Ubuntu Likely Won't Appear On Phones Until 2015

[DrYak]

Samsung doesn't need Canonical they will ship their own OS on Phones in 2014 called Tizen.

And Jolla has already shipped with SailfishOS (another Meego/Meamo-derivative, just like Tizen), running on Wayland - my neighbour at work is actually using one.
And as you can expect from former Nokia people, this phone looks like a finished product, and it does actually work.

Contrast this with canonical, who both suffer from a case of NIH-syndrom (the whole Mir debacle) and completely lack any experience in shipping actual hardware.

My reasoning is simple: Downloading a free iso and installing it on a computer, quite likely next to an existing OS, is one thing. Shelling out money for a new phone is quite another.
A little story: When Nokia came out with their "internet tablets", the N770 and N800, I was like "whoa, duuuudes, put that system on a phone, put it on a phone! and I'll buy it". The system was Maemo. And Nokia did eventually put it on a phone, the N900. Did I buy it? Nope. Why? I couldn't justify shelling out big bucks when the phone I already had at the time served me well.

I think it's not only a case of price, it's also a case of timing.
When the N770 arrived there were something rather new: have a full blown computer that hold in you pocket!
Of course people would ask for more...
By the time the N900 (which still wasn't actively marketed by Nokia and more considered as an evolution of their experimental platform) and N9 (the first actual Maemo-powered phone to be sold as a phone with all Nokia marketing), the world has changed a bit:
2 years after the N770, the iPhone was out, and 1 year after that the first hardware running Android (HTC dream) launched commercially too.
The N900 (which its launch still targeted at enthousiats) and the N9 (even later) had to compete against rival which were already established by that time.
The masses already had "high-tech phones doing lots of stuff!" (even if those phone aren't powered by full fledged OSes, but cut down version - at that time the masses won't understand the difference nor see any advantage in having a GNU/Linux powered phone, instead of a Android/Linux).
Also these concurrent had big ecosystems of apps, so why should joe 6-pack buy a phone that doesn't run the million of apps out-there, but only something different?

The situation is somewhat different now: although smartphone are now well established as a technology, users starts slowly craving for more (more capabilites, better multitasking, etc.) and the current crop of platform start to show its limitations (Although there are massive effort at Google and Apple in making the latest version of Android and iOS up to expectations).
There could be room for a new competitor having better capabilities and more room for improvement (instead of trying to add another layer of lipstick on the same old turd).
Also, the need to "join" an established ecosystem is now understood: see how blackberry insisted on bringing android app compatibility.

Canonical has shown some creative idea for Ubuntu and Android co-existence, like running Android inside a chroot on Ubuntu and having both OS collaborate. Thus leveraging the full-blown GNU/Linux under the hood of Ubuntu, while having compatbility with the Android eco-system.
But Canonical ended-up incapable of delivering a device, partly because they wasted time and credibility on the Mir fiasco, partly also because they completely lack any experience in shipping actual hardware. (The only way the multi-million crowdfunding campaign would have succeeded is if they'd used the money to buy an established hardware company).

Meanwhile, other like Jolla have come and completely eaten Canonical's lunch by providing professionnal-grade hardware (it's former Nokia guys, they know their stuff), that runs a full blown GNU/Linux disto like SailfishOS (and it runs on a modern Wayland server, no less), and has good eco-system support:
- it has a Dalvik VM and is compatible with Android Apps (lessons of N900 and N9 learned - Jolla HAS access to the million of Apps out there)
- but it also supports HTML apps (thus can leverage the work done on WebOS, Windows RT, FireFoxOS, ChromeOS, etc. and the speed of development of such architecture)
- but it also supports QtQuick, meaning good interoperability with desktop widgets and the like (KDE Plasma use this too, as well as lots of small projects)
Meaning that users and devs aren't locked into specially crafted software that only target the Meego/Maemo/Tizen/SailfishOS family.

The next point is just my personal view, but I think it has some merit: I don't buy the convergence thing. Desktops and phones are simply too different. On a desktop, I want full-fledged, feature-rich applications. You absolutely can't put such applications on a phone, there's simply no real-estate to accommodate the GUI. So you make simpler applications for the phone. Where's the convergence here? Microsoft did it, but their approach was gimping the desktop. No way would I use Metro apps on a desktop. And I'm not the only one, Win8 hasn't exactly been a success. And there's several third-party start menu replacements, so there's clearly a demand where people want their desktop to be... a desktop. How's that for convergence? It's the "in" thing nowadays, but in practice it has yet to deliver, and I doubt it's even doable.

My personal opinion, is that "convergence" is actually pretty much doable, but has just been done wrong. The "one size fit-all" approach doesn't work: you can't indeed use a desktop and a smartphone in the same way...
But the "dual use" approach is much more likely to succeed. Pioneered by concepts like the Smartbook from Always Innovating, or even some demos from Canonical: in your pocket, it's an android phone, but if you plug it (either into a laptop shell like Smartbook, or to a screen&keyboard like Canonical) and you get a full Linux desktop.

Just the same way as the first smartphones and PDA were criticized for web browsing (browsing just can't work with so little screen estate) but eventually, most website started to provide different themes (wikipedia/google/etc. all look differently when in mobile mode or when looked from the desktop) and the web became browseable from smartphone while still benefiting from desktops, the same would probably happen to next-gen smart-phone: in your pocket they're mainly smartphone and if you try opening a document, you have a small document viewer editor. But once docket you get a full desktop experience. Opening the document will launch a full-featured office suite.
Different front-ends for different uses.
You always carry around your local document and your key to your own piece from the cloud. But you have differnt way to interact with them depending on the screen you're looking at.

