Digital Publishing Interest Group Teleconference -- 27 Aug 2013

<trackbot> Date: 27 August 2013

<tzviya> Scribe: <tzviya>

<ivan> scribe: tzviya

<ivan> Guest: Jean (JeanKaplansky) Kaplansky, Aptara

<karen_> Guest: Bernard Grunow, Cengage Learning

Markus: opens meeting

<karen_> Vlad is AC rep from Monotype

vladimir: from Monotype
... provides software solutions for customers, e-text and digital solutions
... member of W3C and fonts WG

<ivan> Guest: Bill (Bill_Kasdorf) Kasdorf

Bernard from Cengage, an educational company, dir. of systems analysis

<mgylling> prev weeks minutes: https://www.w3.org/2013/meeting/dpub/2013-08-20

Markus: approval of last week's minutesL approved

<ivan> Use case directory

markus: all members of group are free to add use cases to directory
... we have a few people who are responsible for specific areas of use cases
... Jean oversees MathML, Rob oversees annotations
... Markus is looking for a lead on pagination
... our structure is still in beta. We will use this for about 1 month. If there is something that is annoying, point it out
... we have 1-2 new faces weekly. Growing steadily

<ivan> Guest: Bernard (Bernard) Grunow, Cengage Learning

markus: scribing. This week had 2 volunteers. So, we will stick with volunteers for now
... everyone is asked to scribe from time to time
... reminder of TPAC meeting. Register if going. Visas are often needed

Karen: main thing for TPAC: register and VISA letter can take time. Contact Karen with questions

Use Cases

<ivan> Use case directory

Markus: move on to wiki activities http://www.w3.org/dpub/IG/wiki/UseCase_Directory
... structure of use case directory is also somewhat arbitrary, expect it to change later on
... two initial use cases from naka-tobira needs input
... and transclusion
... both are still state "new" and no need to change yet
... we will probably leave use cases as "new" until we have a large sample and blessing from larger group

<scribe> ...new use cases on math from jean

<ivan> Subtopic: Math use cases

<ivan> Math Use cases

Jean: there are 2 different kinds of mathML - content-based and presentation-based MathML
... most browsers have taken a stab at presentation-based MathML. It is easier.
... We talk of separation of content and presentation. In math, presentation can change meaning of content.
... There needs to be a way to render MathML in general. Lack of native support in web engines.
... second use case http://www.w3.org/dpub/IG/wiki/MathML_UC#MathML_2 is for dynamic use of math
... this is done with presentational MathML. user interacts with math. It is only when get to very high level math that require content MathML
... 3rd case http://www.w3.org/dpub/IG/wiki/MathML_UC#MathML_3 involves using MathML with interactivity that uses results
... connects computational results of research outside of the publication
... these use cases were written by Peter Krautzberger of MathJax javascript library

Bill: a common use case: publisher input variables on graph, and changes visual display. Does this require content MathML or is presentational OK?

<Zakim> Bert, you wanted to mention (1) Math WG is interested in case 2, and (2) ask how case 3 would be authored.

Jean: to start with, presentational may be OK. The goal is to get something supported

Bert: I am on Math WG. Use case 2 is already in progress in Math WG
... question about use case 3 - who will author this? who will add semantic part?

Jean: I can't say. The authoring tools that exist are designscience, LaTeX, and something else
... Authoring tools are behind the scene. This needs to be part of use case. Will talk to Peter and bring back to Math WG

Bill: Scholarly publishing often asks how researchers are using web to do their work and how they publish their work
... consider whether mathematicians are doing the work or publishing it

Jean: there is a distinction. That is use case 3 - the interactive book.
... whether this gains traction remains to be seen, but it is a use case

Markus: can we make use case 2 more specific in terms of interacting with math?
... we expect to have many use cases of interaction. This may benefit from having specialized examples.
... How does this interact with math?

Jean: will talk to Peter about adding examples.

Ivan: general question: Have heard that current status of MathML implementation is bad.
... what kind of additional documentation would be helpful to publishing industry?
... what would improve the state of MathML support?

