DPUB IG Telco, 2016-11-21: DPUB-PWP: locators, packaging

See minutes online for a more detailed record of the discussions.

Locators

The current text in the editor’s copy is grossly outdated. It originates in the discussion last spring, before the major reorganization of the IG’s thoughts in September. It needs change.

Although the alternative of removing it altogether came up, it was finally agreed that this section should stay, albeit renewed. The discussion thread that happened on the mailing list (e.g., thread on locators) contains a lot of valuable thoughts that should be incorporated into the document. Furthermore, Dave Cramer also jotted down a set of more general thoughts that are related to the subject, and that should not be lost either.

It was therefore decided to redo that section summarizing those discussions. First results should be expected next week.

The term “WP Processor”

As a side track of the discussion the usage of a “(P)WP Processor” came up and it was questioned whether that term was really necessary as opposed to simply use “User Agent”. After a brief discussion it was agreed to remove that term from the document.

Packaging

There has been a renewed interest on Web Packaging lately (see the “explainer” in GitHub). The question is whether it fundamentally affects the work around WP. The (temporary) consensus it that the exact packaging format is not really of importance for a possible specification; profiles can fix one or the other. However, the work on Web Packaging highlighted the issue about package signing which may be of importance for a WP-related security model in the future. (Although it is probable that the same approach could be used with ZIP-based packaging formats, too, i.e., this is not dependent on the Web Packaging format.)

DPUB Category on Discourse

A separate dpub category has been set up. The downside, however, is that in the current discourse set-up it may not be possible to set a submission into several categories. That may be a problem insofar as it would isolate the dpub related discussions. To be explored further…

DPUB IG Telco, 2016-11-28: DPUB and WCAG

See minutes online for a more detailed record of the discussions.

WCAG 2.1 and Digital Publishing

The WCAG WG is working on WCAG 2.1. There are a few items that are relevant to digital publishing that have been proposed as additions or revisions. DPUB members have been working with WCAG to to include this material in the WCAG 2.1 First Public Working Draft. Our next steps are to:

  • Propose an example of a publication in the section about set of web pages. Matt Garrish has proposed “Example: A publication is split across multiple Web pages, where each page contains one chapter or other significant section of the work. The publication is logically a single contiguous unit, and contains navigation features that enable access to the full set of pages.” This language is intentionally broad so that it can encompass any existing and future specs.
  • Discuss whether accessibility metadata in schema.org will be included as a best practice, AAA Success Criteria, or some combination.
  • Revise the existing Appendix on Metadata as appropriate

DPUB IG Telco, 2016-11-21: First review of new DPUB-PWP version; CSS actions

See minutes online for a more detailed record of the discussions.

First review of new DPUB-PWP version

The grossly outdated DPUB-PWP document has to undergo changes as a result of the changes in the UCR document. A first draft of the changes is the editor’s draft: at this moment, it is hardly more than a reshuffle of the old version with a new order of the sections, a new organizations. Discussions have begun on what to do in the document. The meeting addressed some details.

  • Title of the document: there is an agreement to use simply “Web Publications”
  • EPUB, EPUB4, Profiles: the relationships with EPUB and, in a more general sense, reference to Profiles has to be re-considered. Profiles of WP-s may become an essential feature, and EPUB (eg, a future version) may become a profile of WP, alongside other possible profiles. This section has to be thoroughly rewritten
  • API-s: there are two, currently empty sections in the documents on API-s, as a result of the discussions that took place mainly at TPAC. It was agreed that these sections should probably remain in the document, although it is not clear which group would pick those up
  • Locators: the whole issue of locators should probably be re-considered. The old version (that is also in the text) was influenced by a much more rigid view on online/offline and packaged states, and maybe that should be re-thought. Work for the future…

To be continued…

Miscellaneous

The group reviewed the pending actions that were defined at the joined meeting with the CSS WG at the TPAC: Media queries for MathML (currently pending, because it is not clear that this is the right solution after all), character based alignments in tables (collecting use cases), and hanging punctuation.

The usage of W3C’s Discourse as an ‘incubating’ tool came to the fore, with an item on list heading and captions recently raised on that list by Tzviya. The evolution of that item shows the value of using that tool, which may be used more extensively in the future.

DPUB IG Telco, 2016-11-14: Service Workers in action 2.

See minutes online for a more detailed record of the discussions.

Service Workers in action 2.

(Followup of last minute’s meeting on the usage of service workers.)

Hadrien Gardeur made a presentation of the proof-of-concept work he has done.

