Re: BIBFRAME and schema.org

Hi All,

Let me clarify a few things from my point of view.

Firstly an apology for not posting the document on list as soon as it could have been.  I was aware that the publication this report to the BIBFRAME community was imminent from Jean Godby, but the actual timing, with dependancies on when the document could be loaded on the web etc., was not known.  Flight schedules and airport wifi conspired to prevent me from monitoring the BIBFRAME list closely enough to be able to repost it on this list as soon as I would have liked.  Again I thank Karen, and her observant following of these lists, for doing my job for me.

Secondly, I believe that we should take note of what prompted the report and therefore its context.  To quote Jean 'The analysis presented here was prompted by the call at the end of the December 2012 BIBFRAME Early Experimenters Meeting for a set of Point or Position papers that work out technical issues and make recommendations for a number of sketchy, difficult, or controversial aspects of the BIBFRAME model.' and 'This draft is being released as an OCLC report, but it is intended to be read as a working paper for the BIBFRAME community.'

Jean in producing this report to the BIBFRAME community listened to the published SchemaBibEx discussions, read our public email conversations, and the documentation on the wiki over several months.  Having reread her document I am not aware of any topic she attributes to SchemaBibEx discussions that has not at sometime been discussed in the group.  In the refining of her thinking Jean interviewed some of her colleagues, particularly Jeff Young and myself who are involved in Schema.org.  We were also able to comment with others who are interested in the importance of linked data for the library domain, on early drafts.

Next I should address my personal position, as an OCLC employee, in this.  Personally I believe that the issues, challenges, and opportunities being discussed in groups such as SchemaBibEx and BIBFRAME of vital importance to the library, bibliographic, web and semantic web communities. That importance is way larger than any individual organisation, be they commercial, cooperative, individual or government backed. I am fortunate enough to be able to provide some hopefully unbiased facilitation in these areas & communities and the backing of my employer to invest the time (and some of their resources) in doing this.

OCLC I believe has two important roles in these debates.  Firstly as a cooperative representing tens of thousands of member libraries, and secondly they research and manage on a day to day basis a huge aggregation of bibliographic data. The significant experience gained into the transformation, analysis, and processing of this data give them unique insight in to the issues and challenges that arise.

Back to the report itself.

I am not going to go into a point by point discussion here, but taken in the round I believe its description of the general direction of travel of our discussions on how Schema.org could be applied to and influence the description of bibliographic data is not far off.  Especially in support of the conclusion that the BIBFRAME community [to whom the report is targeted] should look to take note of our work and consider cooperation with.

As  commentator on our activities I believe that Jean possibly over emphasises the goal of representing FRBR in Schema.org (something she states would not be acceptable by Schema) - I think it is her interpretation of our challenge to use Schema to describe resources and their relationships – relationships we [library folk] understand and often discuss in terms of FRBR.

Again in general, I believe that her report will be an excellent addition to the debate in the BIBFRAME community and hopefully broaden the debate around the use of library linked data, and most importantly its place in the broader web of data.

~Richard.




From: Shlomo Sanders <Shlomo.Sanders@exlibrisgroup.com<mailto:Shlomo.Sanders@exlibrisgroup.com>>
Date: Friday, 28 June 2013 05:10
To: Dan Scott <denials@gmail.com<mailto:denials@gmail.com>>
Cc: "public-schemabibex@w3.org<mailto:public-schemabibex@w3.org>" <public-schemabibex@w3.org<mailto:public-schemabibex@w3.org>>
Subject: Re: BIBFRAME and schema.org
Resent-From: <public-schemabibex@w3.org<mailto:public-schemabibex@w3.org>>
Resent-Date: Friday, 28 June 2013 05:11

+1

Thanks,
Shlomo

Sent from my iPad

On Jun 28, 2013, at 10:46, "Dan Scott" <denials@gmail.com<mailto:denials@gmail.com>> wrote:

Thanks Karen. I generally concur with your reaction (although I have to admit that for the first time in my life I was getting hung up on the _terrible_ kerning of the font in the PDF, so had been reading through it rather slowly).

I was surprised by many of the statements in the paper about the direction, decisions, thoughts, and beliefs of the Schema BibEx group. Perhaps if all (or most) of those statements were modified to say they were the direction, decisions, thoughts, beliefs "of the OCLC employees currently participating in the Schema BibEx community" that would be more acceptable--certainly closer to the truth.

I very much value the opinions (& Richard's leadership) of the OCLC participants in this group, but cannot endorse this paper as an accurate reflection of the group's positions, direction, etc as a whole, particularly with respect to BIBFRAME.