The last chance for Canonical to not become irrelevant in the race for "convergence" is to provide convincing dual-use environment.

(currently the Meego/Maemo/Tizen/SailfishOS family is still mainly centered around the "internet tablet" paradigm and hasn't started to develop a "docked" experience. Yet.)

At least, I hope it will be ARMv8-only, if they are going to launch it in 2015. At least use that for marketing, as the "first 64-bit only phone" or something, although I'm not sure it will do much at that point.

I don't think that "64-bits" would still be a strong selling point. Users won't probably notice it (the bit wars is over and not much attention is paid to it since the old "8bits vs. 16bit vs 32bits gaming console" era). And while on the long term 64bits is important and interesting, it won't bring an imediate advantage to the user experience neither.

In short: 64bits is good, but nobody gives a f*ck about it, and nobody will see the difference.

It would be very interesting if it was one of the first phones with the 16nm FinFET Tegra K2/M1 (Denver/Maxwell), since Nvidia will probably still have issues with selling its chips to many OEMs, so that could definitely set them apart, but they need to do it as soon as the chip is available.

That has much more chances of being a game changer. Good quality and high resolution graphics and high frame rates are definitely something that could be noticed, specially if canonical's device has also to be able to run on PC monitors and TVs.

[zanny]

(currently the Meego/Maemo/Tizen/SailfishOS family is still mainly centered around the "internet tablet" paradigm and hasn't started to develop a "docked" experience. Yet.)

Do they have to? I see a world where you dock the phone and the init daemon (I'm using systemd euphemisms here) just stops the mobile stack and starts the desktop stack. IE, stop mer start gdm or kdm. If you are using the same user account data can be app-based shared.

It isn't as whole stack integrated as Ubuntu Phone, but we can get devices running at least Sailfish right now.

[DrYak]

Do they have to? I see a world where you dock the phone and the init daemon (I'm using systemd euphemisms here) just stops the mobile stack and starts the desktop stack. IE, stop mer start gdm or kdm. If you are using the same user account data can be app-based shared.

Yeah, indeed, one they that should be the case.
The only thing is that right now, it doens't have anything like that, it's a tablet-only experience.

Whereas Canonical (and Always Innovating, too) has already had some proof-of-concepts and demos (and even a few fast prototypes of SmartBook for A.I.).

But on the other hand, Jolla is shipping actual end-users right now.

It isn't as whole stack integrated as Ubuntu Phone, but we can get devices running at least Sailfish right now.

Yup Indeed, Jolla went for concentrating on actually having a delivrable device rightnow. And it pays.

They'll have time to jump on the "convergence" movement later on.

(specially since they could benefit from some industry standardisation:
- Smartbook prototypes are built around HDMI.
- Nowadays MHL is gaining traction (a special "smartphone-to-TV" communication which can share the already existing standard micro-USB and normal-sized HDMI connectors)
- Wireless solutions will be hapenning soon)

[akincer]

Yeah, *if*. How can my point not make sense "at all" when its rebuttal depends on an *if*?

Because yours does too -- i.e. "If you AREN'T getting ready to buy a new phone".

You can't carve out one group from a whole and argue it negates the other group which is what your argument is attempting to do. Mine isn't carving out a group and attempting to negate the other. I'm arguing there could be a market whereas you're arguing there isn't. Yours is absolute. Mine is not.

[sarmad]

Do they have to? I see a world where you dock the phone and the init daemon (I'm using systemd euphemisms here) just stops the mobile stack and starts the desktop stack. IE, stop mer start gdm or kdm. If you are using the same user account data can be app-based shared.

That is different from switching between mobile and desktop WITHOUT logging your user out or closing any of the open apps.

[Gusar]

Because yours does too -- i.e. "If you AREN'T getting ready to buy a new phone".

You can't carve out one group from a whole and argue it negates the other group which is what your argument is attempting to do. Mine isn't carving out a group and attempting to negate the other. I'm arguing there could be a market whereas you're arguing there isn't. Yours is absolute. Mine is not.

My argument is that the number of people who are in the market for a new phone at a given moment is small. A lot smaller than the number of people who aren't. Then, Ubuntu users are already a small portion of the whole population. So your market, at the moment UbuntuPhone launches, is a small number of people from an already small group. And I seriously doubt it's big enough to make UbuntuPhone a success. For it to succeed, it'll need to attract a larger population that goes beyond just Ubuntu users. *That* is my argument. That you can't rely on the existing Ubuntu userbase alone.

[DDF420]

My argument is that the number of people who are in the market for a new phone at a given moment is small. A lot smaller than the number of people who aren't.

7 billion people on the planet and in 2013 we had almost 1 billion smart phones sold.

[TheBlackCat]

My personal opinion, is that "convergence" is actually pretty much doable, but has just been done wrong. The "one size fit-all" approach doesn't work: you can't indeed use a desktop and a smartphone in the same way...
But the "dual use" approach is much more likely to succeed. Pioneered by concepts like the Smartbook from Always Innovating, or even some demos from Canonical: in your pocket, it's an android phone, but if you plug it (either into a laptop shell like Smartbook, or to a screen&keyboard like Canonical) and you get a full Linux desktop.

But the problem is, they aren't the only ones interested in convergence. KDE has similar, or probably even more extensive, use of this concept. And they already have a working interface and working hardware shipping this month, and are probably much closer to having their tablet ready than Ubuntu is to having their phone ready. And it has the capability of working on top of existing Mer-based devices, unlike Ubuntu-based devices.

Whether plasma active will succeed or not is not known. But it could very well be a problem for Canonical, since in most (if not all) regards they are further along in the process, have similar ideas taken to probably an even greater degree, and has more in common with existing platforms making adoption easier.