First of all, Hadrien has defined a version of manifest using JSON-LD; this work was done, originally, as part of the EPUB3.1 Working Group at IDPF, under the heading of “browser friendly manifestation”, together with Dave Cramer. It contains the various data that EPUB3.1 has in the OPF file (spine; further resources; “links”, i.e., various type of metadata). The manifest file can also refer to a separate Web App Manifest. A manifest can be discovered through direct access, through an HTML link element, or via an HTTP header. The various prototypes he has built make use of this manifest format.

One prototype is a “Web Publication JS” (or “Progressive Enhancement”, see a specific book) is some sort of a Web Publication that carries a specific JS code using Service Workers, and also links to the manifest files. The code injects some navigation signs into the content and displays it in the browser.

The other prototype is a “Web Publication Viewer” (a live demo is also available) which is more kind of a “reading system” that uses its own Service Worker based JS code and displays the content of a book in an iFrame.

Subsequent discussions included:

  • The mutual relationships of the Publication and the Web App Manifests. Hadrien’s approach separates these two, whereas the presentation last week essentially built “on top of” a Web App Manifest via extensions. It is not clear at this moment which approach is better, but it was agreed that a feedback on the current problems should be provided to the Web App Manifest spec developers.
  • Problems around the usage of Service Worker in the second, “Web Publication Viewer” case. The problem is the strict separation of the iFrame context, which makes it difficult to properly implement the situation when one chapter has a direct link to another chapter; that operation happens within the iFrame, i.e., the cache control on the top level browser has no information about it. (Note that Daniel Weck referred to a separate discussion thread that shows that this is an active issue for developers). Again, feedback to the Service Worker spec developers would be a good idea.

DPUB IG Telco, 2016-11-07: Service Workers in Action

See minutes online for a more detailed record of the discussions.

Service Workers in action

Dave Cramer made a short presentation of a proof-of-concept work he has done: a “reader” page that:

  • had a number of references to books; each book:
    • is a collection of HTML/CSS etc files
    • has a manifest: a small extension to the Web App Manifest, also containing a list of resources and a “spine”
    • can be displayed via a javascript application bound to the reader page (as opposed to be part of a book) in an iFrame
    • may have book-wide user preferences (eg, night mode)
    • can be stored offline using a Service Worker based functionality
    • can be dumped into a zip file
  • the reader page has interaction buttons for ‘next’ and ‘previous’, using the “spine” in the book’s manifest, as well as buttons for the offline storage and dump to zip

The discussion following the presentation concentrated on issues like how one would store, on long term, data like user’s preferences, annotations, etc. The value of having the reader javascript application attached to the reader page, ie, the fact that the “book” itself does not (necessarily) carry javascript was also emphasized.

One specific issue that came up (for future discussion) is whether the fact that that Service Workers rely on HTTPS (as opposed to HTTP) is, for publication, limiting or future looking?

Obviously, this is (at this moment) only a proof-of-concept; the goal is to see if a Service Worker based approach is feasible in the first place. There are other, similar experiments around the same subject, with different design decisions; these should be looked at in the future, too.

DPUB IG Telco, 2016-10-31: TPAC UCR review, DPUB ARIA, Task Force reviews

See minutes online for a more detailed record of the discussions.

UCR document review

There was a new version of the UCR submitted last week, some issues remained to be discussed:

  • issue on manifests and links: the various issues got mixed up after all; it was agreed to separate an issue on addressing constituent resources and the other on mapping.
  • the term “versioning” means different things for different communities; we will use the term “iteration” (which is neutral enough)
  • the first section on security (4.2) will be removed and, instead, the use cases will be added to the relevant use cases in the document (just like what happened with the accessibility use cases)
  • 4.3 is really related to the security model for packaging, and that should move to section 3
  • the order of subsections in section 2 may have to be removed
  • there were some discussions on the use case of 2.2.6 (on Buffy) whether that is the right place for it; agreed to keep it there but reformulate it so that it is not creating a new WP, but filtering it

There were also some mostly editorial issues sent to the mailing list.

Next step

Leonard, Tzviya, and Ivan have the pen, in that order; the document can then go for wider review (hopefully by the end of the week).

DPUB ARIA Review

The current document is ready to go for CR soon, and a wide review is in order. Implementations are needed, see the “Exit Criteria” section in that document.

