Even if not locked into an ecosystem, we're still locked in what the app providers support. That includes a lot more companies that do apps, like for :
* games
* trains
* banks
* messaging (Whatsapp-like)
* ...
(For banks, trains and a lot others, mobile websites usually do a great job, and lower this requirement, if web browsing is correctly supported on the chosen platform)

I agree on this common denominator though.

Hi there,

I think this is part of the reason why Neo900 and others got lost. too many conflicting requirements.

I bought my Nokia N900 in 2009 because I loved what it could do, the way it did it and the form factor it was presented in. The wow factor was a bonus. Especially when compared to iphone 3 and later 4.

But I was locked in to Nokia's systems, plus the various apps skype, dropbox, etc,etc.

At the time there was an amazing array of app's being created, many of the wow in themselves.

I then followed a route N9 (x2), back the N900, Jolla, back to N900 and now Galaxy S7. Finally the S7 has enough cool features that it has some wow.

The N900 was retired from main phone duty because most of the app's no longer work due to API's being updated and the browser, once the phones killer feature compared to apple, no longer works well with many sites.

If we demand a closed phone system then there is no incentive for any companies to invest in it, so it will be a niche. If it is a niche, no developers will develop apps for it.

This was the folly (or deliberate destruction) of Nokia. If the N9 supported Maemo rather than Meego then the app community would have had a bigger market, instead they had 2 smaller markets. Jolla tried again with new OS demanding new apps, but tried to offset this with Android support. A cunning move, but annoyingly Android apps on Sailfish are hit and miss if they work, can be installed, or have access to GPS, etc,etc.

I backed Neo900 when it was billed as a Maemo replacement for N900. I withdrew my backing after the Golden Delicious fiasco, and by which time the Jolla I backed was coming.

I fear that the problem will remain, the same as what killed off Nokia in the first place. Without develop support for new apps and constant revision of old favourites anything with unique OS will be a niche product.

If I want to run Linux and do some hacking on an IOT device, my N900 is still the tool of choice, it is still perfectly able to function in that role. However I need so much more from the device than that for it to be my main phone.

What do I want? I would love to see modern hardware version of N900 with oLED display. But I would also love to see developers flooding back to maemo OS to support the device. One without the other for me would be the same as the ZX Spectrum emulator I run on my N900. Simply a trip down memory lane.

Bingo!

One of the reasons Android is so successful is that it is the same Android everywhere. Motorola, Sony, Samsung, Alcatel - they all may use different launchers but underneath it is the same OS. From the developers' point of view this is perfect since they have to build the application only once to reach all Android devices in the world. The user benefits.

Enter the wonderful world of Linux with mutually incompatible distros. Is anyone surprised that developers and users do not flock to Linux in their millions?

N900 tried to make this a bit less painful by maintaining the compatibility at least at the source code level. I could take a 5 years old source for a desktop Linux application, type make and voila, it worked on the N900. Not always very well due to the menu style etc but it worked. The N9 broke this compatibility and then Jolla broke it again. One would think that N9 applications would work seamlessly on Sailfish but no, Jolla in their infinite wisdom had to reinvent the wheel again.

Sorry about that little rant. I could not help it

Honestly I don't think Linux fragmentation is too bad. At this point it is more or less Red hat based vs Debian/Ubuntu distributions, almost everything is based on one of those bases. To solve this Redhat should acquire Canonical and essentially kill Ubuntu. MIR needs to die already. Ubuntu should just exist as a desktop environment as that is the only value it provides. Ubuntu Touch should just be a user Interface for Tizen. Sailfish should also be an interface for Tizen.
The Suse operating system should be acquired from who ever owns it by Redhat.
Accepting the superiority of Redhat and by extension Fedora is the solution to our problems! It would eliminate Desktop and Mobile fragmentation.

But that'd in effect mean that RPM would dominate over DEB, right?
What about all the people who like the admittedly superior debian packeting format, hmm?
Of course to solve that RedHat could change over to using DEBs

> admittedly superior debian packeting format
Did they already implemented some analog of admittedly superior delta-rpm updates format?

Ok, so summarised: No new infos on what the best Handheld gnu/linux machine for 2017 is. A Pity, really.

Not really reinventing the wheel, just using up to date technology which happens to be somewhat incompatible with the old one. I say somewhat because it's only on a phone that it becomes more trouble than it's worth. On a desktop it doesn't really matter as you can run anything with minimal effort and there is enough horsepower for any compatibility layers. The only exception is the one railroadmaster pointed at, which is Canonical (the developer of Ubuntu) reimagining perfectly round wheels as cubes all the time. If they invested their money in helping others fix the real problems with new software, which is sometimes incompatible because it's impossible not to be, I think we would be several years ahead of where we are now. They have lots of money, but choose to spend it on making knockoffs of popular software, with "© Canonical" in the credits. I wouldn't be surprised if one day they forked the Linux kernel so they could call it the Canonikernel.

The biggest difference between distributions is the software they include by default. New microdistributions pop up all the time because there are always some lunatics who think they need to "provide a choice for the people" by making a distribution that does exactly one thing different to the parent distribution just because after an argument on a mailing list the parent distribution (or developers of a particular piece of software) ended up not changing a certain feature, and so the people will be freed from the oppressive chains of the parent distribution. In reality, they waste a whole lot of time trying to keep up with the upstream development, all the while making sure their particular preferred feature still works and evangelising on mailing lists and forums about their heroic efforts to save x from certain doom, only eventually noticing no one really cares about those efforts. But then two more heroes of the people will have risen up to repeat the cycle.

This is not to say people can't disagree, but it would help a lot if those disagreements didn't have a guaranteed end result of split efforts.

Really well written piece with wonderful points that I just have to agree with, thanks!

However; personally I have to confess I don't consider splintering, forking, or wild goose hunts waste at all. As long as everything is kept open nothing people make is ever lost but stays in recach as a resource to be tapped into if needed; it is a richness that just cannot be had in corporate-governance commercial closed source world.

It is pure silliness to think we would fare better if all swdevs started to pull in the same direction, to do that would be to lose the passion that drives you forward!

All hail to chaos and wildflowers!

And therein lies the problem. As long as the main killer of OSS will remain to be seen by the juicemes of this world as a good thing, OSS will never prevail.

OSSers of the world, unite!