfNS where the date pops up in a separate box. That’s the Knowledge Graph at work) It works for famous books such as The Time Machine and The Divine Comedy and others–although for me, it worked for Moby-Dick only after I misspelled it by mistake: “moby dic”. For less famous books, it doesn’t work.

With some however, the actual date pops up within Google itself, along with the normal list of links that (I guess) go to the answer. Of course, as a searcher I wouldn’t look at those links anymore because Google has already given me the answer. Therefore, the list of links become pretty much irrelevant. Who would click on those pages after seeing the answer? I am sure this has not been lost on Google, especially as “the answer” will pop up more and more often.

“The” answer, or an “Answer Machine”. Such ideas should make librarians stop and think. What questions will get “the” answer? No matter what, standards underlie all of it. If libraries want their data findable in the Googles (where people will increasingly get the single search result), they will have to use schema.org because it is the standard they have imposed.

What else does it mean to enter the Semantic Web? Here is the latest version I have seen of the LOD cloud (http://lod-cloud.net/versions/2014-08-30/lod-cloud.png) We will be entering this incredible world as simply one extra circle, with a few lines going out to a few other circles. Or, if libraries go into it individually, it will be a lot of much smaller circles with lines, mostly going to each other, where our RDA/FRBR/Bibframe stuff will work. (Except for where it won’t work) That’s a scary scenario, but it is important to have a realistic view of the relative importance of library standards–once we enter that world.

How will we expect that our users will enter this world? Will they get there by going through our catalogs (that people don’t seem to like anymore) or will they use some as-yet-to-be-created library-made tool that we will expect everyone to use? Will they go through the Googles? Some mixture of all of those? What will their experience be as different sites with differing standards in that incredible Semantic Web/LOD cloud are mixed together?

Right now, I confess I cannot imagine how to create anything out of this that would be anything other than utter chaos. I may be wrong, but I can imagine that the Google engineers have already confronted the reality of creating something coherent from all of that, and then gave up deciding it’s much easier to create a “single answer”!

-547

Share