We test, of course, but it's mainly the programming APIs and SDK. The user space stuff etc., I have the impression Nokia is not that interested us developers to test. Otherwise we would have more recent firmware.

I stress again, we got the devices to develope our own apps. For end user usability Nokia has, I would assume, different people.

I get what you're saying; however I think the whole plot has been overlooked quite a bit.

For instance...

If you cannot test MS Exchange connectivity, then you lose the ability to code in that arena - you cannot code an app for MS Exchange if you use only IMAP, or POP3.

If you don't have access to Wi-Fi 802.11n secured with certificates, then you cannot test that kind of connection and how it might alter your application's ability to communicate out if it makes web service calls.

If you do not have access to a Mac, then you cannot code an app that might overlook the always intrusive hidden files (.Trashes, .DS_Store for instance) once you connect to a Mac and you were to not have the logic that removes or ignores those files.

True story, my Pioneer car stereo plays music via microSD. On the first iteration of the software, they forgot to "ignore" those aforementioned hidden Mac files and you would get the weirdest issue with those files. I had to pull over, mount the microSD in my Motorola Xoom, remove each and every one of those from the root music folder, then replace it into the stereo before having some of the oddest errors and gaps in music.

If you do not have access to a lot of different, real life scenarios, then who are you coding for? Mainly yourself, within a very limited environment. Mind you, I code mostly on OS X, but I have to test on: WIndows XP, Windows 7, Windows Vista, Windows 2003 Server, Windows 2008, HP-UX, Ubuntu, Red Hat Fedora, iOS, Android, PlayBook (QNX) and most of the ecosystems, networks and/or connecting devices, workstations and protocols therein to complete my QA run to ensure I hit the target on those environments.

Not saying that's what you should do. But if you limit what you can and cannot test, then you are implicitly hoping that everybody has a similar network, device, setup, or predicament as yourself. Once you step outside of that, then what?

I've seen folks have problems with certain routers with their N900 around here. Some routers, unfortunately, aren't as "nice" in regards to how they handle DHCP, QoS, or other important protocols that might inhibit access to the internet.

Not saying that you should test everything. But once your environments get too narrow, your QA is also as narrow. And it might end up being a situation where if you need connectivity, a particular writing speed down to your memory card (say you ignore Class 2 cards, don't specify it and you assume we all have Class 6 cards like you and there's an error... then what?) or some other IO read/write or web connectivity necessity that automatically excludes anybody that does not have the same equipment as yourself... that's my point.

Test in more environments, have a true user case study, do proper QA and expand into situations that might/might not raise problems later.

That's my point. Not to test the machine(s) for us, but expand what you should test the moment you have to access the web. A nice "Might not work with 802.11g routers" with a piece of software is better than not knowing, having errors and we feel stupid because we have countless other things that can go out to the internet on our jacked up network just fine... just your app is a problem and we find TMO, or your e-mail and scream, curse, froth at the mouth unnecessarily when all it takes is more testing.

That is all I mean... sorry for the long response.

As individual developers, there is no way that we can hope to approach the exhaustive testing environments of a commercial software organisation. That's why we have community QA (extras-testing). We don't have that for Meego-Harmattan atm, but those of us with N950s can of course test each other's work using our own environments and report bugs. There are also some non-developers that have N950s, and I would hope that they will get involved in testing and reporting bugs, in addition to providing us with those lovely unboxing videos for a product that nobody can buy.

And I have quite the problem with that. Oh well... perhaps it'll be different next time.

There was a program for commercial developers also. So there are some commercial entities that are using N950s. I don't expect that they will respond to any questions here, though. That's the flipside of the coin. I am happy to see N950s in the hands of people that have been active in this community. I can always get an iPhone if I want something that 'just works TM'.

The "Just Works™" part... LOL

In an ideal scenario, I totally agree with your post, it'd be great to be able to do more testing. But you do realize that most of us "developers" with N950 do so on our spare time, and do not make any money out of it.

How am I to find a 802.11n router if I don't have one at home? How am I to test if network connections work in my app through all the GSM/CDMA/3G/4G operators in the world? I don't have the budget, nor the time to have a proper QA setup. The best I can do is to test in my own environment, get my apps as ready as possible, and when the N9 is available to the general public, to be responsive in fixing bugs that will (not might, will) happen. The N950 just gives a little lead-out time to devs so that people buying the N9 on the first day have a few apps to play with.

At the core of my statement(s), that's what I've been saying all along.

This isn't an ideal situation as far as the community device program goes in that one regard. It was merely an observation. One I've actually come to regret making because all was "well" before I stated anything it seems.

Don't panic. Carry on.

Oh I'm sure voice chat will be integrated, that's pretty clear.
However 1-1 or multi-user video chat is still very murky.
There's already a thread for it at the MeeGo forums.
But so far no clarity from anyone... sigh... oh well.

There is? How the heck did they get one?
I'd hope they're still doing their utmost to contribute in whatever way they can.

Exactly.
And IIRC Nokia's also developing a raft of apps internally, & alongside 3rd parties.
Long before these N950 dev. programs commenced...
Albeit probably scaled-back significantly, since the change in strategy

The Ovi Store beta for the N950 just released today and it's got a depressing amount of apps available...

At first I was really excited about developing for the N950, but I've come to realize that it only supports QML for interface designs. Having to learn a new scripting language so I can rewrite my Maemo 5 apps for the N950 is just frustrating.
But I'll keep on going with a steady pace.

And back to the N950 vs. N9... I have no idea which one will be better as I haven't been able to use the N9 yet.
I have to say that although the hardware keyboard is nice. The slideout mechanism really sucks in comparison to the N900.
It's great that the screen is lifted a little, but to slide it out takes quite a bit of effort. It just never feels fluent and easy to do.