On Thu, Jun 27, 2013 at 3:13 PM, Karen Coyle <kcoyle@kcoyle.net<mailto:kcoyle@kcoyle.net>> wrote:
Richard, now that I've gotten further along in reading this, I think that "posting it to the list" is the least that you should have done. This document, written by OCLC and not vetted by this list, attributes to the schema bibex group a number of decisions and thoughts that I do not recognize. It uses "we" to mean not OCLC but the bibex group. I find this more than just problematic - this is at least arrogant and possibly dishonest. I now find decisions attributed to this group that I cannot condone, yet as a member of the group one could infer that they are mine as well.

OCLC cannot speak for this group, and it definitely cannot speak for this group in a document that this group did not even see. Godby stated that this would be presented at the BIBFRAME session at ALA. If it is presented as the thoughts of the bibex group and not OCLC, you should be ashamed.

Here are just a few examples from the document:

"This shift in focus implies a decision by the SchemaBibEx community to defer to the important standards initiatives of the library community, including BIBFRAME, to develop vocabulary required for detailed descriptions of library resources." p. 11 - I do not think we have discussed this at all. In fact, we haven't really discussed the relationship of schema.org<http://schema.org> and BIBFRAME in any detail, and I'm not sure it is necessarily appropriate for us to do so in this forum. There may be some folks on the group who aren't even paying attention to BIBFRAME, but who wish to mark up bibliographic displays unrelated to libraries.

"Though the BIBFRAME initiative needs to develop its own policy with regard to the Product Types Ontology, the SchemaBibEx community sees little need to define and maintain a competing vocabulary for content types and carriers." p. 17 - Again, a decision that I do not recall. Also, AFAIK, no one except Jeff has promoted the use of the product types ontology, and we haven't discussed its use in any detail in the group.

"The SchemaBIBEx community is exploring the possiblity that schema:IndividualProduct ... corresponds reasonably well to the definition of FRBR Item." p. 17 - We haven't really touched on the item level yet. This is one idea, but it is premature to attribute this thinking to the group.

"To move forward, two issues must be resolved. First, we must reach agreement on working definitions of key concepts. Then we must solve the technical problem of mismatched expectations about domain and range values..." p. 18 - I object to the use of "we" here because it is talking about the work of the SchemaBibEx group, not OCLC. This implies that the document is coming from the bibex group, not OCLC. That is not true.

Sorry for the blunt talk, but this document must be re-written to reflect that it is the thoughts and opinions of OCLC, not the bibex group. And that absolutely must be made clear at ALA.

kc



On Thu Jun 27 10:41:15 2013, Wallis,Richard wrote:
Thanks Karen for posting this to the list.  Travelling got in the way of
me ensuring that it was published here and on the BIBFRAME list at about
the same time.

~Richard.

On 27/06/2013 11:59, "Karen Coyle" <kcoyle@kcoyle.net<mailto:kcoyle@kcoyle.net>> wrote:

All,

If you are on the BIBFRAME list you will have seen a message from Jean
Godby with a link to her paper:

Godby, Carol Jean. 2013. The Relationship between BIBFRAME and the
Schema.org<http://Schema.org> ŒBib Extensions¹ Model: A Working Paper. Dublin, Ohio: OCLC
Research.
http://www.oclc.org/content/dam/research/publications/library/2013/2013-05

..pdf.

This 41 page(!) paper is an excellent analysis of the possible
relationship between BIBFRAME and schema.org<http://schema.org>. This is a topic which we
have not discussed directly in this group, and I would like to propose
that we could merge this discussion with our consideration of
"instanceOf" and "has Instance" -- which we decided to push to this list
at the end of our last webex meeting on Tuesday, June 25.

The paper presents the alignment of schema.org<http://schema.org> and FRBR as a primary
goal of this group [1]. I take exception to that, as may others. But I
believe that the underlying question is the coordination of BIBFRAME and
schema.org<http://schema.org>, and that should be discussed first. There are obvious
benefits to the library community to bringing these two into alignment,
but we should also discuss whether we can do so without silo-ing library
data once again.

kc

[1] "The main objective of the redesign is to improve the representation
of the FRBR hierarchy using concepts already defined in Schema.org<http://Schema.org>.
Since the application of the FRBR hierarchy requires the association of
descriptions with differing degrees of abstraction, the schemaBibEx
community has also proposed the properties hasInstance and isInstanceOf,
whose semantics resemble the BIBFRAME properties with the same names."
(Godby, p. 11)
--
Karen Coyle
kcoyle@kcoyle.net<mailto:kcoyle@kcoyle.net> http://kcoyle.net

ph: 1-510-540-7596
m: 1-510-435-8234
skype: kcoylenet






--
Karen Coyle
kcoyle@kcoyle.net<mailto:kcoyle@kcoyle.net> http://kcoyle.net

ph: 1-510-540-7596
m: 1-510-435-8234
skype: kcoylenet