There were 4 arrow keys on this keyboard. They were placed on the lower right corner of the keyboard.

So the up cursor was on row 2 and the other ones were on row 3.

No. And all of the development devices had the same "form factor". Again: These were no polished devices. Just the hardware put together so no outlook on the final product possible.

3 rows of keys

So having (hopefully) addressed some specifics... here are my thoughts:

Wow!!

And that is not really for the device - it's for the people!

I've worked for a major UK telco for a while and unless you've been in a similar position it really is hard to understand how much effort is clearly going into being open with Maemo... and when you meet these people in person... that impression just gets stronger and stronger.

However... the exclamation also applies to the device

My impressions... kinda replaying the timeline of my experience....

It's solid but more manageable than the N800. The plastic casing is a bit lumpy and bumpy and you can't really say how smooth it will be to handle. A black theme is going to look very cool (but someone commented that the SDK theme is not going to be on the product).

The screen feel is smooth and precise. The UI is different and I spent many many seconds figuring it out

I was impressed with the responsiveness (though I expected no less) and the sharpness of the image.

The "click outside the box" thing is different but works well on the device... (later I thought it could be annoying if developers start nesting things too deeply). It is really nice to see too; the blurring of the background is very cool and *very* effective as a usability hint. The whole gui experience was great.

There was no other application SW installed on mine (#7) but IIRC the Application Manager had much improved categories

It has a stylus... small but present... so that's good. I don't recall actually using it though.

There's a camera on the front.... and another on the back and flash leds?
I tried to look inside and count the pixels but gave up at 2 million

No stand obvious though... but the final case may provide that.

Keyboard.... as I posted, the keyboard is great. I typed happily on it and the soft keys on the x-terminal worked well. Shame about the tab but I can see the sense to omit that from the hard keys.

At this point (having spent a while moving around the UI) I actually noticed the screen size as I launched xterm. It felt a bit small for hacking in; I think that's because the font was big (pixels) and it reminded me of an N800 xterm with a virtual keyboard. Having thought about it a bit more I did a little test and when I held the N800 at a 'normal' distance (about 50cm) I found that holding the new device a few cm closer (45cm... not exactly tip of your nose!) gave the same perceived screen size... I reckon this is why it worked.

It was odd though ... I had to lie it on top of my N800 to convince myself the screen was actually smaller and it still didn't compute. Having used it I personally do not think the screen size is an issue.

OK.... later on I had the chance to get the Mer version (arm 5) of Qt running on it and that worked well (although I can reveal that you run out of space on the flash when you install big chunks of Ubuntu).

There are some bugs when you rotate the screen (nb there was no automatic accelerometer to RandR link in the prototype) and the scroll and redraw speed was really bad... apparently that has fixed but the demo version didn't have it.

My Mer/Qt application ran nicely on it, looked really good and "just worked".

I note that the discussions around 3D suggested to me there was some internal frustration with the hardware limitations and they'd like to do more; I don't know if this relates to the driver or the HW.

There was no opportunity to look at the functionality of the GPS or camera(s).

Overall I had very little to want to change... I think the only bug I'm going to file is that when I lie it on the table with the keyboard open then I want the screen to wake up when I pick it up

Well, it was very informal and lots of people made comments and I'm 100% sure that they were listened to.

Hardware: There's almost no point feeding back on hardware. OTOH I had nothing to feedback as long as it works.

Software: Immense interest and communication around how to make it better... Verification to us that the SDK UI works. Also suggestions were made (by us) as to how the docs and guidelines could be developed. I personally shared thoughts on API for screen orientation - but that's on the -devel mailing list too.

Other things... suggesting the UI is different enough that videos and galleries would help... I think that was taken on-board.
Again though, when they release API docs ASAP then they aren't finished. We understand that and I'm happy that releasing early docs that aren't polished is much better than releasing perfect docs later.

We also discussed mechanisms about how we could self-support Nokia tablets after Nokia begins to end-of-life them and how this benefits the community and Nokia.

Nokia are making this easy - I have plenty of devices that simply get abandoned and freeze... my tablet won't be one of them.

From twitter

http://twitter.com/PeterMaemo/statuses/1988845545
http://twitter.com/luovanto/status/1989283476

Does the task switcher emulate the fennec model in the slightest way, e.g. you can pull the window over and get stuff (controls) from the side?

I am assuming this is how they are saving screen real estate.

jandmdickerson, what task switcher?

remember, we saw essentially the SDK that you can download and install now with the exception it was running on an omap3 touch based device itself.

I guess it's hard for us "normal people" to realize how far from "standard practice" the Maemo guys inside Nokia went for this developer session. From what I understand they took quite a risk, I appreciate it and hope everyone involved maintains the discipline so they don't get in trouble for that...

First off, thanks all for the views of those devices. I'm an analyst by hobby and day-job, I look for everything in everything (unfortunately).

That being said, what Nokia is doing is a biggie; developing a platform/ecosystem just isn't done in this open of a manner. And honestly, I don't know if many other companies will be able to pull it off (in mobile; in web, there are a few folks that can and have done this).

The impressions in this thread have me second-guessing a likely N97 purchase. Not because I like the IT more, but because I *know* more before I've gotten my hands on it. To that end, Nokia/Maemo has granted this person some considerable good will that will be returned several ways.

I guess the next thing is to continue to sit back and watch. Hopefully, I'll be able to get in on a few more projects here and see things a bit more under the surface. Until then, sessions and threads like this feed the beast nicely.

Thank you.