See also: IRC log
<trackbot> Date: 02 June 2014
<liza> http://www.w3.org/2014/05/19-dpub-minutes.html
Liza: OK to approve minutes?
... Minutes approved.
... Madi put together interviews at Pearson about Metadata
... complements Bill Kasdorf's interviews
... then we can talk about renaming task forces
... making the names goal-oriented
... so we don't have to solve every problem in publishing.
madi: is BIll here?
... I did some casual interviews across lots of context
... what are your challenges and pain points as you move from print to
digital
... Bill talked to lots of publishers
... I talked to lots of Pearson folks (educational publishing)
... I interviewed 12 people
... across global schools/primary schools, vocational, higher ed, and
english language teaching
... they were very candid
... about their pain points.
... the report is breezy and easy to read
... even executives can read it!
... the top priorities will be no surprise
... when I combined my interviews with Bill's
... the results were very different
... lots of differences between edu and trade
... trade books don't get atomized
... trade talked about ONIX, BISAC, PRISM
... Edu publisher want to offer modularized content around particular
subjects
... so different components can be mashed up
... they talk about learning objects
... this is true across all types of education
... students even using this material at home
... publishers are focused on getting these resources into all sorts of
devices
... top issue reported is (drum roll)
... governance
... there's enough industry standards
... but the right to refuse is still embedded in traditional publishing
process
... so publishers will refuse to add metadata
... there's no mechanism from that moment format
... in content creation and distribution
... there are rights to refuse along the entire workflow
... and they all exercise that right
... they want some kind of mechanism that says you can't go any further
until you complete A and B
... it needs to be incorporated/integrated across entire workflow
... the next two are ...
... true for trade, edu, stem
... free access is something they're dealing now
... so the question is about expressing rights
... to express rights for digital mobility, digital downloads, digital
views
... which may come from many producers
... royalties are a problem.
... if we're chunking content
... how much do we charge for the download of a chapter
... tracking things at the low level is hard
... the next thing is Flow. Workflow...
... content creation starts at the signing of an agreement.
... there are some rights agreed with the author
... but those rights are stuck in the contract, and don't flow with the
content
... we need to migrate metadata from one part of the work to the next
... at the end of the workflow
... you need to distribute the content
... but they don't know what the rights and metadata are
... so they have to research, translate, convert metadata
... and this is labor-intensive
... they want a more holistic approach
... so metadata doesn't just come in at the end
... "metadata is a myth"
... "there's no metadata, so there's no value, so there's no use case"
Bill_Kasdorf: metadata schmetadata
<Luc> Do they sell any book?
madi: half the respondents don't really know
that much about metadata
... they don't understand it enough to put it into practice with their
teams
... so it's back to governance
... so there's no one saying "this is how it's gonna go"
... "I don't get a bonus for metadata, I won't do it"
... there needs to be authority to make metadata happen
... needs to happen from top down
... businesses are ready to change
... standards was another common issue
... people can't make this up as we go
... edu publishers must gather around industry standards
... we want to offer an oasis to offer recommendations
... other answers that were'n't as popular
... inconsistency
... not just between metadata schemas, but between specific terms
... lack of incentives
... which relates to government
... working for the greater good doesn't help THEM
... a need for learning objectives
<liza> We can offer them an OASIS
madi: from standard curriculum authorities
<liza> (metadata joke)
madi: testing etc. are based on these
curriculum authorities
... everyone wants someone else to do metadata
... they want a 'metadata laundry service'
... the conclusion is that everyone has a better understanding
... but no one is ready to do something
... but they want someone else to do that something
... so let's synthesize my stuff with Bill Kasdorf's
... and there's the content and markup task force
... and avoid a piecemeal approach
Liza: thanks!
<pkra1> +1
Liza: what do edu publishers grapple with?
... trade has been behind on atomic content
... how much is this around having to distribute pieces of content?
... and how much is around the basics
madi: they're still grappling with where
they are and where they're standing
... but they now have a vision of where they need to go
... and can articulate that.
... some even said that the legacy stuff should be abandoned
Ivan: sigh
<Luc> sigh
Bill_Kasdorf: Can I add some thoughts?
TimCole: Go ahead bill
Bill_Kasdorf: First, fabulous report. It
rings true to me, even though I focused on different slice of publishing
... and trade and edu focus on different things
... let's be careful not to lose that distinction
<pbelfanti> Apologies, I need to drop
Bill_Kasdorf: a third axis: most were
thinking about trade, a subset were from STEM
... so we need to think about STEM, as they are different from both
trade and edu
... first, I finally posted my interview with Carol Myer from crossref
... it's now in the wiki
... what resonated is something that came up in idpf board meeting
... STEM is part of an ecosystem
... many of them feel that basic metadata is a solved problem
... the xml models all have extensive metadata headers
... they have loads of metadata associated with their content
... like the trade publishers, it tends to be product focused
... but it's an article rather than a subsection of something
... so not quite like edu
... it works well because there's an ecosystem
... they have to use the metadata
... if they don't have crossref and DOIs they can't so anything
... so it's not a waste of time
... now we're in this hybrid of open access and subscription
... the content needs to carry rights info
... and crossref is building a system to do that, and NISO has a
standard
... and crossref has set up fundref, a reference of funders
... so authors can publicly ack. the funders of the research
... so the big difference with STEM Is this ecosystem
... one other thing:
... the rights thing is reallly interesting.
