Re: New commit of the Portable Web Publication draft, to be reviewed...

Leonard, some comments inline.

On Wed, Sep 30, 2015 at 4:01 PM, Leonard Rosenthol <lrosenth@adobe.com>
wrote:

> Actually, Bill, the reasons why we are “working on this now” is that EPUB
> 3.x doesn’t address a number of the requirements that have been put
> together by this IG for a “Portable Web Publication” [1].   The
> online/offline situation is just one of them – perhaps the most important
> one, but certainly not the only one.
>

Agreed. and it would be nice if (somewhere) other deficiencies in today's
EPUB 3 (i.e. that keep it from being a full "Portable Web Publication"
solution) were enumerated. I certainly am not suggesting that there are
none (and I'm really not trying to go all "Nobody puts EPUB in a corner" on
this!).

>
> So in my mind, the PWP document [2] should be using [1] as the basis for
> discussions.  Any  reference to EPUB 3 should be as an example of something
> that the community has been using to date that doesn’t meet their needs
> (and perhaps why).   That to me sounds like an appendix.
>

It is far too strong to say that "community has been using to date *that
doesn’t meet their needs*" particularly when this IG represents not so much
of that community and when said needs are in some cases speculative.

And EPUB is not just "an example of something", this IG was created as part
of collaborations between IDPF and W3C and I think it's critically
important both that we not undermine EPUB 3.x adoption and that we don't
give up on the potential for an evolutionary vector. After all, HTML didn't
meet a number of needs for interactive and rich media content, but the
chosen solution was not to start over from scratch ... to the contrary,
it's arguable that a key principle that led to HTML5's relatively rapid
proliferation was to maximize backwards compatibility (something that many
purists were very unhappy with at the time).

You might prefer this group to embark on a de nova effort on a brand new
format, without making compatibility with EPUB an explicit consideration,
and I can imagine there may be some others who may agree with you, for a
variety of reasons, but that is not a consensus position., but I just don't
think it's practical that we

... whatever technologies are chosen to deliver PWP is completely open for
> discussion and debate.
>

That is true but that EPUB exists, is adopted, and that we should not
needlessly reinvent any wheels are IMO not open for debate. The first two
are facts and the last is both a general principle for standards
development in W3C and elsewhere as well as again part of the partnership
between IDPF and W3C that led to co-authorship of the original EPUB+WEB
white paper as well as the current charter for this IG. We want to keep all
our options open, including the very real possibility (some might say
likelihood) that a future PWP format may not be compatible with today's
EPUB 3.x, but that is not at all the same thing as ignoring what has
already been done in EPUB 3, which has benefited from over a decade of
collaboration among key stakeholders and experts in the publishing and
accessibility communities. Even without my IDPF hat on, I believe that
doing so would be a big mistake, particularly as so many publishing
industry stakeholders are not yet part of this IG and since EPUB 3 is still
evolving relatively rapidly. Let's not do an XForms here (unless we have
to, which I think it's premature to agree on).

>
> And on that note, I find the current document draft offering too much
> proposed technical direction, which is VERY VERY premature at this time...
>

On this I agree with you. As I said earlier, I think it would be best if
this document described the ideal goal state rather than how we will get
there from here.

