Actually I think no N9/950 and no Jolla would make the Maemo community stronger as there would be nowhere to go.

And by not loosing many talented people to the "green pastures" of the N9/Jolla many of the Maemo related project that have been talked about but never materialized due to insufficient manpower (updated autobuilder, rewriting more closed stuff, rebasing Maemo, running the PowerVR blog on new kernel, SHR on the N900, newer Python/Qt/Glibc, porting to other platforms, etc.) might have materialised and moved Maemo forward. Many of these projects are still moving forward, but slowly - but they could have been already finished long ago in this alternative past and a lot of miracles could be built on top of them.

Haha - WRONG! Mer has already happened by then - and you should be the one to know it.

I still remember hanging out on the Mer IRC channel back in 2009, testing Mer on my SmartQ 7 chinese tablet (and possibly writing a part of this page), just as some rumors about a mysterious "Fremantle" device started floating around - and you were there too on that channel. And then the N900 came, sucked out all Nokia employed developers working on the Mer project, killed it for all purposes.

So no change there ? :P

So what would I have done if there was no N9 & Jolla ?

Probably concentrated more on the N900 versions of my apps & on desktop versions. Ports to other platforms that didn't get much of my attention might get more of it (BB10) and I might concentrate more on the Android ports.

I might also have got more involved in some of the mobile open OS projects, such as open-Maemo or SHR.

Oh I think that's quite simple in principle - study the current market leaders & find what people hate about them and what are their major shortcomings. then do the thing properly so that people don't hate it any buy from you instead of them. And that's all. No idea about trivial stuff like how to fund it, how much work would it take, etc. though.

I'm actually only half joking BTW - I think there are some pretty glaring flaws in how Android and iOS do things and exploiting them could be a nice attack vector (its actually interesting in itself that you can "attack" someone by doing things properly ) for an entity wanting to change the current status quo.

Also many of these flaws/behvaior patterns are so ingrained that they can't be ever changed, so the "incumbent" just wont be able to react once you do the thing correctly and you can then forever point out how your solution is better and how theirs is flawed.

tl;dr;doing stuff properly in the mobile market
List of random stuff you can do better to kill Android and or/iOS (including some pretty non-realistic items):

• quick/immediate security updates - more important than ever these days
• long term support for your devices - possibly with paid extended very long term support
• everything open source and auditable - also more important than ever due to security concerns + open source becoming the norm at least in some segments
• hard push on open drivers and up to date kernels - even Android vendors are starting to wake up as they are basically drowning in out of tree patches by this point
• no adds whatsoever anywhere in the OS and apps
• no leaking of personal information in default setup, possible opt it "I really want to do that"
• strong policy on handling of user data by apps
• incentives to make as many apps open source as possible for security and auditing reasons
• hardware that's actually durable and not only cheap at all costs
• hardware that can be actually repaired, not thrown away
• possibly extensible & modular hardware & actually selling extensions and modules for the hardware !
• strong aim to make the hardware and/or an abstraction generic enough to make desktop/server style generic distros a possibility on mobile devices