Amazon objecting to page breaks

[Tex2002ans]

Originally Posted by jhowell

The statement about two level of nesting is not exactly truthful. I took a look at a book published long ago that I know has three levels of NCX TOC (Legends B005LVO6FS) on various apps and devices running the latest software. Only two levels show on a Kindle Oasis and Kindle for iOS. However, all three levels show using Kindle for PC, Kindle for Android, a Fire tablet, and Kindle Previewer 3.

Thanks for the info. Mind taking screenshots of <h3> and deeper displaying on the various devices/apps?

So hard to get real-life info on a lot of this stuff... since their own documentation doesn't tell the exact truth.

Originally Posted by jhowell

Version 2019.2 of the Amazon Kindle Publishing Guidelines (November 2019) added a new guideline:

Thanks. Last version I had was 2019.1. Guess I'll have to scan through and see what's changed in the latest.

But this 2-levels-deep TOC is bullshit.

For example, I'm working on journals right now which typically include:

• Year
• Volume + Number
• Article Title
• 1-3 levels of headings

Yes, some of those can be condensed. Ideally, you would have:

- Vol. #, No. #.
-- Article Title
--- Heading 1
---- Subheading 2
----- Subsubheading 3

If you didn't want to go full-TOC navigation, you could remove some of the lower subheadings:

- Vol. #, No. #.
-- Article Title
--- Heading 1

that only gets you to <h3>.

... but cutting it down to 2 levels, or trying to condense even more of that information into a single heading, would just lead to actual readability issues:

- Vol. 1, No. 1: Article Title
-- Heading 1
-- Heading 2
- Vol. 1, No. 2: Article Title
-- Heading 1
- Vol. 1, No. 3: Article Title
- Vol. 2, No. 1: Article Title

Originally Posted by jhowell

So, this is a new requirement and they consider it to be important.

Anyway, in anticipation for this TOC-readjustment hell, I recommended an enhancement in the Sigil 1.0 thread.

Hopefully something along those lines gets implemented which would allow you to mass shift headings to other levels.

Originally Posted by Hitch

So, they suggested archiving using The WBM or The Internet Archive,which would give you an archived page link, yes, but it also means, for longer/bigger/more complex books, doing this one page/url at a time...rather laborious.

🖕

... and the more and more of these changes occur, the more out of sync Print<->ebook becomes (which leads to a hell of a lot more headaches).

More Link Rot Rants: In many cases, websites even change their underlying structure over the years. So something like:

Old: examplenews.com/article/123.html
New: examplenews.com/full-title-of-article-is-here.html

or there's all this appended garbage to their links which means archive.org "never archived" the exact original:

examplenews.com/full-title-of-article-is-here.html#appended&garbage=1&gobbledygook&searchterms=author.concept

(Also why when I'm creating ebooks, I try to preemptively strip the URLs to their barebones.)

So it requires looking up each thing in a search engine, trying to find the latest working URL. (And speaking of... paywalls. Now a lot of the times these news articles are locked so I can't even see if it's correct!)

[jhowell]

Originally Posted by Tex2002ans

Thanks for the info. Mind taking screenshots of <h3> and deeper displaying on the various devices/apps?

I don’t mind and will do it when I have some free time. I doubt it will do you any good though.

[Tex2002ans]

Originally Posted by jhowell

I don’t mind and will do it when I have some free time.

Thanks.

Originally Posted by jhowell

I doubt it will do you any good though.

I think I'll be able to find a use for it...

*brainstorming article ideas*

[j.p.s]

Originally Posted by Hitch

So, they suggested archiving using The WBM or The Internet Archive,which would give you an archived page link, yes, but it also means, for longer/bigger/more complex books, doing this one page/url at a time...rather laborious. I can't WAIT to see what self-pubs and authors will say when I start telling them THIS stuff. Oy....

The Internet Archive's wayback machine is a wonderful resource and I'm sure they've poured enormous resources into it, but large portions of it have been lost over time. The gremlins seem to prioritize disappearing pages that I try to find.

[Hitch]

Originally Posted by j.p.s

The Internet Archive's wayback machine is a wonderful resource and I'm sure they've poured enormous resources into it, but large portions of it have been lost over time. The gremlins seem to prioritize disappearing pages that I try to find.

AND...yup, got another one this morning. ONE bloody link, whole new author and they're insisting that she gives them a REASON why she "can't" fix it.




Hitch (who is very, VERY grumpy about this...)

[jhowell]

Originally Posted by Tex2002ans

Thanks for the info. Mind taking screenshots of <h3> and deeper displaying on the various devices/apps?

Here are some screenshots from the book I used for testing, Legends: Short Novels By The Masters of Modern Fantasy (B005LVO6FS).

The Kindle for PC version 1.27.0 (56109) screenshot shows the NCX TOC on the left and the TOC that is part of the regular book content on the right. The NCX view shows all three levels when expanded.

The Kindle for Android version 8.26.0.100 also shows the three level NCX TOC. I didn't do a screenshot of the The Fire Tablet because it looks the same.

Kindle Oasis version 5.12.3 and Kindle for iOS version 6.26 both show only two NCX TOC levels for the same book.

[JSWolf]

So basically, Amazon has screwed up the NCX levels by not making them all consistent. Way to go Amazon.

[hobnail]

Originally Posted by Hitch

I am the first, truly, to agree that the usefulness of links in eInks is limited, but I feel compelled to point out that they--links--are not without some purpose and use.

Hitch

Link rot is a well known phenomenon. To the extent that if I were in your position I would be absolutely haranguing the authors to not put them in their ebook, regardless of whether it's for a tablet or eInk.

The only workaround that I can think of is to use a link to a search for their page, rather than a link to their page. E.g., www.booknook.biz, which goes to google searching for that web page.

