Does *anything* support linear=no?

[Quoth]

This seems still to be the case in Oct 2019. linear="no" has no practical use as it doesn't work on the majority of epub readers and apps, even though this is in the epub2 spec.
The epub3 looks more like a spec for tablets and phones as an alternative to HTML, not IMO a sensible spec for ebooks, because naive publishers/developers will do stuff like fixed layout, video, audio etc all of which are useless for eink. Fixed Layout needs big screens and is better served by PDF.

As a publisher I see no point at all to epub3 over epub2 for real ebooks.

[Doitsu]

Originally Posted by FrustratedReader

linear="no" has no practical use as it doesn't work on the majority of epub readers and apps, even though this is in the epub2 spec.

Actually, the epub2 spec states:

For each itemref, the publication author may specify the optional linear attribute to designate whether the associated OPS Content Document is primary (linear="yes", which is the default when linear is not present) or auxiliary (linear="no".) [...]
Reading Systems are not required to differentiate between primary and auxiliary content, and for the requirements and recommendations given in this section may consider all OPS Content Documents in spine to be primary, regardless of the value of the linear attribute.

[Quoth]

In general a totally optional item a spec is often one that gets left out if any of the following apply:
1) Implementation spec writer can't see the point of it.
2) It's awkward to implement, or more work.
3) Project timescale is ambitious (every SW project I ever worked on in 30 years)
4) The other guys never implemented it
5) It was never used by other people that did implement it.
6) It was implemented, but hardly anyone tried using it, so we never fixed the bugs, doesn't matter as it's optional.

Item (6) does seem to apply to some epub implementations as putting the tag simply moves the item to the end rather than making it only accessible via links in other pages (or possibly the TOC).

So it's in the spec, but in a way that ensures it will either not be implemented or not as intended (like the mpb:section no cross tag). Having it in the spine does sort of make more sense than a fake HTML tag in the content, like mobi has.

[Quoth]

Originally Posted by Doitsu

Actually, the epub2 spec states:

Thanks for wayback link. I wonder why idpf.org is gone, or does Daisy Group have that content?

[DNSB]

Originally Posted by FrustratedReader

Thanks for wayback link. I wonder why idpf.org is gone, or does Daisy Group have that content?

Going on close to 3 years ago, the IDPF website had an announcement that the merger between the IDPF and the W3C had been completed and the IDPF no longer existed as a membership organization and the idpf.org website now existed as an archive. A Google search came up with the following (used cached data):

IDPF Has Combined With W3C
January 31, 2017
On January 30, 2017 the combination of IDPF and W3C was completed (W3C press release: http://www.w3.org/2017/01/pressrelea...nation.html.en), and IDPF has ceased operations as a membership organization. This website is now an archive. For current information about EPUB and more visit https://www.w3.org/publishing/ .

Read more about IDPF Has Combined With W3C

[Quoth]

The Internet is useful.
However, it's not reliable, persistent, very honest, open access, democratic etc.
The wayback machine aka archive.org is patchy and often important pages lack the vital content because of how the original site worked.

Don't rely on bookmarks or so called cloud or a forum if something is important. Do your own backup. Don't waste money or subscriptions on DRM based content that needs a live server.

Geocities long gone. Now Yahoo groups are getting closed.

"wget" (standard on Linux, can be installed free on Windows) is your friend for backing up websites better than archive.org does. Also what assurance is there about archive.org or any service from any big USA multinational?

[JSWolf]

Originally Posted by FrustratedReader

The Internet is useful.
However, it's not reliable, persistent, very honest, open access, democratic etc.
The wayback machine aka archive.org is patchy and often important pages lack the vital content because of how the original site worked.

Don't rely on bookmarks or so called cloud or a forum if something is important. Do your own backup. Don't waste money or subscriptions on DRM based content that needs a live server.

Geocities long gone. Now Yahoo groups are getting closed.

"wget" (standard on Linux, can be installed free on Windows) is your friend for backing up websites better than archive.org does. Also what assurance is there about archive.org or any service from any big USA multinational?

And if you are not careful how you backup the website, you could end up filling up your hard drive with all kinds of useless crap thanks to links to this that and the other thing.

[Quoth]

Absolutely. You need to read wget manual / help. It can limit the madness.

It's got tricky to even READ the content of some sites when you have sane protection like noscript, umatrix or ublock, all of which are better than Antivirus SW for protecting against web based malware.

[j.p.s]

Originally Posted by FrustratedReader

Absolutely. You need to read wget manual / help. It can limit the madness.

In over 15 years of use, I've never been able to get --no-parent to actually work on some sites.