Meet 20121017

Agenda

• Comments on previous meeting
• Scope
• Topics - See Areas for Discussion
  • Publication Event
  • Work/Manifestation
  • Identifiers as things
  • Content /Carrier
  • Controlled Vocabularies - SKOS concepts & schemas - Jean Delahousse
• Use Cases
• Documenting proposals
• AOB
  • Meeting scheduling

Actions

• Provide examples and/or use cases from a Publisher's point of view - Laura Dawson
• Describe schema.org purpose (see if Dan Brickley can help) and how SchemBibEx purpose fits with it. - Richard Wallis
• Reorganise Wiki to better represent a set of themes for discussion/recommendation. - Richard Wallis
• Provide example of how library holdings could be described with schema.org. Describe in wiki in a way that could be used as a template for future examples/recommendations. - Richard Wallis / Jeff Young
• Create a meeting schedule for next three meetings.

Call Recording

To view a recording of approx 1 hour call: Click Here
Note: will play using a WebEx browser plugin

Unfortunately, due to a technical problem, the first few minutes were not recorded. The recording starts after I return to the call having been kicked out. Fortunately there was little of substance discussed prior to that point. Richard Wallis

Chat transcript from Call

From eric miller to Everyone(04:06:28 PM)
is the telecon bridge working?
From eric miller to Everyone(04:06:50 PM)
/me hears what sounds like background noise
From kefo to Everyone(04:06:57 PM)
I've been wondering the same.
From eric miller to Everyone(04:06:58 PM)
/me thinks he now hears jeff ;)
From Owen Stephens to Everyone(04:19:37 PM)
+1 to Karen
From Owen Stephens to Everyone(04:23:51 PM)
Agree with that Richard - but that's why I think the use cases are the most important starting point
From eric miller to Everyone(04:25:16 PM)
(add me to the list of folks interested in use cases)
From LAURA DAWSON to Everyone(04:25:38 PM)
Also me.
From eric miller to Everyone(04:25:45 PM)
but as to 'why' ... schema.org is dealing with describing products to improve search and discovery. Some of the products relate to resources that libraries care about. There is currently grey area in how best to describe these resources as there are many stakeholders involved. To the extent we can recommend (though consensus but perhaps even more via publishing data on the web) a pattern for describing these resources that complement the descriptive practices that are common for libraries, the continuum between discovery and description that GLAM care about can be more effectively managed.
From LAURA DAWSON to Everyone(04:26:30 PM)
I would also extend what Eric says to include not just "products", but content itself.
From eric miller to Everyone(04:28:44 PM)
i used products only to make the point that what schema.org is focused on really is about helping accelate more effective discovery of things that can be bought / sold. We (in the library community) can recognize this and leverage this direction to support our (more content focused) requirements.
From LAURA DAWSON to Everyone(04:29:57 PM)
Agreed!
From Corey Harper to Everyone(04:30:55 PM)
Eric: What gives you the sense that schema.org is primarily about finding things to buy & sell? Seems to me the goals of the SE's in general are broader than that.
From Corey Harper to Everyone(04:32:06 PM)
By way of example, the first real application of Google's rich snippits was faceting on cooking times & ingredients in recipies, right?
From LAURA DAWSON to Everyone(04:32:32 PM)
Good point.
From Jerry Persons to Everyone(04:33:42 PM)
hmmm, might then there be value in starting from what exists (the world of GLAM metadata), seeing what cannot be expressed in schema.org, determining if what's missing is of "such high value" that it "must" be included (whatever the criteria for "must" might be) and crafting recommendations around adding the must-have stuff that's missing
From jean delahousse to Everyone(04:34:04 PM)
finding a "recipe for banana bread" in Google USA is the most astonished thing you can do. It was the best demo I found to explain the value of schema.org to music librarians
From Owen Stephens to Everyone(04:34:26 PM)
Specialist materials could be rich here
From Owen Stephens to Everyone(04:34:30 PM)
and different
From eric miller to Everyone(04:34:31 PM)
Corey ... this is based from extrapolative from my discussions with schema.org originators, observations from the first F2F schema meeting and the money involved in understanding the advertising space.
From Corey Harper to Everyone(04:35:12 PM)
Eric: That makes sense. Particularly the advertising peice, which I always (somewhat willfully) forget about...
From eric miller to Everyone(04:35:42 PM)
what schema.org is doing isn't *bad*... in fact its quite good for the web (in terms of accelerating resource description). we should simply be clear that we have different goals
From LAURA DAWSON to Everyone(04:35:55 PM)
It wasn't a TV, it was a developer
From LAURA DAWSON to Everyone(04:36:05 PM)
A LOUD developer
From Owen Stephens to Everyone(04:36:08 PM)

)

