Re: Scope of this group's work

On 11/14/12 2:22 AM, Adrian Pohl wrote:
> As already mentioned in the last mail, here are some questions I am
> interested in regarding a definition of the scope of this group's
> work. Most of these questions have already been posed but I think it
> is important to specify an answer and define the scope a clear as
> possible on the wiki.
>
> - Is this group only about creating a schema.org extension for
> bibliographic data in a narrower sense, i.e.: descriptions of
> bibliographic resources plus person data/authority data in general?

Adrian,

As a starting point for discussion, I would separate bibliographic 
resources and authority data into two different options. There are a 
number of efforts (ISNI, ORCID) to identify named agents, of which 
library authority files are one. These are separate data sources from 
what we usually think of as "bibliographic data," and in current data 
not always linked in an actionable way. Similarly, subject authority 
files are another separate (but potentially linkable) source of data 
that provide additional information (broader, narrower, related) that is 
not carried in the bibliographic record itself. (We need to remember 
that schema.org is mark-up for web pages -- and ask: what information is 
on the page?)

This is why I feel uneasy about including things like "pseudonym" or 
"fictional person" in our work, because that information generally 
resides in the authority data but is not be present in the bibliographic 
data, and not displayed on the bibliographic display web page (today). 
By discussing these distinctions as part of bibliographic data I think 
that we are mixing bibliographic and authority data as if they are one 
and the same, which they are not. It DOES make sense to develop 
schema.org properties for authority data, but our discussion will be 
less confusing if we talk about bibliographic and authority data 
separately (IMO).


> Or is this group's work also about providing information about
> holdings, offers, availability, price, services, sites etc.? As
> access in general was already discussed here and seen as a desirable
> use case, the group probably at least has also to cover the services
> resources are provided by, their access restrictions, locatio &
> opening hours (if applicable) etc.

This reflects my question about whether we are talking about 
bibliographic data in general (which would include citations, 
bibliographies, etc.) or specifically LIBRARY bibliographic data. If we 
are focused on library bibliography data then we need to think about how 
we anticipate that data will be used by search engines (which is your 
next question). WHY do we want to surface library data (or bibliographic 
data in general) to search engines?


- Another important question
> regarding scope was already discussed on this list: Which data
> providers do we have in mind to use this extension? Do we focus on
> library data or do we want to propose a standard that's useful  for
> most of the agents that provide bibliographic information on the web
> (authors, publishers, booksellers, libraries, social cataloging
> websites, universities etc.)? Although there already seems to be
> consensus that the extension should not only cover libraries as
> publishers,  I see that most group members[1] are somehow linked to
> the library world. Shouldn't we invite more representatives of the
> different publishers of bibliographic data on the web, then? Have
> more people from other organizations already been invited?

My understanding is that schema.org grows unevenly based on who shows up 
to request extensions. If you look at the current state of schema.org it 
is very detailed in some areas, and not at all detailed in others. There 
is also some fairly uncontrolled overlap between interested parties. 
This is quite different to how we tend to develop standards in the 
library world.

Between the descriptions for schema.org/Book and the data in the product 
category, I believe that booksellers have already expressed their needs. 
(I don't see a way to track who suggested what sets of properties for 
the original set, unfortunately.) The schema.org wiki [1] shows efforts 
underway.  We are not listed on the proposals page [2] yet, but note 
that there are proposals for scholarly article and for comic books, both 
of which have considerable overlap with general bibliographic concepts.

All this to say that my impression is that schema.org development takes 
place by and for particular communities (booksellers, car sales agents, 
medical services) rather than being organized around "things." Its focus 
is on the surfacing of particular services and offers on the web, not 
describing the world. (This is my interpretation, of course.) The 
medical area [3] is an interesting example that might be closer to 
libraries than the product-oriented ones. It appears to reflect the 
kinds of medical information that exists today in web pages.

We COULD define our target as library web pages, and/or as library 
catalog web displays. It makes sense to me to model our work around web 
pages rather than data in databases. I also think we should consider the 
display of library catalog data separately from library "web pages" -- 
those pages that have information about the library. I'm not saying that 
we shouldn't also consider that data, but, like in the medical example, 
it may be a different set of elements.

kc

[1] http://www.w3.org/wiki/WebSchemas
[2] http://www.w3.org/wiki/WebSchemas/SchemaDotOrgProposals
[3] http://schema.org/docs/meddocs.html
>
>> From the foregoing discussion it sounded much like...
> a) the intended schema.org extension should be useful by for diverse
> individuals and organizations publishing bibliographic data on the
> web. b) the term "bibliographic data" is interpreted quite broadly as
> it not only covers descriptions of bibliographic resources, of
> authors etc. but also information about where and how an item can be
> obtained (lend, bought, streamed etc.)  and by whom.
>
> I already made a first start defining the scope on the "Scope" page
> [2]. As said, we should also think about what people to invite to
> this group that are not from the library world.
>
> Adrian
>
> [1] http://www.w3.org/community/schemabibex/participants
>
> [2] http://www.w3.org/community/schemabibex/wiki/Scope.
>
>
>

-- 
Karen Coyle
kcoyle@kcoyle.net http://kcoyle.net
ph: 1-510-540-7596
m: 1-510-435-8234
skype: kcoylenet

Received on Wednesday, 14 November 2012 11:41:53 UTC