Possible bug. Strange page numbering in ACCESS (sideloaded kepub)

[davidfor]

Originally Posted by arspr

Sure, but how do you tell to other user where it is?

"Chapter 5, about halfway in" or "Chapter 5, paragraph starting with ...". Back in the pbook days, I wouldn't assume they had the same edition of the book. I would assume even less with ebooks. Add in the ability to search the text, and saying "Page x" isn't as needed.

Of course I wouldn't have "fixed" ADE numbering. I would have just adopted a "fixed" version of it in the Kobo ACCESS rendering (where they have already adopted a different system BTW). So no greater confusion, just the same one we have right now: two systems. I just think a per chapter and relative numbering method is not a good idea.

And one of the common questions is "how do I know how much more in the current chapter". So, a per chapter numbering is desirable for some people.

Yes, I can choose but it is an all-or-nothing decision. Like I said about statistics, I don't really care about page numbers. And my turning point to ACCESS is based (or not) in other prominent features.

BUT despite of that fact, I think the current page numbering adopted in ACCESS is really pointless and quite worse over all than the original RMSDK/ADE one.

And there is the thing: It is your opinion and preferences. As I have pointed out, others don't agree. But, you do have some choice. Unfortunately, the choices carry other features that you might not like.

[DNSB]

Originally Posted by arspr

A joint answer to both of them.

Koboish numbering is dynamic, relative, based over your current book layout. ADE/RMSDK is static and absolute based on whatever you want (character count, file size, compressed file size, they are all more or less equivalent).

Actually, the ADE/RMSDK algorithm uses the compressed size of the chapter files. Every 1024 bytes of compressed data is one page with any leftovers in a chapter file rounding up to the next page (1024 bytes = 1 page, 1025 bytes=2 pages) with a fudge factor added for DRMed epubs. And I would disagree that character count, file size or compressed file are are more or less equivalent. Depending on the compression algorithm, you can get wide variances in the compression ratio.

An epub with a lot of inline styles will appear to have more pages than an epub with most of the styles in a stylesheet. One public domain cookbook I modified went from 270 pages to 112 pages according to the Adobe algorithm since the original did not have an embedded stylesheet, depending on 200+ lines inside <style></style> tags at the beginning of each recipe/chapter file and two or three <span></span> tags at the beginning of each paragraph.

The only advantage to the ADE/RMSDK page numbering in my opinion is that it is consistent on identical epubs -- if I am reading a ebook on a Kobo mini, Aura HD, iPad or computer, the same ebook shows the same page numbers. No worries about font size, line spacing, margins, etc.

Regards,
David

[speakingtohe]

Originally Posted by DNSB

Actually, the ADE/RMSDK algorithm uses the compressed size of the chapter files. Every 1024 bytes of compressed data is one page with any leftovers in a chapter file rounding up to the next page (1024 bytes = 1 page, 1025 bytes=2 pages) with a fudge factor added for DRMed epubs. And I would disagree that character count, file size or compressed file are are more or less equivalent. Depending on the compression algorithm, you can get wide variances in the compression ratio.

An epub with a lot of inline styles will appear to have more pages than an epub with most of the styles in a stylesheet. One public domain cookbook I modified went from 270 pages to 112 pages according to the Adobe algorithm since the original did not have an embedded stylesheet, depending on 200+ lines inside <style></style> tags at the beginning of each recipe/chapter file and two or three <span></span> tags at the beginning of each paragraph.

The only advantage to the ADE/RMSDK page numbering in my opinion is that it is consistent on identical epubs -- if I am reading a ebook on a Kobo mini, Aura HD, iPad or computer, the same ebook shows the same page numbers. No worries about font size, line spacing, margins, etc.

Regards,
David

That was the first thing I liked about my Kobo mini. The same page numbers as my Sonys and ADE so I could switch readers easily.

I'd like to have a goto page number, but don't care all that much. In 60 years of reading I have never told a friend to go to a certain page in a fiction book or had one tell me. I/they just say it is a great book, you'd like it etc.

I do recall teachers saying "class go to page" or "we are starting on page", or "read pages x-x" so no Kobos in the classroom I presume

Helen

[DNSB]

Originally Posted by speakingtohe

I do recall teachers saying "class go to page" or "we are starting on page", or "read pages x-x" so no Kobos in the classroom I presume

One of my great-neices is enjoying a university education. The textbooks for one of her courses are ebooks. What her professor for that course tends to say is read the first three chapters on the Krebs cycle leaving up to the student to find the chapters. For specific bits and pieces, he gives the first words of the section and the students are expected to search for them. A generation for whom Google was always there seems to have no issues with that procedure.

