Kobo typesetting...

[graycyn]

Originally Posted by davidfor

And I think that's the thing. Most people using these devices would not know what a ligature is. And if you told them, they are unlikely to care. Unfortunately, that means that a once the it is pretty good, the company is probably not going to do more without a good reason. It will be a case of diminishing returns. Improving the typesetting beyond the current level is probably a lot more expensive than it was to get to this point.

This. You said exactly what I've been thinking. The average person doesn't care all that much. If they did, being locked into justification on Kindle in pre-KFX days and some VERY ugly word spacing for many, many years, would've put them off. It didn't.

If there was customer demand, then manufacturers would be ALL OVER making improvements to compete with each other. But the demand isn't there. It's not that people might not enjoy more pleasing typography were it offered, but they aren't out there badgering for it or complaining loudly about its lack.

There is, for instance, a CSS property for hanging-punctuation. And it has virtually no support, not in readers, not in browsers. Probably will end up deprecated. Yet I'd LOVE to see that in our e-readers. Not to mention better font support for spacing characters, you name it. So much could be done, but probably won't be.

[lensmann]

Originally Posted by JSWolf

Actually, the simplest way to go ePub > KePub is with the KoboTouchExtended driver in Calibre as it does the conversion on the fly when you send ePub to your Kobo. You also don't end up with a KePub copy on your computer.

It depends on the context in which you're doing the conversion. On my office machine, where I'm typically just transferring a couple of academic articles at a time, it's a lot simpler and quicker to use kepubify (rather than launching Calibre).

[sherman]

Originally Posted by graycyn

There is, for instance, a CSS property for hanging-punctuation. And it has virtually no support, not in readers, not in browsers. Probably will end up deprecated. Yet I'd LOVE to see that in our e-readers. Not to mention better font support for spacing characters, you name it. So much could be done, but probably won't be.

Do you know what else has a CSS property that basically no browser (but Safari) implements? Dropcaps. Specifically, initial-letter.

So, instead we have to rely on hacky floats with specially crafted font size, margins, line height, and break whenever a different font is used.

(No one on the web cares, they just use a Javascript solution and call it a day).

[John F]

Originally Posted by davidfor

And I think that's the thing. Most people using these devices would not know what a ligature is. And if you told them, they are unlikely to care. Unfortunately, that means that a once the it is pretty good, the company is probably not going to do more without a good reason. It will be a case of diminishing returns. Improving the typesetting beyond the current level is probably a lot more expensive than it was to get to this point.

Originally Posted by graycyn

This. You said exactly what I've been thinking. The average person doesn't care all that much. If they did, being locked into justification on Kindle in pre-KFX days and some VERY ugly word spacing for many, many years, would've put them off. It didn't.

If there was customer demand, then manufacturers would be ALL OVER making improvements to compete with each other. But the demand isn't there. It's not that people might not enjoy more pleasing typography were it offered, but they aren't out there badgering for it or complaining loudly about its lack.

There is, for instance, a CSS property for hanging-punctuation. And it has virtually no support, not in readers, not in browsers. Probably will end up deprecated. Yet I'd LOVE to see that in our e-readers. Not to mention better font support for spacing characters, you name it. So much could be done, but probably won't be.

I can't say as I'm that interested in improved typography, but it always seems a little strange to me that Kobo puts a lot effort into rearranging the UI and so little effort into improved typography.

I would think that better typography would be more of a selling point then geegaw features. I would think the typical reader just wants to do linear reading.

[graycyn]

Originally Posted by John F

I can't say as I'm that interested in improved typography, but it always seems a little strange to me that Kobo puts a lot effort into rearranging the UI and so little effort into improved typography.

I would think that better typography would be more of a selling point then geegaw features. I would think the typical reader just wants to do linear reading.

I think it could be a selling point as well, but HOW to sell it might be tricky. Just saying 'our device supports better typography' is not enough, since, all those users who have no idea what a ligature is, may equally be clueless about typography.

Speaking to improved readability just isn't going to have the impact that 'NEW and IMPROVED CONTRAST' does. Or battery life or what have you. Getting into things like 'our device supports CSS property x, y and z' isn't wallet opening ad copy, LOL!

And I *didn't* know about the CSS drop-cap property mentioned by sherman. That's another thing I'd love to see well-supported!