... it's not just about discovery. You need to know the rights once you
discover something.
... ??? has a rights working group. It focuses on agents and publishers
and authors
... the news people from IPTC are also focusing on rights metadata.
... the magazine people... PRISM has a rights vocab
Bill_Kasdorf: it's all siloed; they're not working together
Liza: let's talk about discovery
... the word discovery is not suffiecient
TimCole: Scholarly publishers were meeting in Boston last week.
<liza> "Discovery and Attribution"? "Discovery and Access"? "Discovery and Rights"?
TimCole: the ecosystem comment is
interesting
... there is an ecosystem in trade, but it's seen as a one time event
... Librarians do use ONIX
... Librarians add holding info, access info...
... so metadata gets completed from several sources
<liza> "Discovery and Authority"?
TimCole: so there's a governance challenge
... we need to make sure that metadata is augmented/enriched over time
... need to do that in a way that's managed
... first step is that the metadata record at pub time is not the final
metadata
... it will update and change over time
... we need more organized way to do that
... part of it is mind-set in different parts of the workflow; what
they're responsible for
Ivan: one remark
Ivan: without minimizing importance of
crossref the tradition of the scholarly world
... they already used metadata extensively
... I was in research
... having an exact citation to another pub is very important
... this is how scholars are judged.
... a precise citation is metadata.
... crossref is important, but they came into a community that was
already ripe for doing it on computers
... and that's a big difference
Bill_Kasdorf: publishers demanded crossref
be created
... that could be done because metadata was already fundamental in that
group
... the header metadata is a required element, the body is optional in
these XML vocabs
Liza: discovery is optional but important
... there is a 2nd set of metadata is absolutely required for process
... like rights information
... maybe that's a useful split: what MUST be there, and what just helps
it be discovered
TimCole: There were DOI-X experiments done
before crossref
... when we first submitted records, we typed those records
... we were told we included too much metadata
... we didn't want to include too much metadata for a particular
function. Do the right thing in your system, but don't step on other
toes
Bill_Kasdorf: crossref is the minimal
possible metadata do do the xref function
... because they exist, they assume it has ALL metadata
... there's an assumption it's a global repo, it's not.
... getting back to Madi's report
... from edu side
... first, the granularity of content is fundamentally important
... and the other worlds don't want to think that way
... but edu has to use that.
Liza: don't want to get hung up on what the
publishers think of this
... in Safari we need to integrate things from all those different
worlds
... the common ground is more important
... and what users get from it
... what does metadata deliver to the reader?
... that's more important that what the traditional publishing silos
think.
Bill_Kasdorf: Who does the work? That's
important
... where does it come from? Who does it?
... it's someone elses problem
<TimCole> DOI-X experiments -- http://www.librarytechnology.org/ltg-displaytext.pl?RC=7867
Bill_Kasdorf: address at granular level
... different folks might be the authority for different sorts of
metadata
... but that doesn't exist right now
Liza: what do you think are most relevant to OWP?
<pkra1> 42
Ivan: We can say these things are really
relevant for OWP
... or things that we can only draw attention to
... I don't know how much energy it would require to go through docs to
identify what OWP could do
Bill_Kasdorf: We need to do this
... Having this report is great
... can we look at this over the next few weeks?
Madi: yes
Madi: I do believe we're circling the horses
... once we put it together and share with the group
... I think recommendations will emerge
Luc: I'm wondering shows that questions of
organization and governance
... but can we do anything about that?
Madi: Can you repeat question?
Luc: the questions are questions of governance. Can this group address such questions? It doesn't seem a natural activity for W3C.
Madi: Governance within standards is still
an issue
... if we can build a coalition among different standards groups
Bill_Kasdorf: There's a deer-in-headlights
reaction
... it's so mind-boggling they can't process it.
... providing a clearer path forward would be helpful
gcapiel: adding metadata at different points
in lifecycle
... there's a project called the learning registry
... AAP is doing a pilot with it
<liza> http://learningregistry.org/ ?
gcapiel: It's a technology platform from
Dept of Ed that allows metadata about a resource to be submitted by
anyone
... we worked on it from a11y perspective
... someone aligned videos with common core
... we came in and described them with schema.org
... multiple people added different types of metadata
... it's a project worth looking at
Bill_Kasdorf: that's a good example of the ecosystem thing
<gcapiel> http://learningregistry.org
Bill_Kasdorf: the existence of a service is good
Ivan: that's true for schema.org, too.
Liza: next step is tying it together with Bill's work, and then deciding which group at W3C could help with which parts
Ivan: done by next week :)
Liza: sure :)
... let's talk about renaming for metadata on mailing list
... we can also bring that up on next call
Ivan: I think an email discussion would be useful.
Liza: I'll start an email thread about it.
<pkra1> +1
Liza: I think that's it. We're adjourned.