Jaakko Roppola, senior designer at Jolla, writes:
I get asked this a lot so I did a post about it.
Simple really.
A Sailfish application has a much higher UX potential than any other platform counterpart. The whole operating system is designed around an unobstructed and efficient use of applications. What you as a user want to do.
That’s all well and fine, and we all know why native applications are superior to less-than-native counterparts – which in the case of Sailfish comes down specifically to Android applications, which it supports quite well. The reality, however, is that these reasons are not even remotely enough to draw developers of native applications to Sailfish.
Early this year, I wrote a comprehensive review of Jolla and Sailfish. Since then, a lot of people have been asking me to revisit that review, and go into the current state of the platform. It’s something that I’ve been wanting to do for a while now, but I’ve been putting it off because to be honest – there is very little to tell.
The general conclusion of the review was that Sailfish was a good operating system considering its age, with a comprehensive user interface that was a joy to use, and that was both fast, smooth, and intuitive. Being a new platform, its biggest issue was of course the lack of third party applications – but even there, the platform got off on a good start with a few high-quality applications such a WhatsApp client, a great Twitter client, a barebones but decent Facebook client, and a few others. For a platform that was only a few weeks or months old at the time, that was a great running start.
Sadly, even though we’re almost a year down the line, the state of the platform is still pretty much exactly the same. The operating system itself has improved even further, and continues to do so at a decent pace. Every time I boot my Jolla, Sailfish delights me with its intuitive and smooth, one-handed interface. Between the review and now, we’ve seen like 10 proper operating system updates, and each of them have improved the operating system in noticeable ways. It’s nowhere near as complete or full-featured as Android, iOS and WP, but it will certainly cover users of those platforms just fine.
As far as third-party applications go, however, the situation is – and let’s not sugar-coat it – dire. The applications I highlighted in my review and again a few paragraphs above are still pretty much the only proper Sailfish applications today, with the note that the Facebook client hasn’t seen any development in months and is, as far as I can tell, abandoned. Other than that, virtually every time I get my hopes up when I see tweets about “new” Sailfish applications, it’s yet another Android application that also works on Sailfish.
Of course, we all knew this was going to be the hardest – and largest – piece of the Sailfish puzzle. However, I did expect more than what we have now. I’m sure the lack of support for paid applications plays a role here, discouraging more professional, non-hobby developers from joining in on the fun as a side-project. Whatever the cause, we’re still looking at a third-party applications landscape that isn’t much better than what I saw back in January.
It could be that there’s a lot of stuff going on behind the scenes to get developers interested, but so far, we haven’t seen much sign of that. I’m obviously not going to write off the platform or anything like that – the operating system is too good and fun for that – but progress on the application front is sorely, sorely needed.
Right now, my Jolla spends most of its time in my device drawer, only to be booted up when there’s an update or when I’m bored. Sailfish deserves more, but I’m not sure how they’re going to get it.
I’m the developer of the Lighthouse process monitoring app for Jolla and I can only agree with your conclusion.
Jolla has built and missed it’s own boat. It’s a really sad story and I can’t even tell what caused them to go to sleep mode. I guess they tried hard to get into the eastern markets or something.
There has been 0 activity for the past 1-2 months, all the new regressions are still there (like the bug where CPU activity is reported wrong) and nothing new and exciting is happening. I have lost all interest in putting my time into developing for the platform. It’s still my main phone, but given how it’s limiting me to 2G only here in Canada I don’t see using it for much longer.
You mean from Jolla? I think there was a long holiday break in the month of August in Finland. After that, there was supposed to be a big update in September, but it was pushed back a bit due to, among other things, the recent Bash vulnerabilities.
I’ve been itching for my first big Sailfish update.
So, you’ll be selling it then?
Unfortunately it takes more than a few good ideas or good people to bootstrap an entire mobile ecosystem.
Here’s me in 2013:
And that seems to be what’s happened. It takes a lot of money to spin the fly wheel so to speak, and patience, and time. Microsoft, who is beyond well capitalized is having a hell of a time cracking this particular nut.
While some of you have gone all “LOL METRO” as the reasoning why MSFT is having trouble in mobile, I think the overarching reason is that it is a tough market to get into. Especially if you’re late. Especially if you’re trying to be different.