[Hitch]

Originally Posted by jhowell

Here are some screenshots from the book I used for testing, Legends: Short Novels By The Masters of Modern Fantasy (B005LVO6FS).

The Kindle for PC version 1.27.0 (56109) screenshot shows the NCX TOC on the left and the TOC that is part of the regular book content on the right. The NCX view shows all three levels when expanded.

The Kindle for Android version 8.26.0.100 also shows the three level NCX TOC. I didn't do a screenshot of the The Fire Tablet because it looks the same.

Kindle Oasis version 5.12.3 and Kindle for iOS version 6.26 both show only two NCX TOC levels for the same book.

Yupperdoodle, this is what we see and I've got those starting to pile up too. And do we have books with three levels? Oh, you betcha your bippy we do. Probably...a thousand, or more? Sh**T!!!

Originally Posted by JSWolf

So basically, Amazon has screwed up the NCX levels by not making them all consistent. Way to go Amazon.

It's the TOC.html and the K4PC/Mac apps that are causing the issues.

Originally Posted by hobnail

Link rot is a well known phenomenon. To the extent that if I were in your position I would be absolutely haranguing the authors to not put them in their ebook, regardless of whether it's for a tablet or eInk.

The only workaround that I can think of is to use a link to a search for their page, rather than a link to their page. E.g., www.booknook.biz, which goes to google searching for that web page.

I'm actually pondering what it would take to just sent up a SANE link-archiving operation. I mean...who sets up a "commercial" link archiving op, and forces you to sit through a seminar, before they'll let you set up an account, even??? I mean...you search ALL over Archive-it.org, and there is not a single post, FAQ, nothing, that explains what you have to do, to get a login. What to do once you have one, sure. But to GET one? I couldn't believe it.

Hitch

[hobnail]

Originally Posted by Hitch

I'm actually pondering what it would take to just sent up a SANE link-archiving operation. I mean...who sets up a "commercial" link archiving op, and forces you to sit through a seminar, before they'll let you set up an account, even??? I mean...you search ALL over Archive-it.org, and there is not a single post, FAQ, nothing, that explains what you have to do, to get a login. What to do once you have one, sure. But to GET one? I couldn't believe it.

Hitch

If you go to http://web.archive.org/ there's some thing on the right called Save Page Now that sounds like it does what you want; does that avoid the hoops that archive-it.org requires?

[Hitch]

Originally Posted by hobnail

If you go to http://web.archive.org/ there's some thing on the right called Save Page Now that sounds like it does what you want; does that avoid the hoops that archive-it.org requires?

For a single page, sure, although upthread, we spoke about the fact that all of us here have seen archived pages on TIA go walkabout.

It's for books with hundreds of links that I fret.I have many non-fiction books with links by the hundreds and even thousands.

Hitch

[hobnail]

Originally Posted by Hitch

For a single page, sure, although upthread, we spoke about the fact that all of us here have seen archived pages on TIA go walkabout.

It's for books with hundreds of links that I fret.I have many non-fiction books with links by the hundreds and even thousands.

Hitch

I still feel like putting links in a book is a big mistake. I suspect that Amazon feels the same way and that's why they're so hard-nosed about making authors fix them.

[Tex2002ans]

Originally Posted by hobnail

I still feel like putting links in a book is a big mistake. I suspect that Amazon feels the same way and that's why they're so hard-nosed about making authors fix them.

Clickable links are one of the enormous advantages of ebooks over print. It's why you have hyperlinks all over the internet, linking directly to other sources.

Originally Posted by Hitch

It's for books with hundreds of links that I fret.I have many non-fiction books with links by the hundreds and even thousands.

Yep. Last year I digitized a history book: 1900 footnotes, 550 URLs.

As the articles above discuss, some % of links naturally die every year... even 5% dead would be 27 links. And that's just one book. Each one would require you to hunt down the article, try to find the original, make sure it's the same, make sure you don't make a introduce new errors, etc. etc.

And in many cases, like Youtube copyright takedowns, who knows if the working link you found is what the author intended.

On Link Rot

Wikipedia also lists some ways to "solve" this problem:

https://en.wikipedia.org/wiki/Wikipedia:Link_rot

but even in their case, they LEAVE THE ORIGINAL LINK, then link to an archived version too. (Which in the case of Non-Fiction, is good practice + may be mandatory depending on your citation style [Chicago, MLA, etc. etc.]).

This solution would be a complete no-go with this new Amazon crackdown because... dead link!!!

[hobnail]

Originally Posted by Tex2002ans

Clickable links are one of the enormous advantages of ebooks over print. It's why you have hyperlinks all over the internet, linking directly to other sources.

Hyperlinks in web pages are trivial to fix since they're on a server somewhere. At least comparatively trivial compared to an ebook that the purchaser downloaded to their tablet, etc. There's probably some software that the sysadmin can run that scans the web pages and sends email when links are broken. Whereas the ebook user won't know until they tap on a link and then they'll wonder what to do.

[Hitch]

Originally Posted by hobnail

I still feel like putting links in a book is a big mistake. I suspect that Amazon feels the same way and that's why they're so hard-nosed about making authors fix them.

With all due respect, I think I can say with 99% assurance that they don't. When we had to undertake that tests that we had to pass, to become authorized vendors to Amazon itself, the test books were in languages that we didn't speak; they were each 300-400 pages, with images on pretty much every page plus x-reffed captioning, and thousands of links (and again, internal links with cross-referencing as well). [nb: you're not allowed a SINGLE mistake on the books. One mistake="you're out." You get 72 hours to make all 3.]

If Amazon doesn't like external links, they have a damn funny way of showing it--and, if they didn't, why wouldn't they just tell the publishers getting these notices to remove the links, rather than all this goat-frackery around link archives?

Hitch