Sailfish – An open source mobile OS(sailfishos.org)
sailfishos.org
Sailfish – An open source mobile OS
https://sailfishos.org//index.html
50 comments
Rejecting a new thing out of hand because it isn't the current popular thing is a classic example of "middlebrow dismissal".
Android is not perfect and was not always a popular thing and there is room for more than one mobile platform.
Android is not perfect and was not always a popular thing and there is room for more than one mobile platform.
I agree with you. Too bad HP did not. WebOS was really fantastic and just a bit ahead of its time performance wise. I feel like I'm one of the few users that "got it".
It is good to see another opensource OS step into the arena.
It is good to see another opensource OS step into the arena.
I think there were a lot of us who got it. Palm's hardware division, marketing dept. and whoever sucked at wooing carriers didn't get it.
How about rejecting a new thing because it lacks the backing necessary to succeed? Both Firefox OS and Ubuntu are backed by organisations with larger pockets, and Firefox at least has agreements with phone providers.
Sailfish has the backing of China's largest mobile phone retailer D.Phone Group, with 2000 stores and 150 million customers. That can be much more valuable to get the OS in the hands of real customers than a good brand name in other IT sectors.
True, but there is an argument that could be made that mobile ecosystems have become heavily relient on a strong third party development community which is likely to be almost non-existent on a new OS. As well expectation have been raised as far as what a mobile OS does and the level of polish.
There is certainly things wrong with Android and iOS and I'm not saying that a new popular mobile OS is not possible but there are certainly a lot more challenges that need to be over come then even a couple years ago. I am pretty sure that Android and iOS v1 would not have been overly successful if they had been released into the mobile OS landscape that currently exists.
There is certainly things wrong with Android and iOS and I'm not saying that a new popular mobile OS is not possible but there are certainly a lot more challenges that need to be over come then even a couple years ago. I am pretty sure that Android and iOS v1 would not have been overly successful if they had been released into the mobile OS landscape that currently exists.
Current mobile ecosystems definitely have become heavily reliant on 3rd party apps. That doesn't mean it's the only way a mobile OS (or any OS) can succeed, nor does it imply that you need to hit a critical mass of app developers for it to be successful.
I think the Ubuntu phone will run regular Ubuntu apps and the Firefox OS treats web apps as first class citizens. Those two are leveraging pre-existing applications.
Aside from that, the only apps I use on my phone are the browser, google maps, and google voice. Who's to say the massive 3rd party application pool is even needed for an OS to really be successful? What if Sailfish just had very few really well-written and privacy-respecting apps that they sold for like $20?
I think the Ubuntu phone will run regular Ubuntu apps and the Firefox OS treats web apps as first class citizens. Those two are leveraging pre-existing applications.
Aside from that, the only apps I use on my phone are the browser, google maps, and google voice. Who's to say the massive 3rd party application pool is even needed for an OS to really be successful? What if Sailfish just had very few really well-written and privacy-respecting apps that they sold for like $20?
I'm not disagreeing with the fact that taking a different approach to a mobile OS could be successful, simply that there is a certain expectation that people have about mobile and if you are not following that then you will have to convince people that this way is better.
Is this possible? of course, if its better it will win out. But this is definitely a hurdle that needs to be over come. Convincing people that they way they are use to is worse then an alternative is often a difficult task, especially for a device that pretty much never leaves someone's side. I even notice that people often have a lot of difficulty switching between Android and iOS if they have used one for a significant period of time, and most of the differences between these two platforms are fairly minimal.
Is this possible? of course, if its better it will win out. But this is definitely a hurdle that needs to be over come. Convincing people that they way they are use to is worse then an alternative is often a difficult task, especially for a device that pretty much never leaves someone's side. I even notice that people often have a lot of difficulty switching between Android and iOS if they have used one for a significant period of time, and most of the differences between these two platforms are fairly minimal.
being skeptical isn't being dismissive. There's alot of software that simply never takes off for one reason or another, so an air of skepticism is reasonable.
It looks nice, but simply being "nice" or a small amount better won't amount to much.
While that's a fair statement, an OS--mobile or otherwise--doesn't really need to be among the top few in terms of adoption to be valuable. Lots of people still use their N900 and N9 phones and want something to be a successor to them. This is likely the way I'll be going next.
I'm surprised there are not more viable OSs. The domination of computing by mobile devices is still in early days. There is definitely room for improvement versus both iOS and Android.
Among the would-be competitors, the Jolla team is best equipped to mount a challenge.
Among the would-be competitors, the Jolla team is best equipped to mount a challenge.
Firefox OS has potential. At least you can load it onto a Nexus device, although it is extremely limited.
I'm surprised there are not more viable OSs.
I'm not surprised. Mobile OSs are faced with an enormous challenge, which is hardware support.
The PC platform (while there are variances) is relatively uniform environment compared to the current crop of SoCs (system on chip) used by phones and tablets. There is hardly any standardization on things like power management, security, and other on-chip peripherals.
Each port for an OS is a lot of effort, so only the most popular OSs (sometimes only one of them) is supported by the manufacturer of the SoC.