So kudos to the Jolla team for getting as far as they give, I wish them nothing but the best. I installed the Jolla launcher on my N5 and was blown away by how nice it is.
Yes, Microsoft still seems having trouble filling their store with quality apps. However, when I talk to people trying to build apps for Microsoft, they say the same things:
1. They can’t figure out how to do this or that. The API seems underdocumented, contradictory, or just plain confusing.
2. The ad API doesn’t seem to work. Nobody wants to develop apps if they can’t do some kind of advertising, preferably Google’s. Charging money for apps seems like a thing of the past.
3. The programming model(s) seem to be endlessly confusing. .NET? Win32/64? DirectX? HTML5? Silverlight? Half of those, if they’re not dead, smell funny. Android and iOS have simple models and tools (Java and Objective-C).
I’m sorry that something that sounds so cool as Sailfish isn’t moving forward. My own main interest is Firefox OS, which still seems to be moving fast and furious. Unfortunately people in the USA don’t know about it because it’s not available. But their latest info speaks of “15 countries with 4 hardware partners, 7 operators and 16 launches.” The programming model is … HTML5, making it easy for people to create apps easily. I can’t tell if there’s any money in apps yet, the app-to-market time is pretty short, especially if you use engines like Construct 2.
I think the main problem with Sailfish is that there isn’t room for many more players. I think Firefox OS is smart in aiming at markets that aren’t yet dominated. #1 in Ecuador might sound silly, but it’s a toe-hold, and Sailfish doesn’t seem to have even a toenail-hold.
That’s unfortunate. I mean, I’ve found some parts of it to be under documented. At worst though its been a matter of digging through a Sample App. Doesn’t happen often, at least to me.
dev.windows.com and MSDN in general have pretty good documentation.
Absolutely not true. There’s nothing wrong with the Ad API, there’s even support for AdMob.
WinRT is the underlying platform. That defines the types, the sandboxed environment, the capabilities, etc.
DirectX is a COM API, so its native code. WinRT is a variant (in some ways) of COM.
.NET sits on top of WinRT and interops pretty well with it. They share a common metadata format, so you can write WinRT types in .NET and expose them to non .NET environments (HTML5 and C++).
This is why WinRT is the unified programming model, and why there are so many buzzwords.
Win32 really only exists in the sense that some APIs are made available to ease porting of native libraries. Silverlight only exists for pre WP8.1 apps.
Not only that. WinRT is actually quite similar to Ext-VOS, the genesis of .NET, when they were still thinking of doing everything native.
http://blogs.msdn.com/b/dsyme/archive/2012/07/05/more-c-net-generic…
With the going native wave, Microsoft is just going full circle.
Thanks for laying out a roadmap. I’ll pass it along.
Just promise me they won’t change it all again!
I just saw some headline saying that Satya says that the Office API is the most important to the future. The summary for that is easier:
1. Underneath almost everything is COM and OLE.
2. .NET is glued onto that. Loosly.
Anyway, myself, I prefer HTML5, which is somewhat supported by Windows. I used to do assembler, C, and COM.
Why do you say loosely? WinRT’s metadata format is the same one that .NET uses precisely to enable a better mapping between them. Compositional inheritance was added to WinRT to map better to .NET semantics as well.
If you look at the interop for .NET, it’s actually quite lightweight:
http://msdn.microsoft.com/en-us/magazine/jj651569.aspx
And its actually faster to interop between C# and WinRT components than it is to interop between C++ and a WinRT component due to the way the JIT optimizes code. Its hardly bolted on.
Confusing only for wannabes that don’t care about reading documentation.
There are enough Build sessions, MSDN articles and Channel 9 videos explaining when to use what.
Sounds like a lot of work! Apple, Android, not so much. Firefox OS, hardly any work at all!
Yeah, say that to anyone who’s tried the unintuitive mess that is Interface Builder. Or had to pour through Android documentation.
Come on, this type of thing exists to varying degrees on every platform.
Jolla DID solve the problem of not having many applications: by having very good Android application support, and by making Android applications extremely easy to install (they’re included in the Jolla store).
Is it bad to run Android applications on Sailfish? Why?