Regards,
David

[speakingtohe]

Originally Posted by DNSB

One of my great-neices is enjoying a university education. The textbooks for one of her courses are ebooks. What her professor for that course tends to say is read the first three chapters on the Krebs cycle leaving up to the student to find the chapters. For specific bits and pieces, he gives the first words of the section and the students are expected to search for them. A generation for whom Google was always there seems to have no issues with that procedure.

Regards,
David

I'm sure I'm equally sure that there are more than a few old-school teachers at all levels in the education system, who will be adamantly opposed. You may say fire those teachers who are not up to helping their Grade one students search with all of their various devices but we know that isn't going to happen and I don't really think it should.

As an aside, what University is your Great niece attending? Last summer I worked with a 24 year old student attending the University of Victoria and majoring in English. She has never used an ereader and looked a bit aghast when I offered to lend her one of mine. She has a laptop and an iPad so is not a complete technophobe, and very Google literate, just totally not interested. Currently she is working her way through the classics and is fully aware that she can get most of them free, but despite the fact that she is paying her own tuition and has loans to pay, is just not interested and seems to feel that she is in a majority group in this regard at UVIC at least among English Majors.

How do your great niece's professor's deal with that I wonder?

Helen

[arspr]

Well, as a lot of people seem to prefer per chapter and relative numbering, I've told myself "Well, DO try it for a time, maybe you are just not used to it and that's all".

OK so I've dissabled Full Book Page Numbers and then I've discovered ANOTHER related issue in this case with the navigation bar. And, as I will try to explain, it is also related to the support of nested ToCs.

The issue arises when you have html files which include several "subchapters". (Normally each chapter in a book is a different html file. But it is also completely normal if one file contains several chapters, usually because they are "subchapters").

I explain the issue with an example. I'm currently reading a long book called "Las Legiones Malditas" (title translated to English: The Cursed Legions, and yes, it's a historical novel about Romans). This novel is split in books and then in chapters. (If you prefer, think about "The Lord of the Rings" and its six books and then its chapters).

The "Las Legiones Malditas" epub has the following structure:

• Each book is a different html file.
• In each book, its title is a <h1> heading.
• Each chapter is included in its parent book html file. There are no different files for each chapter.
• Each chapter title is a <h2> heading.
• toc.ncx is also built following this structure. It's a nested ToC with 2 levels. The parent <navpoint>s link to books and the child <navpoint>s link to chapters.
• Just in case it's important somehow although I suppose it's fully irrelevant. Of course, child <navpoint>s, (the chapter ones, the h2 ones), need anchor links as they refer to positions in the middle of the html file. (So they are all like <content src="Text/Legion-7.xhtml#toc-anchor-1"/>). But parent <navpoint>s, (the book ones, the h1 ones), could have gone with just a link to the html file as they are always in the beginning of the file. But nevertheless, they also use anchor points.

Well then the incoherences I get are:

• In the footer of the book I get the known "Chapter whatever - Page xx of xx". I intentionally say chapter, because I do get the h2 titles (well telling the whole truth, the <navLavel> of their associated ToC <navpoint>).
• But in the navigation bar, if I press << or >> I don't go to the next chapter but to the previous/next book (which is the previous/next html file). Am I the only one who sees a little incoherency here?
• And this situation is worsened if I disable the undocumented and not-officially supported Full Book Page Numbering. Because in the Kobo default page numbering (per chapter), the navigation bar ONLY covers the current chapter. Do you see the problem? There's no way to jump to the next/previous chapter, because << or >> jumps to the next/previous book and the navigation bar is limited to my current chapter.
• I've told a little white lie. You can actually jump to the next chapter through the Table of Contents. But as Kobo flattens it, you have a mess of entries. (In this book with just two levels of hierarchy, I already have TEN pages of ToC).

[PeterT]

Remember... Kobo designed the solution they use for ACCESS reading for content that THEY AND THEY ALONE release; not for material that has been converted by end users using a best guess reverse engineering of their epub format.

Yes this might be an oversight in their handling of multiple chapters in one file but seeing we do not (and likely will never know) their formatting rules, calling it a bug is misleading.

