Amazon says having their own format permits faster page turns

[Jellby]

Originally Posted by Amazon

We chose a format that we felt would give us better performance and superior ease of use.

No, they chose a format because they had previously bought it. From Wikipedia:

Mobipocket.com was bought by Amazon.com in 2005
EPUB became an official standard of the International Digital Publishing Forum (IDPF) in September 2007

Originally Posted by Amazon

Because we control our own standard, we can develop applications that let customers read Kindle books on the iPhone, Android tablet, iPad, BlackBerrys and PC.

Well... it's not false, but they can develop applications for any open format too, they do not need to have their own format for that.

There's obviously no real connection between "proprietary format" and "faster software", unless it means "faster than others at reading this proprietary format". About whether Mobi is faster than ePUB, it might be, but because it's a simpler format (and designed for simpler devices), not because it's proprietary. The Kindle would be even faster if it read only TXT files.

[MV64]

Honestly, if a page turns faster than I can turn a real page with my hand, I'm fine with it. A 1 second page turn doesn't bother me and I probably wouldn't notice an decrease in .5 seconds.

[screwballl]

Originally Posted by Purple Lady

Well my page turns are .3 seconds. I don't know if the newer kindles can beat that, so of course this is just another excuse.

Yup, I have several formats and only PDFs run slow. Otherwise all the AZW, MOBI, TXT, RTF files that I have used, even large files (2-3MB) still turn the page in almost no time.
Sometimes it takes a touch longer to display the next page of books, than it does to turn the page in the book itself.

[OtterBooks]

No, they chose a format because they had previously bought it.

I am confused. That doesn't seem like the usual order of events when choosing something. Seems like they would choose it first, then buy it. Did they initially buy it for something else?

[bhartman36]

Originally Posted by Jellby

No, they chose a format because they had previously bought it.

You might be mistaking cause and effect. They probably bought it because they planned to use it. Also, as you point out, EPUB didn't become an official publishing standard until after Amazon had made this decision.

Originally Posted by Jellby

Well... it's not false, but they can develop applications for any open format too, they do not need to have their own format for that.

While it's true that they can develop applications for any open format, having their own format gives them control over what features the format supports, and how they're supported.

Originally Posted by Jellby

There's obviously no real connection between "proprietary format" and "faster software", unless it means "faster than others at reading this proprietary format". About whether Mobi is faster than ePUB, it might be, but because it's a simpler format (and designed for simpler devices), not because it's proprietary. The Kindle would be even faster if it read only TXT files.

I have no idea why MOBI is faster than EPUB. It might be because it's simpler. But doesn't that also highlight one of the problems with the EPUB format? If you've got two identical books, and one loads faster on the Kindle because it's in MOBI/AZW format, doesn't that say that the EPUB format is needlessly complex? And (correct me if I'm wrong) doesn't that also mean that a company who wants to simplify the EPUB format for use in their devices can't?

[OtterBooks]

You might be mistaking cause and effect. They probably bought it because they planned to use it.

I'd venture to say it is even more than probable, since they started developing the Kindle in 2004, a year before purchasing mobi. Did epub exist as a format in 2005? If they were primarily interested in proprietary control, rather than the performance of the format, could they not have just used a proprietary DRM with epub? And wouldn't that have been cheaper than buying another company?

[Andrew H.]

Originally Posted by OtterBooks

I'd venture to say it is even more than probable, since they started developing the Kindle in 2004, a year before purchasing mobi. Did epub exist as a format in 2005? If they were primarily interested in proprietary control, rather than the performance of the format, could they not have just used a proprietary DRM with epub? And wouldn't that have been cheaper than buying another company?

Epub did not exist until 2007, and I don't think that there was an epub reader until 2008. Sony only moved away from its proprietary format at the end of 2009 (although its readers may have also been able to read epub before then).

[HarryT]

Originally Posted by OtterBooks

I'd venture to say it is even more than probable, since they started developing the Kindle in 2004, a year before purchasing mobi. Did epub exist as a format in 2005? If they were primarily interested in proprietary control, rather than the performance of the format, could they not have just used a proprietary DRM with epub? And wouldn't that have been cheaper than buying another company?

At the time that Amazon bought Mobipocket, Mobi was THE dominant commercial eBook format, and was the format that pretty much everyone who read eBooks was using. Most people were reading on PDAs of course back then; this was before eInk hit the market.

[OtterBooks]

Epub did not exist until 2007, and I don't think that there was an epub reader until 2008. Sony only moved away from its proprietary format at the end of 2009 (although its readers may have also been able to read epub before then).

At the time that Amazon bought Mobipocket, Mobi was THE dominant commercial eBook format, and was the format that pretty much everyone who read eBooks was using.

Interesting. Thanks Andrew and Harry for the info.