You say that the WhatsApp and Twitter applications are good, but that the Facebook application is poor. Does any mobile operating system have a good Facebook application? (Serious question. I seem to remember hearing that all Facebook applications are lousy and that they expect you to just use the mobile version of facebook.com…)
What functionality is missing on Sailfish that you can’t get from either a native or Android application?
As for me, I chose not to install any Android applications on my Jolla mobile. It does everything I need with the native applications.
In the past year, the places that the Jolla mobile can be officially purchase has expanded to all of the EU, Hong Kong, India, and a few other countries. Have there been any reports on how many mobile devices they’ve sold? They only need to make enough money to keep their small company going. I like to think they are, but I have nothing to back that up.
Have a handful of android apps installed myself, (unfortunately) Skype being the one I run the most. Quite the battery hog.
I have said this before and I will say it again…
It is way to hard to build your OS/app ecosystem share just by selling phones. I own a Nexus 5… Why is there not a official build for that phone? A lot of people install 3rd party Android ROMs on dozens of different phones. I think many of them would give Sailfish a try.
You need to increase the number of people using the OS, whether you are looking to sell apps or push ads. Trying to market a phone and a unproven OS is to damn hard. Their phones aren’t even available in the US!
I own a One Plus One and I would install Jolla on it for sure if there was a ROM available…
If a platform provides an emulation API for one of the major competing platforms it will never get anyone targeting their native API.
As consequence, it will loose any appeal for both developers and users to switch to it.
This has never succeeded.
This is a valid point. If Qt would become popular as a cross-platform toolkit, then people might put in the effort to make native sailfish apps there as well.
Unfortunately, I don’t really see that happening.
They still have a lot of catching up to do.
For example, only the upcoming 5.4 version will support Android native controls, and you need to use QML for it. Not C++ Widgets.
http://qt-project.org/wiki/New-Features-in-Qt-5.4
Currently if you try to use a file picker for example, you will get the desktop version.
I’d rather use Xamarin/AppMethod than Qt at this point. They have been taking way too long to come up with a mobile cross platform toolkit.
Xamarin also has its own set of issues.
Personally I would rather use C++, as it is available in all vendor SDKs.
But it surely feels like a stepchild on Android.
Well, it all starts with device availability. I really would have liked to buy one, but they are hard to find. And besides, I am not a big fan of the design of the device. Or just release a version that runs on a nexus 5.
I am really bored with ios and android and wanted something else.
Now I have just bought me a blackberry z30. What a joy to use. A great device relying on gestures for user interaction a lot like the jolla does. Android apps run great, so I’m not missing anything. And the keyboard is magnificent. But now I’m getting off topic
I bought one and really like it, but I can’t deny that the hardware is a drawback. It isn’t really that inspiring at the end of 2014. Comparing it to my LG G2 just makes me wish they’d make sure I could run sailfish on that instead.
Next time I can’t help buying new hardware, I might also opt for blackberry if I’m still feeling adventurous.
What concrete apps do you miss ?
I mean currently we developers are a little bit binded by the strict rules of the harbour jolla store and missing of some api elements like for example a standard file open dialog.
So almost all devs are now waiting for Qt 5.2 to arrive.
On the multimedia front we are waiting for gstreamer 1.x as 0.10 has lots of problems with certain codecs.
When it comes to structural changes there were discussions about paid apps, other third party repos (which are controlled in contrast to openrepos) and so on.
So pretty much its just a matter of time.
With getting their Jolla Phone to other countries and parts of the world more and more users and also potential developers are getting involved.
So I guess there will be more applications in the near future.
But its true that premium apps (a.k.a. apps known from other plattforms) won’t come to it as sailfish os first needs to get a wider user base for them to recognize a need for it.
And of course there are some who will just say: Use our android app it works fine with the jolla anyways.
Edited 2014-10-07 22:12 UTC
For me the biggest issue is that it’s still not fully open source. I don’t care about junk like Whatsapp.
I don’t think so. The main barrier now is not the lack of applications, but the lack of Jolla’s push for spreading the platform. So far they were extremely cautious to avoid Northern American market because of patent trolls and racketeers like MS and Apple (a concrete example how patent racket stifles innovation and reduces choice).
However may be the latest Alice ruling will encourage them to do bolder moves. Until then the situation won’t change much.
Edited 2014-10-08 02:46 UTC
I don’t think the NA market would make much difference…