I'm not surprised. Mobile OSs are faced with an enormous challenge, which is hardware support.
The PC platform (while there are variances) is relatively uniform environment compared to the current crop of SoCs (system on chip) used by phones and tablets. There is hardly any standardization on things like power management, security, and other on-chip peripherals.
Each port for an OS is a lot of effort, so only the most popular OSs (sometimes only one of them) is supported by the manufacturer of the SoC.
MeeGo was pretty great at running a lot of Linux binaries, wasn't it? I never had a phone with it or played with it, but it always seemed it was a big step closer to a mobile OS that allowed your phone to do most anything your desktop could do. If that's the case, I hope it takes off. Being able to run most BackTrack tools on my phone could be nice.
The N9, which only runs a Meego instance, i.e. provides the Meego API on Maemo (the biggest difference is likely that the package manager is dpkg rather than rpm as for Meego proper) can do pretty much everything the desktop can do, modulo the missing keyboard, libraries and architecture. Really, it’s mostly like any other Linux distribution with a little weird hardware.
My experience on the N900 was that desktop software runs down your battery like crazy. Phones need to run software that's optimized for power.
If i'm remembering correctly, it's not entirely the meego successor. Sailfish is a new project from the people who worked on meego, but the actual meego project turned into Tizen.
N9 'meego' was Harmattan, the direct descendant of Maemo 5, planned before the link-up with Intel to create Meego 'proper'. Harmattan had some backported compatibility pieces to make it nominally Meego compliant, but it was really Maemo 6.
Tizen came from the Meego proper project, which was more Moblin than Maemo.
This is what happens when you let marketing... do anything.
Tizen came from the Meego proper project, which was more Moblin than Maemo.
This is what happens when you let marketing... do anything.
Successor is probably the wrong word - it's more accurate to say that it grew out of some parts of that project along with Sailfish's sibling, Tizen.
First Sailfish device will be released next month (May): http://techcrunch.com/2013/04/17/jolla-to-show-first-phone-i...
May be some devs can get a hand on it
May be some devs can get a hand on it
How do I put it on my phone?
That would be nice to know, wouldn't it? We see things like this every now and then, Ubuntu TV comes to mind, where the idea looks great but at the end of the day it is targeted at OEMs, not the consumer.
I have been looking for a mobile OS that adheres to the principal of least astonishment for a while now, but have come up short. I would pay money.
I have been looking for a mobile OS that adheres to the principal of least astonishment for a while now, but have come up short. I would pay money.
If you have an Android phone, go to their website; grab the platform binaries, and flash it onto your phone using fastboot. (Disclaimer: you'll have to root your phone first, so be prepped for broken warranties.)
I'm kidding ofcourse. The real answer is, wait until they release some device that supports it!
I'm kidding ofcourse. The real answer is, wait until they release some device that supports it!
You don't need root access on whatever OS you're running to unlock fastboot. You typically only need to access fastboot via USB, how else would you load an OS to a blank device?
I'd love to know. I bought an N9 late in the game expecting its "openness" to work with all this year's upcoming mobile OS's, mainly Sailfish. So far all it can run is android.
The logo seems suspiciously similar to the Seagate emblem.
The tech description lists an "Android Runtime" (https://sailfishos.org//about-technology.html), so maybe it'll be able to run older/smaller android apps like the new blackberry phone. Good for users and developers, and not a bad way to jump-start a new OS.
Got excited when I saw the download and realised it was for an SDK. Ah well, that'll keep me entertained for a bit.
Hopefully they'll be working towards having something releasable so people can start using it in the wild soon, it looks interesting but it's going to need many eyeballs on it before it's ready to play with the big boys.
Hopefully they'll be working towards having something releasable so people can start using it in the wild soon, it looks interesting but it's going to need many eyeballs on it before it's ready to play with the big boys.
Is there actually any information about the OS?
Are there any examples of how this is different than what we already have?
Isn't Android exactly this. Linux with a touch UI.
At least from a developer's point of view there's not much Linux visible in Android.
IMHO, the Android APIs feel like Symbian rewritten in Java... I get the impression that Google has spent five years rushing to add new components to accomodate everything under the sun with no vision of how the stuff fits together, and meanwhile vendors do their own thing that is slightly incompatible with the core. Of course Symbian was worse in every respect, but the end product is worryingly similar.
IMHO, the Android APIs feel like Symbian rewritten in Java... I get the impression that Google has spent five years rushing to add new components to accomodate everything under the sun with no vision of how the stuff fits together, and meanwhile vendors do their own thing that is slightly incompatible with the core. Of course Symbian was worse in every respect, but the end product is worryingly similar.
Of course Symbian was worse in every respect
Except less hardware requirements and better battery runtime :-)
Except less hardware requirements and better battery runtime :-)
And, dare I say it, better multi tasking.