So let me know if I got this right, because I may not have:

-2004: Amazon begins creating a device for a small, nascent market. Though not the first of its kind, they invest in the device and a restructuring of their company to support it to an extent that nobody had done previously.
-2005: Amazon purchases Mobipocket, which they had chosen as the primary format for the product they were developing (along with its supporting products such as mobile phone/PC software), ostensibly optimizing it for this format; a logical format choice at the time.
-2007, September: Two months prior to the Kindles's launch, epub is chosen as the IDFP standard.
-Amazon's venture is a success to the point that for all practical purposes it gives birth the market in its current state, and dominates.
-A handful of other companies capitalize on this success, but choose epub as as their primary format.
-Amazon is criticized for not adapting their device to a different format they claim is inferior for their product, made "standard" after the Kindle's development and then utilized by a small handful of competitors.

[bhartman36]

Originally Posted by OtterBooks

Interesting. Thanks Andrew and Harry for the info.

So let me know if I got this right, because I may not have:

-2004: Amazon begins creating a device for a small, nascent market. Though not the first of its kind, they invest in the device and a restructuring of their company to support it to an extent that nobody had done previously.
-2005: Amazon purchases Mobipocket, which they had chosen as the primary format for the product they were developing (along with its supporting products such as mobile phone/PC software), ostensibly optimizing it for this format; a logical format choice at the time.
-2007, September: Two months prior to the Kindles's launch, epub is chosen as the IDFP standard.
-Amazon's venture is a success to the point that for all practical purposes it gives birth the market in its current state, and dominates.
-A handful of other companies capitalize on this success, but choose epub as as their primary format.
-Amazon is criticized for not adapting their device to a different format they claim is inferior for their product, made "standard" after the Kindle's development and then utilized by a small handful of competitors.

Sounds about right to me, from what I understand...

[Jellby]

Originally Posted by bhartman36

You might be mistaking cause and effect. They probably bought it because they planned to use it. Also, as you point out, EPUB didn't become an official publishing standard until after Amazon had made this decision.

Right. My point was that they didn't choose the format because they thought it was better than ePUB. When ePUB "appeared", they had already made their choice.

I have no idea why MOBI is faster than EPUB. It might be because it's simpler. But doesn't that also highlight one of the problems with the EPUB format? If you've got two identical books, and one loads faster on the Kindle because it's in MOBI/AZW format, doesn't that say that the EPUB format is needlessly complex?

But the two books are never identical (underline is mine). The ePUB version would, for instance, allow user stylesheets. As for "needlessly complex", it depends on what one determines is "needed"; a series of screen snapshots could probably be made to display faster (with the appropriate optimizations), and it would look exactly the same as the book from which the snapshots are taken. Of course, one could argue that the Mobi format is enough and everything further is unneeded, but others won't agree.

[bhartman36]

Originally Posted by Jellby

Right. My point was that they didn't choose the format because they thought it was better than ePUB. When ePUB "appeared", they had already made their choice.

Okay.

Originally Posted by Jellby

But the two books are never identical (underline is mine). The ePUB version would, for instance, allow user stylesheets. As for "needlessly complex", it depends on what one determines is "needed"; a series of screen snapshots could probably be made to display faster (with the appropriate optimizations), and it would look exactly the same as the book from which the snapshots are taken. Of course, one could argue that the Mobi format is enough and everything further is unneeded, but others won't agree.

What I meant was, if you take the same book, with the same content (i.e., the two texts are identical word for word, and font choice for font choice) and the MOBI book loads faster than the ePub book, doesn't that mean that the MOBI book is faster, and that the other things in the ePub book (e.g., the stylesheet) are just weighing it down? If you can achieve the same book with less baggage, why wouldn't you? (Of course, it's arguable that a text with XML formatting without stylesheets is more complex than one with a stylesheet, but for the sake of argument, I'll accept the proposition that the MOBI format is simpler.)

[Jellby]

Originally Posted by bhartman36

What I meant was, if you take the same book, with the same content (i.e., the two texts are identical word for word, and font choice for font choice) and the MOBI book loads faster than the ePub book, doesn't that mean that the MOBI book is faster, and that the other things in the ePub book (e.g., the stylesheet) are just weighing it down? If you can achieve the same book with less baggage, why wouldn't you?

Yes, I understand that. But those two books will only look identical for a given user, a different user will have other preferences, which can be applied to the ePUB book, but not to Mobi. He may prefer blockquotes left- and right-indented, or paragraphs spaced by 0.5em; this can be applied to an ePUB book through user stylesheets (at least in theory, I'm not saying all current readers allow this), but would not be possible with Mobi. More clear is the case of nested elements (a <p class="caption"> inside a <div class="illustration">), which just don't exist in Mobi.

For plain text, consisting of just normal paragraph after normal paragraph, with the occasional chapter break, I grant you Mobi is all you need, and most of the features of ePUB are unnecessary. But for plain text, a plain .txt file would be even faster (and could, with appropriate settings, look just the same as the Mobi or the ePUB equivalent).

Anyway, I very much believe the theoretical performance difference is mostly due to differences in the reading software, more that to differences in the book format, and even then they are negligible in practice.