Task force reviews

A quick roundup on the current task forces:

  • The a11y task force concentrated on the use cases; now that this is essentially done, the other issues are related to additions to WCAG
  • The CSS task force was pending but more volunteers may come to the fore to improve the relationships
  • The Archive task force is a bit pending, may have to be closed down
  • The STEM task force is probably out of steam, to be closed down
  • The Structural Semantics Task Force is behind the DPUB-ARIA work, something that may continue later, albeit the exact form is not yet clear
Posted in Activity News | Comments Off on DPUB IG Telco, 2016-10-31: TPAC UCR review, DPUB ARIA, Task Force reviews

DPUB IG Telco, 2016-10-24: UCR document review

See minutes online for a more detailed record of the discussions.

UCR document review

Most of the discussion concentrated on the second review of the newly organized UCR document. The discussions were based on some earlier comments by emails. The main issues discussed were

  • The accessibility related use cases should be, mainly, spread over the various use cases and requirements, rather than standing on their own. However, it is worth and important to have a separate appendix calling out those to emphasize the importance of accessibility.
  • There were a number of entries related to Packaged Web Publications that, in fact, are generic and not packaged dependent. Those should be moved around in the document.
  • The security related use cases should be collected in one place as a top level section. There was a separate discussion on two use cases to be added under that section (on the integrity of the content and the origin of the document).

There were also a number of minor issues.

Next steps

The next steps will take these comments into account in coming up with a new version as soon as possible.

Posted in Activity News | Comments Off on DPUB IG Telco, 2016-10-24: UCR document review

DPUB IG Telco, 2016-10-10: Post-TPAC UCR review

See minutes online for a more detailed record of the discussions.

UCR document review

Most of the discussion concentrated on the firs “dedup” (for de-duplication) pass on the UCR document that Leonard did in the past few days. That version was in a separate branch (though merged into the main branch since).

The document has undergone major reductions, and was, in general, well accepted by the group. Only some issues were left open, discussed on the call.

Notion of versioning

The notion of “version” turns out to mean very different things for the two communities. Its meaning is fairly loose, and publisher dependent, in the publishing world, in contrast to the software world where its meaning is more strict. After discussion it was agreed that what is needed is to be able to track (in some way or other) any change of a Web Publication; everything else is very implementation dependent.

Referencing the schema.org vocabulary on accessibility

It was agreed that a reference should be made to that work, to reinforce the importance of metadata on accessibility

Definition of a WP

As part of the cleanup the introductory part, that includes the definition, has to be cleaned up. It was decided to use, essentially, the definition put forward at the TPAC meeting in a separate wiki page

Collection of publication

There was a separate use case section about publications that are themselves a collection of other publications. After some discussions it was agreed that this section should be removed; it is largely an implementation aspect that should not impose new requirements on Web Publications.

Next steps

The next steps will include a reorganization of the document, separating the Web Publication and the Packaging aspects. Some new use cases also came up during the github discussions that should be incorporated after the reorganization.

Portable Web Publications Use Cases and Requirements FPWD

How do publications differ from web sites? What are the nuances of publishing on the web and making use of the tools of the Open Web Platform? Do publishers really need more than linked web sites? Yes, we do! Portable Web Publications Use Cases and Requirements provides detailed use cases and requirements from the Digital Publishing Interest Group, focusing on two primary issues. These use cases look at the portability of published works, which allow users to transfer their books, articles, and magazines from state to state and device to device. The document also seeks to define the book or publication as a rightful citizen of the Open Web Platform. Thousands of years of successful history, knowledge and information sharing in easily consumable, producible, and storable formats must be recognized as we focus on the tools of the Open Web Platform and what it means for Publishers, Authors, and Readers today.
We welcome your feedback on GitHub.

Posted in Activity News | Comments Off on Portable Web Publications Use Cases and Requirements FPWD

DPUB IG Telco, 2016-09-12: Preparation for TPAC, UCR Publication

See minutes online for a more detailed record of the discussions.

Preparation for TPAC F2F

These are the important pointers:

The meeting concentrated on some of the details on what should be on the agenda, like

  • WCAG connections
  • CSS Houdini status and work
  • CSS Media queries
  • XSL-FO vs. CSS
  • Changes to be done on the PWP draft
  • Preparation for the meeting on the future of the work at W3C

UCR Document

The FPWD of the UCR document is ready to be published (to be done tomorrow, 13 September). Heather gave some overview on the recent changes (too numerous to list here). Sections have been cut, the section headers are no longer the Requirement descriptions, etc. A number of editorial issues came up and will be added to the github issues’ list for further discussion. There will be a separate session at TPAC on the UCR document.

Posted in Activity News | Comments Off on DPUB IG Telco, 2016-09-12: Preparation for TPAC, UCR Publication