--Bill



>
> Leonard
>
> [1]
> https://www.w3.org/dpub/IG/wiki/Requirements_for_Web_Publication_and_Packaging
> [2] http://w3c.github.io/dpub-pwp/
>
>
> From: Bill McCoy
> Date: Wednesday, September 30, 2015 at 6:42 PM
> To: Ivan Herman
> Cc: W3C Digital Publishing IG
> Subject: Re: New commit of the Portable Web Publication draft, to be
> reviewed...
> Resent-From: <public-digipub-ig@w3.org>
> Resent-Date: Wednesday, September 30, 2015 at 6:43 PM
>
> Hi Ivan,
>
> I would generally be OK to consolidate more detailed references to EPUB in
> one place, feasibly in an Appendix.
>
> However, I am not sure how a section titled "Why work on this now?" can
> reasonably exist in a vacuum from EPUB (as well as other things that exist
> now in the world but especially EPUB since it is the only open standard
> portable document format based on Web technologies, and is widely adopted).
> We want to ground our motivation in practicalities not theoretical
> concerns. And to me one of the key practical motivations to "work on this
> now" is precisely that today's EPUB does not fully address the online use
> cases. Whether or not the fruits of this effort ultimately gets called EPUB
> 4, in a very real sense that's the role it will necessarily end up playing,
> assuming we successfully avoid forking competing efforts which would only
> confuse the ecosystem to the advantage of proprietary solutions. So
> personally I do not see the upside in burying EPUB in fine print at the end
> (at least for those of us who want to see a unified open platform for
> digital publishing).
>
> But perhaps leaving in a briefer mention of EPUB in that motivational
> section would be sufficient, and then all the details about ways and means
> of potentially achieving the requirements, including as they touch on EPUB,
> could be consolidated to one section or Appendix at the end.
>
> --Bill
>
>
> On Wed, Sep 30, 2015 at 11:07 AM, Ivan Herman <ivan@w3.org> wrote:
>
>> Bill,
>>
>> thanks.
>>
>> I fully understand your concerns. On the other hand, it is also true that
>> it is "cleaner" if the technical descriptions, architecture, etc, could be
>> as generic as possible. I was thinking in a somewhat different structure
>> that may be the basis for a good consensus, I wonder whether this could
>> work for you
>>
>> - references from, essentially, the current text to EPUB would be
>> removed, including those parts that you referred to
>> - we would add a completely different section (or appendix) on whose goal
>> is to clarify the relationship with EPUB. That would include, with possible
>> editing, the various remarks on EPUB, completed and/or replaced with the
>> text you propose below.
>>
>> I believe this would make the structure cleaner, and would allow us to
>> make a consistent and clear set of arguments on the relationships to EPUB.
>>
>> WDYT?
>>
>> Ivan
>>
>> PS. I agree with the removal of the EPUB+WEB reference.
>>
>> ---
>> Ivan Herman
>> Tel:+31 641044153
>> http://www.ivan-herman.net
>>
>> (Written on mobile, sorry for brevity and misspellings...)
>>
>>
>>
>> On 30 Sep 2015, at 19:46, Bill McCoy <bmccoy@idpf.org> wrote:
>>
>> Hi, I haven't been on the calls, so apologies if I'm revisiting old
>> ground, and also my comments are not substantive/technical. But, to help
>> avoid fostering unnecessary FUD among current EPUB adopters and
>> stakeholders, I would like to propose some minor word-smithing of one
>> sentence in Section 4.2. In the interests of full transparency, I'm sending
>> my suggestions to the group as a whole rather than just making edits in the
>> source document. Feedback and discussion welcome, as always.
>>
>> The sentence I'm concerned about currently reads:
>>
>>
>> *Although all effort must be taken to keep as much backward compatibility
>> with deployed formats (e.g., EPUB 3) as possible, the requirement(s) of
>> Portable Web Publications will very likely mean a non-backward compatible
>> transition from those. *
>> I have two issues with this:
>>
>> 1. I believe it would be more accurate to state outright that EPUB is not
>> just one of many "deployed formats" but is the primary format for
>> consideration for backwards compatibility in this effort. Proprietary
>> formats like Mobipocket are not really under consideration (and at this
>> point the vast majority of content flowing to Amazon's infrastructure from
>> authors and publishers is in EPUB anyway). Non-web-based formats such as
>> PDF, XPS, and DOCX aren't really in scope for specific backwards
>> compatibility either. So unless someone would like to propose other adopted
>> formats that we should specifically consider for backwards compatibility,
>> it just seems unnecessary to be vague on this point.
>>
>> 2. I would also suggest a slight softening of the "non-backward
>> compatible" part. I think it's important that it not be perceived that this
>> document is the opening salvo of a determined "forking" effort particularly
>> when two of its co-authors are respectively the CTO of IDPF and the
>> co-chairs of the IDPF EPUB Working Group. IDPF understands that the results
>> of work towards achieving the goals of Portable Web Publications may be
>> incompatibilities with today and tomorrow's EPUB, but IMO it is premature