[arspr]

Originally Posted by PeterT

Remember... Kobo designed the solution they use for ACCESS reading for content that THEY AND THEY ALONE release; not for material that has been converted by end users using a best guess reverse engineering of their epub format.

Yes this might be an oversight in their handling of multiple chapters in one file but seeing we do not (and likely will never know) their formatting rules, calling it a bug is misleading.

I haven't called it a bug but an issue...

Nevertheless, unless they have other ways to make kepubs work fine, (methods which we do not know), this is "serious" trouble for Kobo. I mean, at least in common epubs, every time you have a new html file you have a page jump. Because of that if you want a new chapter BUT you don't want a page jump, you need using the structure I've told where each file has several chapters in it.

Maybe Kobo can avoid that structure with some kind of untold trick, or they can make that structure work with some kind of untold setting. But it doesn't look really promising, although as you say we must be cautious.

[davidfor]

Originally Posted by arspr

I haven't called it a bug but an issue...

Nevertheless, unless they have other ways to make kepubs work fine, (methods which we do not know), this is "serious" trouble for Kobo. I mean, at least in common epubs, every time you have a new html file you have a page jump. Because of that if you want a new chapter BUT you don't want a page jump, you need using the structure I've told where each file has several chapters in it.

Maybe Kobo can avoid that structure with some kind of untold trick, or they can make that structure work with some kind of untold setting. But it doesn't look really promising, although as you say we must be cautious.

Sorry, but you missed PeterT's point. A kepub is not an epub. It might look like an epub, and appears to share a lot of the specs. But, Kobo define those specs. Unfortunately, they haven't published those specs, or at least nowhere that I have seen them. The specs might be that there is a one-to-one relationship between TOC entries and files. We don't know. The conversion that we have is based on guesses made when looking at DRM free kepubs from the Kobo shop.

[DNSB]

Originally Posted by arspr

Sure, but how do you tell to other user where it is?

Last time, I wanted to tell someone where something was in an ebook, I just told her to search in the book for the string "If this gets any more cloying,". At least that was how I pointed a fellow David Eddings fan to one of the repetition of phrases found in his work. In this case, in Magician's Gambit and The Sapphire Rose. Given I was reading on an Aura HD and she was reading on a Kindle Fire, any relationship between the page numbers we see is impurely coincidental.

Originally Posted by arspr

BUT despite of that fact, I think the current page numbering adopted in ACCESS is really pointless and quite worse over all than the original RMSDK/ADE one.

Given that page numbers in any ebook border on useless, it could be argued that pages that reflect the amount of text on a page as they do in a dead tree book are about as relevant as page numbers based on an algorithm that has nothing to do with what is seen on the display.

Regards,
David

[DNSB]

Originally Posted by speakingtohe

I'm sure I'm equally sure that there are more than a few old-school teachers at all levels in the education system, who will be adamantly opposed. You may say fire those teachers who are not up to helping their Grade one students search with all of their various devices but we know that isn't going to happen and I don't really think it should.

Oddly, I've seen quite a few primary (K-3) students who have been helping teachers with technology. If nothing else, the kids are not scared of breaking something.

Originally Posted by speakingtohe

As an aside, what University is your Great niece attending? Last summer I worked with a 24 year old student attending the University of Victoria and majoring in English. She has never used an ereader and looked a bit aghast when I offered to lend her one of mine. She has a laptop and an iPad so is not a complete technophobe, and very Google literate, just totally not interested. Currently she is working her way through the classics and is fully aware that she can get most of them free, but despite the fact that she is paying her own tuition and has loans to pay, is just not interested and seems to feel that she is in a majority group in this regard at UVIC at least among English Majors.

How do your great niece's professor's deal with that I wonder?

She's attending UBC and is planning to follow her mother into dentistry. She is a technophile so has no issues with using ebooks but there are some students in the class who are not as fond of technology and have gone for the deadtree alternative. On the other hand, the professor had no qualms in using ebooks for the last couple of years. Considering the traditionally high cost of deadtree textbooks ( a tradition which most of the textbook publishers seem to be upholding for ebooks), the cost of a ereader is not a big chunk of the book budget.

Regards,
David

[speakingtohe]

Probably you can read material as you wish in many schools these days. I just have a different impression, and of course I wish it was like that way back when. I got whacked on the knuckles a few times for reading something different even when it was the same book and I had already finished the part we were on.

