EPUB 3 Working Group Virtual Locators Task Force Telco — Minutes

Date: 2021-05-12

Attendees

Present: Ben Schroeter, Brady Duga, Garth Conboy, Tzviya Siegman, Wendy Reid, Ivan Herman, Pilar Wyman, George Kerscher, Charles LaPierre

Regrets:

Guests:

Chair: Wendy Reid

Scribe(s): Pilar Wyman

Content:

Wendy Reid: https://docs.google.com/spreadsheets/d/1KO-HyLGUUw36F-ruAARHNiPO1aUJCCNeTv3zxGtjuHw/edit#gid=0

Wendy Reid: summarized last session, incl use cases review and categorization, per locator creation and use

George Kerscher: authoring tools?

Ivan Herman: text editors are included in authoring tools

Wendy Reid: index use cases

Pilar Wyman: current practices on embedding incl page numbers just because they’re available

Tzviya Siegman: same with XML, per embedding

Brady Duga: pre-epub vs epub locators

Ivan Herman: some other tools may be called for

Brady Duga: reading systems incl often authoring tools

Wendy Reid: authoring tools, reading systems vs readers

Ben Schroeter: indexers vs authoring tools

Ivan Herman: who puts locators in epubs? that is the question.
… ie, pre vs post purchase of the epub

Wendy Reid: authoring tools include indexing tools

Charles LaPierre: what about post-production indexes?

Ivan Herman: that’s annotation, a separate use case

Wendy Reid: deep-linking cases, how do they apply?

Brady Duga: deep-linking uses cfi, and book identifiers
… user-interface issues are beyond what we are doing here

Ivan Herman: it’s ugly code, where is the spec for it?

Tzviya Siegman: full code combination for such deep-linking isn’t possible yet

Ivan Herman: theoretically we do have it

Brady Duga: disagree. we don’t have the URL specifications yet

Ivan Herman: the reading system is the culprit

Brady Duga: yes, it’s from the reading systems

Wendy Reid: the how here is the challenge, beyond our scope

Ivan Herman: unless we use CFI

Wendy Reid: we are here to find new alternatives for page numbers
… cross-reference/annotation use case

Ivan Herman: this is reading system stuff, annotation servers do this, not our responsibility.
… annotation anchors are separate, and may be our responsibility however

Charles LaPierre: we should invite the Hypothesis people (Dan) who have solved this, regarding annotation anchors

Ivan Herman: the Hypothesis system works nicely but uses a variation of locator references for their anchors, not anything we haven’t already discussed.

Tzviya Siegman: I will talk to Benjamin Young (of Wiley), who also worked on it

Wendy Reid: more index cases
… we return to making locators readable

Ivan Herman: is there a tool that converts EPUB CFIs to something readable? that is what we need

Garth Conboy: not sure EPUB CFI is less readable than other fragment IDs

Brady Duga: this is about labels

Wendy Reid: we need a locator with semantic value to replace printed page numbers. CFI does that, but it’s really long
… how can we convert CFI to something shorter?

Tzviya Siegman: and accessible
… readers like printed page numbers

Brady Duga: ranges too. page numbers are just so good, and none of this other stuff does that.

Wendy Reid: question to tzviya, what was it that readers didn’t like about the non-page number locators?

Tzviya Siegman: they found it confusing

Ivan Herman: wouldn’t it be nice if the reading system displayed page numbers per the viewing/reading system device

Brady Duga: religious war over device pages and ‘real’ pages.

George Kerscher: we are making this too complicated. easy solution is to define fake pages for publishers and we define our hats on those

Ivan Herman: what would be the heuristics?

Wendy Reid: Adobe Digital did that

George Kerscher: it’s an authoring thing, not a reading system

Tzviya Siegman: Adobe Digital was based on bit size

George Kerscher: Let’s make a golden rule, of so many characters = 1 page

Wendy Reid: internationalization would be complicated

Garth Conboy: internationalization is solvable, though

Ivan Herman: I can be convinced that’s the way to go, but we do need non-normative agreement heuristic

Brady Duga: what about page lists? can we just use those page numbers?

George Kerscher: yes

Garth Conboy: if no page list, build one

Wendy Reid: no pages in digital-only

Garth Conboy: make it up

George Kerscher: non-normative, general guideline, let’s make it up

Wendy Reid: think about this, limits that define pages, etc.