In the end I think it is most likely that the inclusion of a speaker was not financially or logistically economical. The LED lights they have installed within the lower part of the device meant that if they wanted to maintain the attractively small size of the Kindle they needed to make some sacrifices, the obvious two being the speakers and the home button. There simply isn’t space for everything.

I imagine that in the future an improved Paperwhite will be released which could be a little bit bigger but include these lost features, and, we can hope, external buttons to control the brightness of the LEDs, as this excellent review suggested. In the meantime they have either yet to develop a model which could satisfy both these dimensional and the spacial requirements, or they are keeping the technology for an improved version, which they could market and sell as a new model once again (and naturally make more money in the process).

While I’d like to wait and see what future improvements we might see, I’m far too excited to be able to finally read my Kindle at night time, and this alone, as Amazon most probably foresaw, has sold this Kindle hands down to me and to millions of other people around the world. Now I just need to sit out the wait until November 12th, when it will be despatched to addresses in the UK!

When making the decision to drop the memory size, I’m sure Amazon looked at usage statistics and found that a vast majority of users where not even close to the 4GB memory threshold. And I suspect a vast majority were not close to using 1/2 the memory (the 2GB size in the new Paperwhite). And those that were, likely were not actively accessing most of the titles on the device. As such, the cloud based solution works to satisfy the small niche of users with large libraries, especially given the increase of WiFi accessibility today. ANd for those users with large libraries on the move, the 3G version is the offered solution. So I think you personally end up in a *very* small segment of users: a large library that requires regular on device (i.e. non-cloud) access. Even then, is the non-cloud access a true requirement, or a perceived requirement? Does the 3G access not satisfy the use case, albeit at a higher cost? It does suck when you fall into a ultra small niche of users. But it is reality. So you may have to pay a higher cost (to get 3G), but at least there is an option. It *really* sucks to be in a niche of users where there is no solution.

As to why the decision to drop the memory was made, my educated guess is that it was a technical design decision that also had some marketing and pricing considerations. The added light comes at a cost of energy use. A major selling point of the Kindle is long periods or use without needing to recharge. Adding the light, even with its super energy efficient design, adds to the power consumption of the device. Less memory requires less energy to access and maintain. So by robbing Peter (memory) they could pay Paul (light). Given the following two options:
1) Long battery life with 21 years worth (1,100) of books on the device and unlimited books on the cloud
2) Shorter battery life with 42 years worth (2,200) of books on the device and unlimited books on the cloud
I am sure option 1 would be picked hands down by at least 99% of users. And it is a stronger marketing position.

Also, when engineering a device like this, they are working on *very* tight margins trying to hit a *very* specific price point. There is no wiggle room. A $119 device will massively outsell a $121 device with slightly better features. The spread is even greater when talking about a $199 (the 3G ads free version) vs a $200 device. It’s a marketing/psychological fact. Even purchasing on extreme bulk, an extra 2GB would add $5 (likely a bit more) to the device. That may not seem like much. But it would push the device over that psychological price point affecting sales. There is already a belief that Amazon is selling these devices at cost or at a loss in order to make money on selling content (the razor blade or ink jet business model). Given they are selling millions of these devices, an extra $5 loss per unit adds up very quickly to the $20-$30 million level. Even if the extra 2GB cost $1, it is a several million dollar decision.

As for the dropping of TTS, in my humble opinion, that was purely a money making decision. They want to sell the “Whispersync” narration add on to the books you buy. They might have also found that TTS was not a highly used feature and saw it as a way to reduce the device’s cost. I personally found it a bit lacking to be able to listen to and understand a book. Mostly in terms of phrasing. This is most evident with dialog. It was almost impossible to follow back and forth dialog since you didn’t know when one character stopped and the other responded. (Spacing in the text provided such when reading.) The sentences ran together. And while there are better TTS engines out there, they would likely add cost to the device. And improving would undercut the sale of the Whispersync narration add-on. I’d love to see a much better quality TTS on the device. But I doubt that will ever happen.