Amazon's a pretty big SW company. I wouldn't be surprised if they started their fork with 2.1, but will bare no resemblance to it. Amazon may have their own internal Android fork built up now and may not merge back with the Google version of Android.

I don't think Amazon wants Google to control their destiny. And considering what Google has done in their 3.x branch (none of which is very impressive), Amazon may well believe they can out execute Google. And considering how few Android tablets have shipped, Amazon could really quickly dominate the Android tablet market -- and then be the one calling the shots.

I honestly don't think Amazon will take that risk - their changes are going to be cleanly designed such that they can pull Google's AOSP code, merge their changes and crank out their own updates. I can't see Amazon developing 2.1 code base without merging Google's bits from newer releases - there is just no way they are going to be able to catch up with all the stuff Google and co. are putting in Android - they will never have that kind of expertise unless they hire people from Google/Android team which I find somewhat far fetched.

Keeping up with the releases from 2.1 and going to 4.0 is going to be a huge pain. And that will be required for compatibility sooner or later.

Nothing will happen. Kindle Fire users will continue to read their books, watch their movies / tv shows and browse the web as before. This is being marketed as a Kindle for a reason.

The GUI will be completely Amazon-built, so the the version of Android probably won't matter all that much to end-users.

Since this is all speculation, we'll have to see when the device is unveiled.

If amazon makes their tablet "smooth" then that will be an achievement independently and despite of android - not due to a specific android version.

And before the android-fanboys crawl out of the woodwork: The "next version" is always supposed to finally solve the lags, by virtue of some awesome software-fix or raw dual-core power.

However, just today I used an SGS2. The latest "next version", dual-core and all. And guess what: It lagged.

There's a more than reasonable reason:

Perhaps Amazon's project froze that portion of the code at some point during 2010. They're shipping v1 of a brand new product based on heavily customized code. At some point, you need to eliminate moving targets if you're ever going to ship.

Amazon's use of Android here is almost an implementation detail. They want to sell you a Kindle so you'll buy content. They're not trying to sell an Android tablet. As a result, I think they're going to sell a ton of them. Enough to make Apple sweat a bit. Very exciting.

Corresponding edit:

>[Edit] Consider JIT (2.2+) and low pause GC (2.3+) that they will be missing. Also consider Flash support which is also 2.2+. Then there is competition from the likes of Archos and you can take their $299 Honeycomb 3.2 tablet and run everything Amazon has to offer on it - Kindle, MP3, Instant video etc.

"No wireless. Less space than a nomad. Lame."

Which is in my opinion a much better idea than trying to compete on price, features AND software with the iPad. Get a couple clear differentiators - price, Kindle bookstore, multimedia content, etc - and actually compete.

I agree completely.

But this isn't a tablet. This is a Kindle. That's why it's the "Kindle Fire" not the "Amazon 7" Tablet Buddy Pro S 1.0." It's going to be exceptionally good at doing Kindle things: reading books, listening to music, streaming movies – and paying for the privilege.

They're going to sell a shitload of these things. They know how to make a great experience around content.

I was surprised too.

http://kellishaver.tumblr.com/post/10646083786/my-ipad

That's not to say it won't sell well - brand and integration might make it work and Amazon never discloses how many Kindles they sell - but it's hard to argue Apple is going to sweat. B&N may be but not Apple and not any other Android Tablet maker.

The game is not about specs anymore. It's about experience.

Amazon has banked a lot of credibility with regard to making great portable hardware and streamlining the content consumption experience. You're seeing this the wrong way around.

People aren't going to buy this because they want a tablet. You and a few hundred thousand other people (technophiles) are the only ones who care about "tablets" that aren't the iPad.

People are going to buy this because they love reading books and love watching movies. And oh, what a convenient and neat way to accomplish that task! What a great gift! etc.

It's going to feel good to use. If that didn't matter to Amazon, it would be running the latest OS they could slap on it, it'd have a spec sheet a mile long, and it would cost a nose more than the iPad.

But that's not their game. They're coming at it from the opposite direction.

And fucking finally – Apple will have a competitor in this space who isn't a complete joke. I'm very, very excited to watch this unfold.

Amazon has good specs for the hardware there - it would not have altered their proposition drastically to ship modern OS on it along with support for Email client and many other experience enhancing things that it comes with - UI acceleration, more app compatibility etc. It's nearly free for them. At $300 price point it behooves them to make it a full tablet along with a reader and media player. Without that it's just not a good value in my mind.

Amazon's customizations are purportedly deeper and better then Honeycomb. Proof will be in the pudding of course but if you were doing a significant fork you can't just keep restarting over and over to incorporate everything Google is doing too.

But let's put that aside - think about compatibility from user stand point. Google's app store is still the main Android app store and is likely to be so going forward. If Google and its partners start pushing out newer releases that Amazon fails to incorporate in their 2.1 tree - users are going to miss out on apps - when they have a 2011 era dual core hardware that they paid $300 for.

Again if this was $100-$179 it would be OK - $300 is what makes this not attractive.

Prove it.

Walk us through the level of customization Amazon has done here, explain at a high level the goals and tradeoffs they sought in forking Android, and show how the product would have shipped on-time, for the holidays, by making the simple change of completely replacing the operating system they'd already spent so much time to customize.

I really don't think you're hearing me, here. This isn't an "Android tablet." This is Amazon's Kindle, which will succeed on its own merits, and just happens to have some deep technical underpinnings in Android.

This isn't about the specs. This is about taking some raw material and shaping it into something that meets their goals. They've done that. Can it be even more impressive? Always. And I'm sure at some point they'll merge in a bunch of the technical junk that only 0.1% of their users care about, in the furtherance of the experience.

But I think, out of the box, they're already going to be heads-and-shoulders above their "competition," such as it is.

You seriously think Apple would have sold as many iPads if they sold it with iOS 1.x version of the browser and no email client, purely based on experience? I don't think so.

So again - our debate here is about price, features and experience. You can change the order but I think excluding any one or more thing at $300 is not really a good thing.

And you and me can argue to no end that it's not Android or it doesn't really matter that it's Android - but Amazon has an "Android" App Store and they sell apps through that and people are going to be expecting to see some apps that they won't either due to them not being compatible or not on the Amazon app store or in the best case just not working good enough due to running on an older OS with deep 3rd party customizations.

And upgrades being nearly free for Amazon - It's Google's code, they just have to rebase and deal with kernel/hardware bits just like every other vendor. The customizations probably are all Java and easily mergeable atop newer AOSP releases - if they aren't then that's bad news already. But let me be clear that even if it wasn't nearly free - as a user I don't really care - I want a reasonably secure and updated OS with no artificial limitations if I am paying $300.

Rest assured, nothing based on Android 2.1 is going to do that.

Seeing as how Ice Cream Sandwich is meant to better unify the APIs, I don't mind if they skip Honeycomb as long as they do get on track eventually.

[Edit: This was on my phone. Sibling response from stonemetal provides interesting counterpoint.]