Helen

[arspr]

Originally Posted by davidfor

Sorry, but you missed PeterT's point. A kepub is not an epub. It might look like an epub, and appears to share a lot of the specs. But, Kobo define those specs. Unfortunately, they haven't published those specs, or at least nowhere that I have seen them. The specs might be that there is a one-to-one relationship between TOC entries and files. We don't know. The conversion that we have is based on guesses made when looking at DRM free kepubs from the Kobo shop.

What I missed is not PeterT's point but my English lessons... You don't seem to have understood me so I re-explain (and I also copy my own original post in your quote).

I had already thought about what you are saying (each chapter must have its own file in Kepubs, and each file must have its own ToC entry too). But if that is the situation, and if there's no other hidden/unknown/Kobo private/undocumented feature, that situation is a clear STEP BACKWARDS from our old, well established, well known common epubs. Because, and the book "Las legiones malditas" is a good and real-world example, in either kepubs and epubs, a new html file always means a page break (unless, in kepubs, there were a hidden option to make it otherwise). So if the ebook author/editor DON'T want a page break before each chapter he is already forced to put them in the same html file. But then this situation leads to a lot of cross troubles and side effects in ACCESS, unless, I repeat, we are missing something hidden about kepubs.

But, do you really think that's the situation (missing info about kepub and ACCESS features in this particular issue)?

Of course that's always a possibility but if I were to create an enhanced epub format because I feel it's good for my device and shop, (a brand new Kobo epub format), I suppose I would have copied ALL the existing features epubs already have and then I would have added the new things I thought interesting. So if something that works in epubs doesn't work in kepubs, of course, maybe the reason is that it has to be done in a different (and unknown, unpublished) way. But I'd really bet money on the alternative possibility of something being broken during the "enhancement" process.



And now I'm going a bit offtopic. And sorry if I don't express fine myself. English is not my mother language and my English knowledge is what it is...

So maybe it's because of my said level of English, because I'm missing some info, or because I'm just dumb. More over, I would really like to think that the fault is in me. But in the little time I've spent in these forums, (since I got my Aura HD), I'm getting the feeling that there's some kind of fear about directly saying something-doesn't-work-in-kepubs/ACCESS. It's like if you are afraid of Kobo getting angry about that kind of posts and then completely dropping sideloaded kepub support. It's something in the air, an unwritten rule, about forum users somehow being forced to say the same sentence in a something-doesn't-seem-to-work-in-kepubs/ACCESS,-but-please-please-please-don't-get-it-wrong-because-Kobo-is-fantastic-and-PROBABLY-the-problem-is-something-I-did-wrong way.

Of course, trolling about Kobo should not be admitted. Posts about ACCESS-is-rubish should be always avoided, but I really think that Kobo should be really pleased when people post the troubles their software or devices have (or seem to have), they explain them, and post a way to make those troubles arise, (even if they happen because of the user's fault).

Maybe my way of speaking (of writing) is a bit harsh (and maybe it also happens because of my English knowledge level), but I always try to explain step by step what the problem is. And I only post what I find while reading my real life books. And so far, I don't say that ACCESS is bad, (I'm currently on ACCESS, so I do think RMSDK is worse), but nevertheless ACCESS shows some flaws. I've detected at least:

• The extra spacing around dashes and other symbols (... at least) when justifying texts. (It can look fine in some situations but it looks horrible in other ones).
• widows, orphans, "page-break-whatever: avoid;" not working.
• The issues posted here which involve Page Numbers (per book or per chapter), ToC and navigation bar.

Of course they can be caused by:

• I am doing something wrong.
• TRUE bugs in ACCESS.
• Things that have changed and should be made in a different way inside kepubs.

And of course Kobo can take the following approaches in either of them:

• Ignore them. Spending any attention in them is worthless.
• Fix them (if they were a bug or just an undesired behaviour).
• Explain them to the world, (if they are kepub changed behaviours from the standard epub), if they thought that allowing the users know how kepubs work is interesting for their business (which is selling ebooks, but also devices which can read books from other places, enhanced by the ACCESS renderer).

But if Kobo is not informed about the issues ACCESS suffer (or seems to suffer), they cannot take ANY decision...

[PeterT]

TL DR

[arspr]

Well, as it is a complex cross effect I didn't expect it modified, and of course it isn't in FW 3.1.0

Kobo default relative per chapter numbering + Several ToC entries in the same html file => Unreachable chapters through navigation bar (at least in sideloaded kepubs).

(And of course unnested ToC to make things worse).