Page numbers in EPUB

[eschwartz]

Originally Posted by JSWolf

But if you have a hardcover, you get one set of page numbers, a trade paperback, a different set of page numbers, and a mass market paperback, yet another set of page numbers. But with the ePub eBook and ADE, you get one set of page numbers and that's it. You do get to know about the page numbers once you've read enough ePub eBooks.

It is based off a specific print edition, and I believe usually the hardcover. But yeah, once you've read a lot using the ADE calculation, you get a feel for how long it is. At least it is consistent.
Although I wonder -- do you then get thrown off by the differences between the ADE count and the books with Page-maps?

As for the time left, that can be wildly inaccurate deepening on how you are reading. Sometimes I read in bed and the time left would be less. Sometimes I read while watching TV and the time left would be more. Getting interrupted without putting the Reader to sleep would give you yet another time. So unless you sit and read the same way in that book, the time left is not always accurate.

Well, YMMV but I usually read while reading and find it to be fairly accurate. Besides, the main idea behind checking page numbers is usually to find out how long you have left to read. The Kindle updates based on your current reading speeds.

It works for me, not saying it works for everybody...

[JSWolf]

Originally Posted by eschwartz

It is based off a specific print edition, and I believe usually the hardcover. But yeah, once you've read a lot using the ADE calculation, you get a feel for how long it is. At least it is consistent.
Although I wonder -- do you then get thrown off by the differences between the ADE count and the books with Page-maps?

I delete the page maps and also all the extraneous ID tags. I don't see why they are needed in a fiction book. Besides, all they really do is make the ePub larger than it needs to be since (IMHO) the page-map isn't needed (and I don't actually want it).

[Notjohn]

Originally Posted by eschwartz

Besides, the main idea behind checking page numbers is usually to find out how long you have left to read

I have a nice copy of Our Mutual Friend (I think that's the title -- the one about the man who made a fortune with a rubbish pile) that my daughter read at Harvard, presumably on assignment. Every twenty or thirty pages there is a penciled calculation, something like this: 534-113=421.

I think she would have loved the Kindle time-remaining feature!

[GHRuffo]

Gentlemen, I would just like to thank you all for the insights and advices, specially to Doitsu. I have tried to use a page list and it did not work. I have even loaded Doitsu's ebook with the page list already built to see if I was missing something on the code, but I wasn't. His ebook, even with a page list, showed the same page in many different screens. So I gave it up, took all reference to pages out and this is it. Thank you all!

[eschwartz]

Originally Posted by Notjohn

I have a nice copy of Our Mutual Friend (I think that's the title -- the one about the man who made a fortune with a rubbish pile) that my daughter read at Harvard, presumably on assignment. Every twenty or thirty pages there is a penciled calculation, something like this: 534-113=421.

I think she would have loved the Kindle time-remaining feature!

Vandalism!

Yeah, sounds like a perfect match.

Originally Posted by GHRuffo

Gentlemen, I would just like to thank you all for the insights and advices, specially to Doitsu. I have tried to use a page list and it did not work. I have even loaded Doitsu's ebook with the page list already built to see if I was missing something on the code, but I wasn't. His ebook, even with a page list, showed the same page in many different screens. So I gave it up, took all reference to pages out and this is it. Thank you all!

Multiple pages is expected and unavoidable. As said previously, digital books are not comparable to paper books. They are reflowable, that means the size of a screenful of text is dependent on a variety of end-user configuration settings.
Your readers will expect pagenumbers (if they exist) to be split across pages.

Anyway, wise choice.

[j.p.s]

Originally Posted by Doitsu

Page numbers require an NCX pagelist whose syntax is explained in the wiki topic that I linked to..

Also each page number needs to be defined in the body text via an anchor whose ID must start with a letter. E.g. <a id="page1"></a>. (You can't just use the number.)

Some time ago I created a simple ePub pagelist test file that demonstrates all of this. (You can ignore the additional files in the archive; they're only relevant for Kindles.)

The above wiki reference and your pagelist test epub both have pagetarget tags with playorder attributes, but your pagelist generator sigil plugin omits the playorder attribute.

Is that because playorder is not required in pagelists?

If something somewhere requires playorder in pagetarget, is it safe to assume that its presence or absence has no effect on kindlegen generated apnx files?

[Doitsu]

Originally Posted by j.p.s

Is that because playorder is not required in pagelists?

According to the IDPF:

NCX documents that do not contain a DOCTYPE may omit the playOrder attribute.

Since the DOCTYPE is optional in epub3 ncx files, playOrder attributes are optional and since my plugin was primarily intended for epub3 books, it omits them.

Originally Posted by j.p.s

[...] is it safe to assume that its presence or absence has no effect on kindlegen generated apnx files?

AFAIK, KindleGen doesn't require playOrder attributes.

[KevinH]

The kindleunpack APNX code unpacks the page information to an Adobe pagemap.xml file. The reason for that was that NCX pagelists were generally not well supported in epub2. In epub3 the nav pagelist not the ncx pagelist is used. Kindle support for converting ncx pagelists is not even part of KindleUnpack.

If you have questions, I would be happy to point you at the appropriate places in the KindleUnpack code.

[j.p.s]

Originally Posted by KevinH

The kindleunpack APNX code unpacks the page information to an Adobe pagemap.xml file. The reason for that was that NCX pagelists were generally not well supported in epub2. In epub3 the nav pagelist not the ncx pagelist is used. Kindle support for converting ncx pagelists is not even part of KindleUnpack.

If you have questions, I would be happy to point you at the appropriate places in the KindleUnpack code.

Thanks, I'll probably be asking questions in the kindleunpack thread when I have some time.

[Notjohn]

By coincidence, I am simul-reading two ebooks that have page numbers. The first is a public domain translation of The Dawn of Day by Herr Nietzsche, the other an advance proof copy of In the Cauldron, a biography of Joseph Grew, US ambassador to Japan at the time of the Pearl Harbor attack. Both have page numbers inserted.

The first inserts the page number in brackets wherever the new page begins, in the form [pg 047]. That is scarcely any interruption at all, and would be even less so if reduced to [047], with equal clarity if the first instance were spelled out [page 002].

The second appears to be a PDF from the uncorrected print edition, and the page number is rather randomly inserted with a space above and below it. Often, therefore, the text line above it is a short line. I don't mind it, but I am accustomed to reading uncorrected proofs, so if I were going to do this, I would follow the Nietzschian example, apparently done by the Gutenberg Project volunteers.

[j.p.s]

Originally Posted by Notjohn

By coincidence, I am simul-reading two ebooks that have page numbers. The first is a public domain translation of The Dawn of Day by Herr Nietzsche, the other an advance proof copy of In the Cauldron, a biography of Joseph Grew, US ambassador to Japan at the time of the Pearl Harbor attack. Both have page numbers inserted.

The first inserts the page number in brackets wherever the new page begins, in the form [pg 047]. That is scarcely any interruption at all, and would be even less so if reduced to [047], with equal clarity if the first instance were spelled out [page 002].

The second appears to be a PDF from the uncorrected print edition, and the page number is rather randomly inserted with a space above and below it. Often, therefore, the text line above it is a short line. I don't mind it, but I am accustomed to reading uncorrected proofs, so if I were going to do this, I would follow the Nietzschian example, apparently done by the Gutenberg Project volunteers.

That's close to asciidoc markup for a link target, e.g. [[p_047]], so converting such a Project Gutenberg text to an EPUB with page numbers would be pretty easy.