Kobo Firmware 3.5.0

[mystika1]

I just noticed after installing the new firmware I can change the font size by sweeping the screen. I used to have to install a hack for that. It has been a while since I have used my glo, so this feature may have been added months ago but.... Yay! :-)

[violent23]

Originally Posted by arspr

It happens with at least the Default Font (whichever it is) but I think (IIRC) it always happens in kepubs, even when they have embedded fonts. The used font type just makes the problem more evident or less.

I've just got used to it and ignore it, but it is a bug nevertheless...

It happens with all of the fonts, but there is a fix in Geoffr's firmware patcher called "Custom kepub default margins".

[DNSB]

Originally Posted by JSWolf

Off means do not fix widows and orphans so they are not widows and orphans. Turn the feature off. 2 is on, 0 is off.

Not quite right. 1 is off. 0 is an invalid value. There is no way to "turn off" widows and orphans.

Regards,
David

[DNSB]

Originally Posted by JSWolf

This is ADE we are talking about ADE allows 0. Kobo uses ADE thus 0 is valid.

Kobo uses RMSDK which is not quite the same as using ADE. When I did some testing setting widows and orphans to 0, I got some odd results but for the most part, they were the same as setting widows and orphans to 2.

BTW, CSS Level 2 specified widows and orphans as being set using positive integers. 0 is not a positive integer and hence not an acceptable value. In CSS Level 3, this was explicitly stated but the definition of widows and orphans is still the same.

Regards,
David

[cybmole]

Originally Posted by DNSB

Not quite right. 0 is not a positive integer.

If you want to turn widows and orphans off, set them to 1 -- it's a bit hard to leave less than 1 line at the top/bottom of a page.

CSS Level 3 makes this clear -- "Only positive integers are allowed as values of ‘orphans’ and ‘widows’. Negative values and zero are invalid and must cause the declaration to be ignored." From a discussion in another forum, it appears that for one ereader if the declaration is ignored, the default value (2) will be used.

Regards,
David

strange, I have been using zero in all my books, it worked fine on my old (ADE based) Sony readers & it seems to also work fine on my Aura HD.

For sure a book with both set to 0 displays differently from one set to 2. So where is the "other forum" discussion & what is the "other ereader" please

I always took 0 to mean " dont do any W & O processing"

additional questions/ observations.

1. is CSS3 the correct spec here - have Kobo ever said that they comply with that ?
2. if I take a definition from about.com:
The orphans property defines the minimum number of lines of a paragraph that must be left at the bottom of a page.
I have no problem conceptualising zero lines left - that just means the whole para starts on next page ?
3. (about.com)
orphans: <integer> | inherit

<integer>
A non-negative number of lines to be left.

zero IS a non negative integer

4. ( my interpretation ). IF zero does cause W/O processing to be turned OFF , that is not the same as turning it ON with a value of 2

[PeterT]

I think I would trust the actual CSS specs over a page at About.com. Sure; About.com does say non negative integer, but so often, people use that term to mean the same as positive integer, but from a strict defintion. the two are NOT the same.

From OEIS
Negative Integers
http://oeis.org/wiki/Negative_integers

The negative integers are real integers that are less than 0. For example, –147 and –4 are negative integers, but –0.4181554... and 10 are not (the former is a negative number but not an integer, the latter is a positive integer). The negative integers are listed in A001478, which can be thought of as the "negative equivalent" of A000027

Positive Integers
http://oeis.org/wiki/Positive_integers

The positive integers (also called the counting numbers or the whole numbers) are most often what the set of natural numbers refers to, but not always. (Many authors consider zero to be a natural number, although it was not even a number for the ancient Greeks!)

Zero
http://oeis.org/wiki/Zero

Zero is usually not considered to belong to the set N, of natural numbers, as it was not even considered a number by the ancient Greeks and Romans. The number zero is represented by the Hindu-Arabic numeral 0. As a digit, 0 is used as a placeholder in place-value numeral systems.

[cybmole]

i could not find the actual CSS3 spec for widows /orphans, so I quoted what google did find.

but surely any device that interprets zero as " use 2 instead" is not following the spec correctly !
the only sensible interpretation of zero would be " do not apply any widows / orphans processing"

[Rev. Bob]

Originally Posted by cybmole

i could not find the actual CSS3 spec for widows /orphans, so I quoted what google did find.

Funny, I found it easily enough. I even quoted it, at some length, earlier in this thread - just a few hours ago.