[ZodWallop]

Originally Posted by John F

...it always seems a little strange to me that Kobo puts a lot effort into rearranging the UI and so little effort into improved typography.

That's been my feeling for a long time.

[sherman]

I don't think they have the ability to improve typography, at least not without spending lots of development time (and therefore $$).

The problem is that core rendering engines for both epub (RMSDK) and kepub (QtWebKit) have been abandoned by their respective creators. As far as I know, RMSDK is basically on life support, and Qt completely abandoned QtWebkit, to the point they removed it from later Qt5 releases. And there is no (production level) drop-in replacement for QtWebKit.

[rtiangha]

Ugh. I wonder if the Readium engine can be ported. I think Kobo's already using it on their Android app?

[MGlitch]

Of course we’re also assuming publishers would follow these standards at least for the average Joe reading ebooks, since it won’t matter what Kobo does if the ebooks don’t use the CSS for the reader to render.

[DNSB]

Originally Posted by MGlitch

Of course we’re also assuming publishers would follow these standards at least for the average Joe reading ebooks, since it won’t matter what Kobo does if the ebooks don’t use the CSS for the reader to render.

And that assumption may be a bit of a reach. One of the last epub ebooks I edited since to me, it looked like crap, had a single html file with all the styling done inline. The massive preamble included a bunch of browser options which suggested it had been done as a web page and then just stuffed into an epub container. It did look fairly decent in Edge & Firefox but on a Kobo and a Kindle, it sucked.

[GERGE]

I think RMSDK is quite good. It supports kerning and ligatures, its rendering is nice with spacing, justification and hyphenation. Drop caps and such are mostly CSS based anyway.

[ZodWallop]

Originally Posted by sherman

I don't think they have the ability to improve typography, at least not without spending lots of development time (and therefore $$).

The problem is that core rendering engines for both epub (RMSDK) and kepub (QtWebKit) have been abandoned by their respective creators. As far as I know, RMSDK is basically on life support, and Qt completely abandoned QtWebkit, to the point they removed it from later Qt5 releases. And there is no (production level) drop-in replacement for QtWebKit.

Originally Posted by GERGE

I think RMSDK is quite good. It supports kerning and ligatures, its rendering is nice with spacing, justification and hyphenation. Drop caps and such are mostly CSS based anyway.

Yeah, I have no complaints about RMSDK layout. It's Kobo's kepub that needs some love. Kepub has nice features that ePub doesn't have and it is Kobo's preferred format. So I wish they would clean it up some.

Fans here are enabling ligatures, providing better hyphenation and are now working on character spacing. Is Kobo incapable of doing the same work that three or four outsiders have done for free?

[JSWolf]

Originally Posted by ZodWallop

Fans here are enabling ligatures, providing better hyphenation and are now working on character spacing. Is Kobo incapable of doing the same work that three or four outsiders have done for free?

I've been reading an eBook in KePub because I may want to do some image zoom. With 4.33 and my hyphenation dictionary, the better spacing, and webkitTextRendering=optimizelegibility, KePub look pretty good.

I do agree these are things that Kobo should be doing to make the typesetting better. There is one more thing that Kobo should do is fix the bug where if the line spacing is small enough, part of the first line of the next page ends up at the end of the current page.

[davidfor]

Originally Posted by ZodWallop

Fans here are enabling ligatures, providing better hyphenation and are now working on character spacing. Is Kobo incapable of doing the same work that three or four outsiders have done for free?

Of course Kobo is capable of doing it. But, Kobo has to decide what to do for all of their customers. Not just what a few who think they know what is going on. Yes, they could turn on the optimize legibility option for everyone. But, then they would get the complaints about performance. Which have been made in this forum. Or they turn on the performance option and get complaints about worse typography. Instead, they have taken the middle road that is probably going to be fine for most people and given a way for those to care one way or the other to make their own choice.

[ZodWallop]

Originally Posted by davidfor

Yes, they could turn on the optimize legibility option for everyone. But, then they would get the complaints about performance. Which have been made in this forum. Or they turn on the performance option and get complaints about worse typography. Instead, they have taken the middle road...

Looks like I'm not the only one who thinks they know what's going on

Could be what you say. Or could be that Kobo just hasn't bothered. I wouldn't presume to know why Kobo makes the decisions they do. But it would be nice to have typography that matches KFX and RMSDK.