>> to suggest so strongly that this will be the case, particularly as we don't
>> yet know how far tomorrow's EPUB (e.g. EPUB 3.1) can evolve incrementally
>> towards achieving Portable Web Publication vision. Also "very likely" is a
>> judgment call that it seems premature to declare consensus on, particularly
>> without any supporting specifics.
>>
>> As I see it, what we are fundamentally doing with this document is
>> describing a desired end goal, not detailing how we are going to get there
>> from where we sit today.
>>
>> So while it's wordier I would propose the following replacement:
>>
>>
>> *It would be highly desirable to deliver on the requirements on Portable
>> Web Publications in an evolutionary manner that builds on, and is backwards
>> compatible with, the existing EPUB format 3, since it is already widely
>> deployed. However, to fully deliver on Portable Web Publications
>> requirements may not be possible without a non-backward compatible
>> transition. At this preliminary stage, this document emphasizes all
>> requirements envisions for Portable Web Publications without addressing the
>> natural tension between goals of preserving compatibility and fully
>> achieving all these requirements in the most elegant manner. But neither
>> this end-goal emphasis, nor the use of the new term "Portable Web
>> Publication", should be taken as implying a recommendation to definitely
>> create a "de nova" new format that would replace EPUB. Further
>> investigation is required, and the ongoing evolutionary trajectory of EPUB
>> must also be taken into account. *
>> Basically I want to preserve (at least at this early juncture) the
>> notions that a) it might turn out to be possible to fully meet the
>> "Portable Web Publications" goals in an evolutionary manner and/or that b)
>> the community might decide that it's better to sacrifice or relax some of
>> the goals in order to preserve backwards compatibility. By way of example,
>> the ZIP+XML packaging could turn out to be fine for all needs of packaging
>> on the Web, and allow preserving that aspect of EPUB compatibility.
>>
>> BTW I would also suggest removing the note about EPUB+WEB, at this point
>> I think that's historical, and not of interest to the broader audience we
>> expect for this document. In which case the clarification I suggest above
>> could perhaps come earlier in the draft, perhaps as soon as where that note
>> is.
>>
>> --Bill
>>
>>
>> On Wed, Sep 30, 2015 at 4:20 AM, Ivan Herman <ivan@w3.org> wrote:
>>
>>> Dear all,
>>>
>>> as agreed on the telco, I have made some changes.
>>>
>>> *The first and most important is that the address/name of the repository
>>> has changed, due to the change of the term Document to Publication.*
>>> Sorry about that. The draft is at:
>>>
>>> http://w3c.github.io/dpub-pwp/
>>>
>>> the github repo being
>>>
>>> https://github.com/w3c/dpub-pwp
>>>
>>> (it was pwd instead of pwp…). Hopefully, this is the last change in the
>>> URL.
>>>
>>> I have made the following changes, mainly on the terminology section:
>>>
>>> - Changed (Portable) Web Document to (Portable) Web Publication
>>> everywhere in the text and, in particular, in the terminology (glossary)
>>> section[1]
>>> - Added a definition for "Content", essentially taking it over from WCAG
>>> - Changed the term 'collated' to 'aggregated' for the definition of a
>>> Web Publication
>>> - Added a bullet item emphasizing the fact that a Web Publication has a
>>> URI denoting the *whole* set of resources
>>> - Added the font example that we used in our discussion to give an
>>> example for the 'fuzziness' of portable vs. non-portable publications
>>> - I have added a separate section on states[2], but switched to the
>>> protocol vs. file and dropped the online vs. offline. Have added the
>>> typical example table, but also added some notes on the other corner cases
>>>
>>> Section 4 still needs work on adopting the state terminology, but I
>>> wanted to get some feedback before proceeding on that one.
>>>
>>> Remember that the goal we have is to get to a version next Monday that
>>> would be acceptable as a first public working draft. A FPWD is not supposed
>>> to be complete or perfect so, at this point, what we have to decide is
>>> whether the document is "good enough". We can and should work on it further
>>> later, of course…
>>>
>>> All comments, (github) issues, pull request, etc, are welcome...
>>>
>>> Cheers
>>>
>>> Ivan
>>>
>>> [1] http://w3c.github.io/dpub-pwp/#pwp_definition
>>> [2] http://w3c.github.io/dpub-pwp/#state_definition
>>>
>>>
>>> ----
>>> Ivan Herman, W3C
>>> Digital Publishing Lead
>>> Home: http://www.w3.org/People/Ivan/
>>> mobile: +31-641044153
>>> ORCID ID: http://orcid.org/0000-0003-0782-2704
>>>
>>>
>>>
>>>
>>>
>>
>>
>> --
>>
>> Bill McCoy
>> Executive Director
>> International Digital Publishing Forum (IDPF)
>> email: bmccoy@idpf.org
>> mobile: +1 206 353 0233
>>
>>
>
>
> --
>
> Bill McCoy
> Executive Director
> International Digital Publishing Forum (IDPF)
> email: bmccoy@idpf.org
> mobile: +1 206 353 0233
>
>


-- 

Bill McCoy
Executive Director
International Digital Publishing Forum (IDPF)
email: bmccoy@idpf.org
mobile: +1 206 353 0233