RE: report: iOS9 adds "print to PDF"

Tons of readers support EPUB 3. See epubtest.org.

From: Richard Schwerdtfeger [mailto:schwer@us.ibm.com]
Sent: Wednesday, July 01, 2015 3:15 PM
To: Peter Krautzberger
Cc: Bill Kasdorf; Ivan Herman; Larry Masinter; Olaf Drümmer; W3C Digital Publishing IG
Subject: Re: report: iOS9 adds "print to PDF"


Android's book reader supports EPUB3. I have not run an exhaustive analysis  of it compared to Apple's but iBooks does a phenomenal job. My hat is off to Apple. Also Google books are published in EPUB format.



Rich Schwerdtfeger

[Inactive hide details for Peter Krautzberger ---07/01/2015 02:00:56 PM---> Maybe it's a problem that PDF works essentially ever]Peter Krautzberger ---07/01/2015 02:00:56 PM---> Maybe it's a problem that PDF works essentially everywhere, whereas EPUB hardly does?

From: Peter Krautzberger <peter.krautzberger@mathjax.org<mailto:peter.krautzberger@mathjax.org>>
To: Olaf Drümmer <olaf@druemmer.com<mailto:olaf@druemmer.com>>
Cc: Richard Schwerdtfeger/Austin/IBM@IBMUS, Bill Kasdorf <bkasdorf@apexcovantage.com<mailto:bkasdorf@apexcovantage.com>>, Ivan Herman <ivan@w3.org<mailto:ivan@w3.org>>, Larry Masinter <masinter@adobe.com<mailto:masinter@adobe.com>>, W3C Digital Publishing IG <public-digipub-ig@w3.org<mailto:public-digipub-ig@w3.org>>
Date: 07/01/2015 02:00 PM
Subject: Re: report: iOS9 adds "print to PDF"

________________________________



> Maybe it's a problem that PDF works essentially everywhere, whereas EPUB hardly does?

Well, iOS and OSX ship with a pretty good epub2&3 viewer, iBooks.

> And that PDF could be made to work for people with special needs?

But do they? I don't have access to any samples.

> And how many websites are out there that really use MathML (as opposed to images of formulas with some Alt attached to them).

Many. Anybody who's serious about math and science, really. From STEM publishers like IEEE to platforms like StackExchange, from individual researchers to MOOCs, from educational publishers like Pearson to blogging teachers, from computational software like iPython Notebooks and MatLab to encyclopedias like, well, Wikipedia (thought arguably not yet the default).

> [...] why it's not good enough for 'mobile'.

Because it throws away most of the information; imho, that's fine for printing but not much beyond that.

>  How realistic is it that a sizeable number of users will actually use EPUB3 (including having access to 'real' EPUB3 reading systems, availability publications in EPUB3, and so forth)?

Apple ships a good epub viewer on all products. They also don't seem to care about other platforms, cf. the iBooks Author format.

> why is there no decent, readily available web page to EPUB3 converter at all?

A google search<https://duckduckgo.com/?q=!g+html+to+epub&t=canonical> turns up a few options. Sure they are not perfect, but neither are PDF generators (though arguably those they'll screw up different things). I would say Calibre has comparable (yet different) quality to print-to-pdf from browsers. Perhaps all it takes is an epub equivalent of print stylesheets to improve the situation quickly (though print stylesheets are probably a good start for epub generation, too).

Anyway, I don't have any issue with having a PDF generator around; it just doesn't knock me off my feet.
Peter.


On Wed, Jul 1, 2015 at 8:18 PM, Olaf Drümmer <olaf@druemmer.com<mailto:olaf@druemmer.com>> wrote:
Maybe it's a problem that PDF works essentially everywhere, whereas EPUB hardly does? And that PDF could be made to work for people with special needs? And how many websites are out there that really use MathML (as opposed to images of formulas with some Alt attached to them). If PDF is used on let's say an iPad and the PDF is captured at the same 'form factor' as when it is being displayed on the iPad, I have difficulty seeing why it's not good enough for 'mobile'.

Maybe there should be more considerations of the combination of  relevance AND feasibility? How realistic is it that a sizeable number of users will actually use EPUB3 (including having access to 'real' EPUB3 reading systems, availability publications in EPUB3, and so forth)? PDF does have the advantage of being relatively widely available, serving over 95% of users well enough for all practical purposes. It took PDF over 20 years to get there. Currently EPUB3 is where PDF was ca. 2 or 3 years into its existence. How do we deal with the other 17 years it might need to establish EPUB3 in the same manner?

And: There is one question that really keeps me thinking - and I have yet to find a good / satisfactory answer: why is there no decent, readily available web page to EPUB3 converter at all? Especially if/as EPUB3 could be described as a packaged web page/site… Or am I missing something?  Or is it too early in the game/am I being too impatient?

