why the fuss over epub?

[pdurrant]

Originally Posted by DiapDealer

(JSWolf in 5... 4... 3... 2...)

Good call!

[pdurrant]

Originally Posted by DiapDealer

And yet the exact same epub code to create those few whistles and bells renders differently (if at all) on different platforms.

And the sad fact is that Mobipocket ebooks can render differently on the different Kindle platforms too.

The ebook industry hasn't managed consistent rendering for any format yet (well, except PDF, which hardly counts).

I'd like to see mich tighter constraints on rendering in the ePub standard. No more 'default is whatever the rendering engine likes' bits in the standard.

[Andrew H.]

It's not a *guess,* although getting the number from Amazon would be more precise. But you can get a fairly accurate number from a properly conducted survey.

Originally Posted by JSWolf

Given that Amazon doesn't release sales figures, that 67% is a guess. Please stop posting numbers that mean little to nothing because they are based on a guess.

[DiapDealer]

Originally Posted by pdurrant

And the sad fact is that Mobipocket ebooks can render differently on the different Kindle platforms too.

The ebook industry hasn't managed consistent rendering for any format yet (well, except PDF, which hardly counts).

I'd like to see mich tighter constraints on rendering in the ePub standard. No more 'default is whatever the rendering engine likes' bits in the standard.

I agree completely. Didn't want to sound like I was cutting the mobi/kindle format any slack. Neither renders consistently on different devices/apps. Mobi's proprietary, so only Amazon can tighten the differences between their devices and app renderers. But someone could certainly eliminate some of the ambiguity and wiggle room left in the ePub specs—especially before unveiling a brand new version!

[JSWolf]

Originally Posted by Andrew H.

It's not a *guess,* although getting the number from Amazon would be more precise. But you can get a fairly accurate number from a properly conducted survey.

There was a survey done on MR and it came out that Sony was the #1 reader among MR users who took the survey. These days, no idea what sort of results that survey would have.

Surveys are just based upon who happens to answer them. If say 75% Sony users filled out a survey asking what reader they had and only 25% were Kindle users, would that be accurate to say that 75% of readers in America are Sony and 25% are Kindle with none of the others represented? No, that would be wrong. So surveys in this case aren't worth much. This is why when someone quotes a number that's a guess as fact, it's just looks all wrong.

[JSWolf]

As for rendering differences, since most readers use ADE, the rendering would be mostly similar enough. I think differences will be due to different versions of ADE.

[chas0039]

Originally Posted by JSWolf

Surveys are just based upon who happens to answer them.

Just to clarify, people doing this kind of work are well aware of sampling bias that you mention. Good pollsters work to eliminate this kind of bias and error along with many others.

[JSWolf]

Originally Posted by chas0039

Just to clarify, people doing this kind of work are well aware of sampling bias that you mention. Good pollsters work to eliminate this kind of bias and error along with many others.

But how do we know these pollsters are good enough? I've seen plenty of polls worded such to try to give the answers wanted.

[chas0039]

I am not making any assertions about this poll. I am just stating that people in this line of work are well aware of errors and biases of the type you mentioned as well as many others.

You said "Surveys are just based upon who happens to answer them." While this can be true, any legitimate pollster knows very well how to avoid this problem, and has for well over 75 years.

[JSWolf]

Originally Posted by chas0039

I am not making any assertions about this poll. I am just stating that people in this line of work are well aware of errors and biases of the type you mentioned as well as many others.

You said "Surveys are just based upon who happens to answer them." While this can be true, any legitimate pollster knows very well how to avoid this problem, and has for well over 75 years.

Until we get a definitive answer from Amazon or from exactly how the poll was conducted, it's just a guess and it's meaningless. The problem is that we get numbers tossed at us from all over the place and nobody says how they arrived at these numbers. So until then, it's a guess and a guess isn't going to cut it.

[Terisa de morgan]

Originally Posted by JSWolf

As for rendering differences, since most readers use ADE, the rendering would be mostly similar enough. I think differences will be due to different versions of ADE.

Sorry, I don't dislike ePub but I dislike Adobe a lot. Id ePub is better because it's open source, and later we see that THE PLATFORM is ADE, sorry, or I've missed some point or somebody is moving the original point.

As I say, I don't dislike ePub (I dislike people who think that a new release with new stylish features is more important than a standard dictionary), but I don't care about ADE either.

[kovidgoyal]

From my perspective, supporting mobi in calibre is about 10x harder than supporting epub, despite the fact that mobi has almost zero features that epub doesn't have and epub has about 1 million that mobi doesn't.

And to backup my claims with numbers: The calibre MOBI plugin currently has 8174 lines of code, while the epub plugin has 1211

I, for one, will dance for joy when mobi dies.

[joenunya]

Originally Posted by RainingLemur

There's the old stand by of DRM. That usually sparks a pretty lively debate.

HEY!!! Are you saying DRM is good? You WANT DRM? HERESY...YOU MUST BE STONED!! take that hopwever works best for you

{just trying to help with the flow in the thread }

[afv011]

Originally Posted by kovidgoyal

I, for one, will dance for joy when mobi dies.

Amen to that!

[TimW]

Originally Posted by afv011

Amen to that!

If you're waiting for the death of Mobi, you will probably be waiting for a long while. The chances are very good that there will never be a universal book format. If not, does it really matter? Not really.