It is always good sport trying to second guess what Google will do with
‘Localness’ is not always a matter of geographical nearness. IP address
ranges for example are rumoured to have some influence in this area.
However it is as much about the relationship between well described
resources - in Schema.org terms having exampleOfWork relationships with
well described CreativeWork instances - and the Organization instances that
hold them. A poorly described schema:library with little geolocation
information that is not linked to other references to the library, in
WikiData or Wikipedia for example, will hold back the relevance of the
resources held as much as poor description of the resources themselves.
Equally if the properties marked up in schema.org are all strings and not
canonical URIs for authors, subjects, publishers, etc. a search engine may
well be able to ingest the data, but linking it to other relevant entities
in their knowledge graph will be a challenge without them.
As to the major search engines tracking and consuming Bibframe. As Kevin
implies, if they find it useful they will. With 12+ Million sites across
many sectors publishing Schema.org data, I would imagine a few libraries
publishing Bibframe would be a bit of an edge case at the moment.
Kevin is also spot on to say “For a variety of reasons, organizations may
choose to use Bibframe as a more library-specific vocabulary but publish
the data for general consumption (read: for Google and others) in the
schema.org vocabulary.” In simple terms choose your vocabulary based on
your audience. If you want to get data to non-library folks, go for the
most widely used and understood vocablary.
Founder, Data Liberate
On 29 March 2016 at 16:09, Harper, Cynthia <[log in to unmask]> wrote:
> Hopefully Google will have a means to let libraries/patrons
> select/deselect areas where they will advertise their resources. We're a
> private institution in Alexandria VA. Our resources are pertinent to other
> people on our single IP domain, but less so to others in Alexandria VA.
> Maybe they'd use the same libraries you choose for Google Scholar link
> Cindy Harper
> -----Original Message-----
> From: Code for Libraries [mailto:[log in to unmask]] On Behalf Of
> Kevin Ford
> Sent: Tuesday, March 29, 2016 10:45 AM
> To: [log in to unmask]
> Subject: Re: [CODE4LIB] Deduping linked data in search - was RE:
> [CODE4LIB] Structured Data Markup on library web sites
> It's probably not safe to say that "all search is local" but there is most
> certainly a strong local component considered for every search.
> For me, every hit on the first page of Google's results for a search for
> "ice cream parlor" is related to Chicago, which is where I executed the
> search. A search for a book (I chose a current bestseller as a test),
> however, does not return a local hit in the first two pages. That's not to
> say it can't happen. It might simply (hah! 'simple') be that Google does
> not know enough about local inventory (books available from a local library
> or in stock at a local bookstore) to offer that type of
> assistance/precision. While this may seem like a theory only, Zepheira's
> libhub initiative has been trying to make this a reality by publishing
> individual libraries' structured data so that Google can make sense of it.
> And, at this point, if anyone from Libhub is on this list, I'll let you
> take it from here...
> On 03/29/2016 08:52 AM, Ruth Tillman wrote:
> > An off-the-cuff response: I've heard it suggested in talks about
> > Bibframe that just as Google tailors your results based on location
> > (i.e. if I put in "pizza," I'll get pizza places in South Bend, as
> > well as pizza recipes and whatnot), they'd tailor your library results
> > based on location. So if I were in downtown DC, and Googled a book, I
> > would see the DCPL holdings but not Indiana, and vice-versa.
> > There are maybe 5 or 10 assumptions happening there that other people
> > can spell out better, but it would be a reasonable solution for
> > deduping assuming the metadata pretty much matches.
> > On Tue, Mar 29, 2016 at 9:40 AM, Harper, Cynthia <[log in to unmask]>
> >> Forgive me if I'm confusing schema.org and Bibframe, but I wonder how
> >> Google is going to dedupe all the sources of a given
> >> document/material when many libraries have their holdings in
> >> bibframe? These sample searches made me wonder about that again. has
> this been discussed?
> >> Cindy Harper
> >> [log in to unmask]
> >> ________________________________________
> >> From: Code for Libraries [[log in to unmask]] on behalf of
> >> Karen Coyle [[log in to unmask]]
> >> Sent: Thursday, March 24, 2016 10:28 PM
> >> To: [log in to unmask]
> >> Subject: Re: [CODE4LIB] Structured Data Markup on library web sites
> >> I worked on the addition of schema.org data to the Bryn Mawr
> >> Classical Reviews. Although I advised doing a "before and after" test
> >> to see how it affected retrieval, I lost touch with the folks before
> >> that could happen. However, their reviews do show up fairly high in
> >> Google, around the 3-5th place on page one. Try these searches:
> >> how to read a latin poem
> >> /From Listeners to Viewers:/
> >> /Butrint 4: The Archaeology and Histories of an Ionian Town
> >> kc
> >> /
> >> On 3/22/16 5:44 PM, Jennifer DeJonghe wrote:
> >>> Hello,
> >>> I'm looking for examples of library web sites or university web
> >>> sites
> >> that are using Structured Data / schema.org to mark up books,
> >> locations, events, etc, on their public web sites or blogs. I'm NOT
> >> really looking for huge linked data projects where large record sets
> >> are marked up, but more simple SEO practices for displaying rich
> snippets in search engine results.
> >>> If you have examples of library or university websites doing this,
> >> please send me a link!
> >>> Thank you,
> >>> Jennifer
> >>> Jennifer DeJonghe
> >>> Librarian and Professor
> >>> Library and Information Services
> >>> Metropolitan State University
> >>> St. Paul, MN
> >> --
> >> Karen Coyle
> >> [log in to unmask] http://kcoyle.net
> >> m: +1-510-435-8234
> >> skype: kcoylenet/+1-510-984-3600