Olaf


On 1 Jul 2015, at 19:49, Richard Schwerdtfeger <schwer@us.ibm.com<mailto:schwer@us.ibm.com>> wrote:

They should have done save to EPUB3 as it is packaged. As you point out, PDF is not the best format for mobile. Also IBook author can import EPUB3. From an accessibility perspective it is more than just tagged PDF that is important. It is also access to digital math to allow for alternative renderings for blind, low vision, attention deficit, situational impairments, and dyslexic users.

Print fidelity is nice but, today, it is about supporting a broader range of users and also due to the uptake of mobile devices in education inclusive access is much more important.


Rich Schwerdtfeger

<graycol.gif>Bill Kasdorf ---07/01/2015 12:13:27 PM---Two comments: Yes, it's the structure that's the main issue—and the structure expressed in a standar


From: Bill Kasdorf <bkasdorf@apexcovantage.com<mailto:bkasdorf@apexcovantage.com>>
To: Peter Krautzberger <peter.krautzberger@mathjax.org<mailto:peter.krautzberger@mathjax.org>>, Larry Masinter <masinter@adobe.com<mailto:masinter@adobe.com>>, W3C Digital Publishing IG <public-digipub-ig@w3.org<mailto:public-digipub-ig@w3.org>>
Cc: Ivan Herman <ivan@w3.org<mailto:ivan@w3.org>>
Date: 07/01/2015 12:13 PM
Subject: RE: report: iOS9 adds "print to PDF"

________________________________



Two comments:

Yes, it's the structure that's the main issue—and the structure expressed in a standard way (i.e., HTML5). That's also fundamentally important for accessibility. So "Save as HTML + CSS" is way better than an alternative "save as X" imo, unless the "X" is EPUB 3, which would be optimal.

The other point is that unless I'm not up-to-date on this (and I may not be), I would be cautious about Apple's iBooks Author format because at least wrt the use of Author itself, I believe there are restrictions on how those files can be distributed and sold (e.g., limited to iBooks). I would love to be informed that that's no longer the case.

--Bill K

From: Peter Krautzberger [mailto:peter.krautzberger@mathjax.org]
Sent: Tuesday, June 30, 2015 2:59 AM
To: Larry Masinter; W3C Digital Publishing IG
Cc: Ivan Herman
Subject: Re: report: iOS9 adds "print to PDF"

> Serious question: if it was “Save as HTML + CSS” or “save as X” for
> any other X, would you be less sad, and why?

Top of my list would be epub3,  but Apple's iBooks Author format would make sense.

Given the quality of the website-to-epub generators I've encountered, that seems like a much harder problem. But even a non-optimal solution might provide a better experience than a page-sized PDF on small screen. In combination with something like readability/pocket/etc or "save selection", the content could even shine.

> What data would you have in other formats that you don’t have for PDF?

I suppose that comes down to the quality of the files, i.e., whether they are "plain old" PDFs (glyphs on a canvas) or pdf/a or even using Flash/JS/etc to represent more complex content. Assuming it's just glyphs with positions, then it seems to me almost all markup is lost whereas HTML/CSS-based formats like epub and iBA can retain parts of the original structure.

Don't get me wrong, I understand why one would ship a PDF generator (i.e., for all the usual reasons); but it doesn't stop me from wondering if whoever decided that this is a good feature for mobile devices also thought: "but really, we need a better way".

Peter.



On Tue, Jun 30, 2015 at 12:42 AM, Larry Masinter <masinter@adobe.com<mailto:masinter@adobe.com>> wrote:
Serious question: if it was “Save as HTML + CSS” or “save as X” for
any other X, would you be less sad, and why?

What data would you have in other formats that you don’t have for PDF?

Seriously. It’s really hard to get down to requirements.


On 6/27/15, 8:48 AM, "Ivan Herman" <ivan@w3.org<mailto:ivan@w3.org>> wrote:

Me too...

Ivan

---
Ivan Herman
Tel:+31 641044153<tel:%2B31%20641044153>
http://www.ivan-herman.net<http://www.ivan-herman.net/>

(Written on mobile, sorry for brevity and misspellings...)



On 27 Jun 2015, at 16:15, Peter Krautzberger <peter.krautzberger@mathjax.org<mailto:peter.krautzberger@mathjax.org>> wrote:
Just something I came across, https://twitter.com/fakebaldur/status/614794685559742464


Quote: "It’s particularly useful for webpages, since it keeps all the text, and makes it searchable and copyable unlike, say, taking a screenshot."

I admit this makes me somewhat sad :-(
Peter.