From Karen Coyle to Everyone(04:36:49 PM)
Eric, yes, but what ARE our goals?
From eric miller to Everyone(04:37:23 PM)
karen, thats why last week i focused on nailing down issues of scope / use cases
From Karen Coyle to Everyone(04:37:31 PM)
me, too
From eric miller to Everyone(04:37:46 PM)
s/week/meeting
From eric miller to Everyone(04:37:55 PM)
(weeks and meetings blur)
From Owen Stephens to Everyone(04:38:06 PM) If we are starting from the "what we have" perspective (which I'm not that keen on, but running with it) wonder if the facets used in s/w such as Blacklight/VuFind would suggest those facets that would also be interesting in web search
From Jerry Persons to Everyone(04:39:01 PM)
owen: yup, a good place to begin
From Karen Coyle to Everyone(04:39:17 PM)
I think if we start with "what we have" we risk not providing what is needed.
From Corey Harper to Everyone(04:39:44 PM)
Owen, Jerry: I think that's one good area. I also think the "holdings/availability" use case is particularly compelling here, and a huge gap in schema.org.
From Owen Stephens to Everyone(04:40:28 PM)
Agree with the Holdings/Availability issue completely - and think clearly expressible use case
From Owen Stephens to Everyone(04:40:46 PM)
"Support users in finding a library that holds a copy of the item"
From Jerry Persons to Everyone(04:41:16 PM)
corey: ok, how about an OCLC id ... that's what they do best, embed it in schema rather than replicate all the OCLC does
From Corey Harper to Everyone(04:41:19 PM)
Also fits in well with SE tendency towards personalized / context sensitive search.
From Karen Coyle to Everyone(04:42:31 PM)
http://kcoyle.net/libRichSnip.png
From Karen Coyle to Everyone(04:42:56 PM)
This is really the only user-based application that I can think of, although perhaps another one would be extracting citations
From jean delahousse to Everyone(04:43:08 PM)
Simple and useful
From pkirly to Everyone(04:44:06 PM)
for Holdings/Availability a starting point would be the NCIP standard
From Owen Stephens to Everyone(04:44:20 PM)
or DAIA
From Owen Stephens to Everyone(04:44:47 PM)
much simpler!
From eric miller to Everyone(04:45:18 PM)
holdings sitemap
From pkirly to Everyone(04:46:53 PM)
I don't have experiement with DAIA. Is it the definitive guide? http://www.gbv.de/wikis/cls/DAIA_-_Document_Availability_Information_API
From Owen Stephens to Everyone(04:47:40 PM)
Yes - that's it
From pkirly to Everyone(04:48:09 PM)
thanks
From eric miller to Everyone(04:48:14 PM)
for those people taking action items, can you say your name?
From eric miller to Everyone(04:48:23 PM)
thank you laura ;)
From eric miller to Everyone(04:49:01 PM)
action item: Laura to work up publisher use case
From LAURA DAWSON to Everyone(04:49:05 PM)
You bet. Did this with ONIX years ago.
From eric miller to Everyone(04:49:18 PM)
bingo
From GordonD to Everyone(04:50:13 PM)
Environment/strings of schema.org is natural language (language of documents); environment/things of library domain is controlled language/identifiers - our task is to find the interface
From eric miller to Everyone(04:50:23 PM)
/me not sure how to capture the action item richard just agreed to
From LAURA DAWSON to Everyone(04:50:45 PM)
Re-jigger wiki can be an item!
From Owen Stephens to Everyone(04:58:55 PM)
I think Karen's point is that we want to be able to say "This book is about Albert Einstein", not to say "Albert Einstein is a term in a controlled vocab"
From Karen Coyle to Everyone(04:59:13 PM)
Owen, exactly.
From Owen Stephens to Everyone(04:59:49 PM)
It seems that saying a book is about or mentions a concrete thing is a stronger use case than saying it is about or mentions a concept From Owen Stephens to Everyone(05:00:04 PM)
Sorry - I meant "it seems to me"
From Corey Harper to Everyone(05:00:14 PM)
Owen, Karen: but I tihnk we want to be able to do so in a way that exposes the connection between Einstein & the additional metadata we have about him.
From Owen Stephens to Everyone(05:01:07 PM)
Small Demons is e.g. of working on this linking books to concrete things
From LAURA DAWSON to Everyone(05:01:31 PM)
Yes, Owen, I was thinking of asking Small Demons to have some input here.
From Corey Harper to Everyone(05:01:40 PM)
Owen++ Small Demons is super cool.
From Karen Coyle to Everyone(05:01:43 PM)
Corey, yes, but I'm not sure how you do that in schema.org -- so that's what I'm trying to imagine. If it's just a matter of using IDs, then we've already got those.
From Owen Stephens to Everyone(05:02:03 PM)
Small Demons is super cool idea, but I've always been left a bit 'meh' by the connections exposed
From Owen Stephens to Everyone(05:02:11 PM)
It may just be me :)
From Corey Harper to Everyone(05:02:23 PM)
Karen: I think it's a matter of ensuring that our "authorities" document a relationship to some of schema.org's abstract classes like thing & intangible.
From Karen Coyle to Everyone(05:02:31 PM)
right, but HOW?
From Corey Harper to Everyone(05:02:48 PM)
Karen: exactly.