And what does "apply no widows/orphans processing" even mean? Spell it out; I bet when you do, it'll come out to the same thing as setting both widows and orphans to 1...as I explained, in detail, in the previous post.

"Turn widows/orphans off" has no logical meaning. You might as well be demanding a spoonful of moonlight, or half a pound of blue.

[Rev. Bob]

Originally Posted by cybmole

but surely any device that interprets zero as " use 2 instead" is not following the spec correctly !
the only sensible interpretation of zero would be " do not apply any widows / orphans processing"

I have already explained two scenarios in which that is not the case:

a. Ignore CSS lines that use invalid values (default value of 2 stays unchanged)
b. Treat a value of 0 as "highest possible value" (results in starting each paragraph on a new page, for reasons I'm disinclined to explain)

[Jellby]

Originally Posted by cybmole

but surely any device that interprets zero as " use 2 instead" is not following the spec correctly !

The spec says that "0" is not valid. If a CSS is not valid, it should be ignored. If it is ignored, the default (or the inherited value, or the previously-defined value) would be used. The default, everyone agrees, is "2".

So in this case 0=2 is a pretty accurate description.

That's what the spec says. A reader behaving otherwise is showing a bug, or an implementation mistake.

[cybmole]

Originally Posted by Rev. Bob

"Turn widows/orphans off" has no logical meaning. You might as well be demanding a spoonful of moonlight, or half a pound of blue.

Of course it has a logical meaning, if you handle me a programming spec, it can say don't write any code to support widows & orphans i.e. code the renderer as if those concepts did not exist..

Just like you would say turn off ( or leave out) support for hyphenation, or for Small Caps - it's just a CSS feature that can be implemented. or not, or implemented as a toggleable preference.

I think you'll find that adobe renderer has an undocumented setting:
adobe-text-layout: optimizeSpeed
that does exactly that, it renders with minimal features, for speed on low powered devices. It causes widows & orphans, and hyphenation to be "turned off"

[cybmole]

Originally Posted by Jellby

The spec says that "0" is not valid. If a CSS is not valid, it should be ignored. If it is ignored, the default (or the inherited value, or the previously-defined value) would be used. The default, everyone agrees, is "2".

So in this case 0=2 is a pretty accurate description.

That's what the spec says. A reader behaving otherwise is showing a bug, or an implementation mistake.

That may be what the spec says but I assert that on every e-reader I own or have owned, ( its a long list),
widows 0 ; orphans 0 ; renders differently to widows 2; orphans 2

The former renders all pages same length, the latter leads to some shortened pages ( giving the false impression that you are on the last page of a chapter, until you turn the page.

Tested with
SONY PRS650, T2, T3,;
Kobo: Aura HD, kobo arc tablet, Kobo app on Kindle tablet.

I don't recall for sure what happens with mobi / AZW conversions for Kindle - but I think Kindle apps & devices do not support any widows / orphans processing

[Rev. Bob]

Originally Posted by cybmole

That may be what the spec says but I assert that on every e-reader I own or have owned, ( its a long list),
widows 0 ; orphans 0 ; renders differently to widows 2; orphans 2

Do you have something against the number 1?

Originally Posted by Rev. Bob

And what does "apply no widows/orphans processing" even mean? Spell it out; I bet when you do, it'll come out to the same thing as setting both widows and orphans to 1...

Go ahead, try it...

[Anak]

Originally Posted by cybmole

strange, I have been using zero in all my books, it worked fine on my old (ADE based) Sony readers & it seems to also work fine on my Aura HD.

That could be right. I also hacked my books setting widows and orphans to 0. The W3C CSS Validation Service always accepted this value. But since a few months it displays an error message (I don't remember exactly when it first had this error message, but from memory at least 6 months ago; since they set the default profile to CSS3 (which was first CSS2.1 and CSS3 was still optional).

If a zero value always were accepted by Kobo readers I don't know, but never noticed that it did not work.
My guess (and just a guess) is that Adobe probably made some changes to the default, mimimum and maximum values that are allowed for specific properties and updated their RMSDK accordingly.

[cybmole]

Originally Posted by Rev. Bob

Do you have something against the number 1?

Go ahead, try it...

well why - 0 works for me - & all I want to do is disable the feature!

why go back through an entire calibre library changing a value that does what I want to achieve.

( the value is set in each book, there is no way to set it /change it on the Kobo devices )

PS I don't bother with a validation services as I'm not a writer; I just like to tweak all new books to use my layout preferences before I settle down to readi them