Jean: there have been some one on one discussions
... Peter K and Jean will do a call to action

for publishers to adequately specify what their needs are ...for browser and reading system vendors

Bert: this is a regular topic for Math WG as well
... Firefox tries to do the right thing. Chrome had some security issues, so shut it off, probably need some engineering help
... if have ideas, let Bert know

<karen_> +1 Jean

<ivan> Subtopic: book semantics

<azaroth> scribe: azaroth

Tzviya: I mentioned this briefly, in HTML structural semantics don't rewflect the issues we have. H1 is the highest structure, and in books that's nested in many other levels
... typically there is part or chapter first, so it would be good to have structure that reflects part and chapter before H1
... Chapter is typically labeled H1 and h1 in print would be h2, making it confusing or it gets tagged incorrectly
... the highest level gets a div, and then h1 so it's not correct

Ivan: To be technical, there is something called a role attribute which is defined and something HTML5 acks as a possible attribute even if not in the WG
... at first glance this attribute could be used for that, question to Tzviya if that is correct. It can be put on any element so can give the right semantics
... the problem with this, and where this group may come up with requirements, at the moment the way that html5 defines the possible values that can be used for role is fairly restrictive

Markus: it's only aria

Ivan: Where we would go with the requirement would be to have some sort of mechanism whereby the allowed values could be extended in view of this requirement
... that might be one way of solving the issue, if I understand

Tzviya: I think roles could be appropriate if able to be extended

Markus: We just need an agreement on what to use. Could be rdfa or microdata etc. At the moment a big ?

Ivan: FOr other reasons rdfa not appropriate

Markus: No need to discuss solutions, just to make sure use case is clear
... Need to answer the question why you want to do this
... GOtten that back from html WG before -- need to understand why you want to do this, so need to document extra reasons

Bill_Kasdorf: Quick comment -- one side is the semantic, the other side is the implied structure of the document starting with H1
... as pointed out publishers want to define structure outside of the H* structure. Not just semantic.

Markus: Gets interesting quickly, eg glossaries and dictionaries and relations that have meaning
... descendants can have implied or explicit semantics depending on their parents.
... Don't think it's a fair request to expect html5/svg to incorporate domain specific rules natively Need to have extensibility independent of the core specs
... No one else on queue, so no more questions?
... Again, request to update, and willing to work with you on it.

<ivan> scribe: tzviya

<ivan> Subtopic: social reading

Rob: social reading use cases
... four basic and simple use cases
... http://www.w3.org/dpub/IG/wiki/Anno_Basic_1 a simple comment made by a user
... may wish to make the comment public

comments are mostly for identity

scribe: use case 2 http://www.w3.org/dpub/IG/wiki/Anno_Basic_2 for tagging "this is mine"
... use case 3 structured comment, review for quality
... use case 4 - private comment, not for public platform but does need to go from one device to another
... such as from my nook to my ipad

ivan: refer only to open annotations
... open annotations does not define all the technical mechanisms for porting data (for example)
... something like that is probably necessary.
... this requirement goes way beyond OA and may involve many other groups

Rob: agreed. Relevant groups need to be filled out. The list could be very long
... Could pull in SVG and media fragments or assume by reference

Ivan: we should take one of simpler use cases and people who have worked on this should do thought experiment to determine what is needed

Frederick: we need to understand that some things are external.
... It would help use cases to understand what has been done already and what needs to be done.
... we need to narrow scope because this is such a broad area

Rob: OA group has some implementation experience.
... Scope question - are we talking about annotating EPUB or web?
... being more specific would be helpful

<fjh> +1 to considering web as part of the scope

Rob: internal/external question - approach in dpub should not be just internal to publication, but approach whole web (IMO)

markus: agreed. Approach the whole web. We are looking to bridge the gap

<azaroth> +1 to bridging

Markus: ask Rob to continue adding use cases

future issues

Markus: if have use cases, talk to Markus or Madi

<karen_> meeting adjourned

<karen_> +1 Tzviya

<ivan> trackbot, stop telcon

<trackbot> Sorry, ivan, I don't understand 'trackbot, stop telcon'. Please refer to <http://www.w3.org/2005/06/tracker/irc> for help.

<ivan> trackbot, end telcon

- DRAFT -

Digital Publishing Interest Group Teleconference

27 Aug 2013

Attendees

Contents

Use Cases

future issues

Summary of Action Items

Scribe.perl diagnostic output