It's much closer to a desktop linux distro - it still uses glibc afaik, and other more "standard" linux bits (eg pulse audio and dbus and x11 and rpm and gstreamer etc - see here for more info: https://wiki.merproject.org/wiki/Architecture - Sailfish is based on Mer, so the architecture should be fairly similar)
Android has way more bits that have been made explicitly for it - many of which were made to BSD license-able (or so I've read) :-)
Sailfish has a Qt-based UI stack. It may also not depend on one specific application runtime environment as much as Android does.
There are also some hints that Sailfish is taking an approach similar to Ubuntu's for touch and GPU support on mobile devices and blending Wayland with Android's compositor.
In other words, Sailfish is likely to be closer to a mainstream interactive Linux OS, and a closer cousin to Ubuntu, Firefox OS, Tizen, etc. than is Android. It's different times now, and you don't have to diverge so much from the Linux mainstream to create a good touch user experience.
There are also some hints that Sailfish is taking an approach similar to Ubuntu's for touch and GPU support on mobile devices and blending Wayland with Android's compositor.
In other words, Sailfish is likely to be closer to a mainstream interactive Linux OS, and a closer cousin to Ubuntu, Firefox OS, Tizen, etc. than is Android. It's different times now, and you don't have to diverge so much from the Linux mainstream to create a good touch user experience.
Most importantly, Sailfish offers real multi-tasking as we know it from desktop operating systems. While in Android or iOS your application is suspended when you switch to another one, in Sailfish they can run in the background.
"While in Android or iOS your application is suspended when you switch to another one"
That's not correct. Android has multi-processing. The Android runtime enables Service components to run in the background, and has preemptive thread scheduling, and has had these features throughout all publicly released versions of Android.
That's not correct. Android has multi-processing. The Android runtime enables Service components to run in the background, and has preemptive thread scheduling, and has had these features throughout all publicly released versions of Android.
In my experience, Android multitasking still doesn't work like a desktop (or even my old n900.) Certain applications (those registered as services) can continue to work in the background, but many suspend entirely.
Sometimes I want to do something else while I wait for a large web page to load over my mobile data connection. On my n900 this worked fine, but as soon as I switch to another task, Chrome on Android 4.2.2 quits loading the page.
Plus, task switching is still a heavy operation on Android - swapping back and forth between two applications is a bulky, slow operation compared to the n900, which, itself, was bulky and slow compared to a desktop window manager.
Sometimes I want to do something else while I wait for a large web page to load over my mobile data connection. On my n900 this worked fine, but as soon as I switch to another task, Chrome on Android 4.2.2 quits loading the page.
Plus, task switching is still a heavy operation on Android - swapping back and forth between two applications is a bulky, slow operation compared to the n900, which, itself, was bulky and slow compared to a desktop window manager.
You may be executing code on the UI thread. Take a look at Loader for a construct designed to preserve state across component lifecycle events: http://developer.android.com/guide/components/loaders.html
Suspending apps in the background of iOS is a _feature_. It's not like the OS doesn't support "real multi-tasking". Apple could give us "real multi-tasking" at the flick of a software switch -- most of us prefer battery life.
This is a piece of common wisdom among the Apple "slavery is freedom" crowd, and is impervious to the fact that N900s and N9s have fine battery lives. I frequently have a couple of browser windows, my media player, bash, and an SMS window open on my N900, and it isn't too tough on my battery.
The problem with battery life and multitasking on N900 is that it has so little RAM that it gets super swappy and starts to thrash esp. under heavy client-side js. I'm dying for an N900 w/2013 specs.
>Apple could give us "real multi-tasking" at the flick of a software switch
This is not true. Task management has to be designed into a multitasking OS. Maemo did a beautiful job.
The problem with battery life and multitasking on N900 is that it has so little RAM that it gets super swappy and starts to thrash esp. under heavy client-side js. I'm dying for an N900 w/2013 specs.
>Apple could give us "real multi-tasking" at the flick of a software switch
This is not true. Task management has to be designed into a multitasking OS. Maemo did a beautiful job.
Have you ever used a N900 or N9? Both of these devices do offer full multi-tasking and base their software stack on the Linux kernel and glibc. The battery life of the devices is actually quite decent.
Android is completely different, everything runs inside a version of the JVM called 'Dalvik'. Linux is incidental and nothing Linux-related is accessible to apps
Android with google isn't completely open-source, and it's certainly not free as in freedom. But this is a role that Firefox OS will fix, I guess. I don't think sailfish or ubuntu are going anywhere.
B2G (ie FirefoxOS) has the same problem as Android as in they both have completely unique weird userland that has very little to do with your "normal" Linux userland, or anything else for that matter. I'm not really sure about Ubuntu phone OS, maybe it will have userland closer to their desktop variant. But on the other hand that won't help all that much as even Ubuntu desktop is veering away from what I'd consider the "standard" Linux distro.
Some people just would prefer having desktop-like OS on their phones with some mobile oriented apps, and that's what Maemo/N900 almost delivered. If Sailfish is anywhere close to that experience then I'll be a happy hacker. Admittedly commercial (mainstream) success probably requires bit more, but on the other hand I know non-linuxy people who were/are quite satisfied with their N9/N900.
Some people just would prefer having desktop-like OS on their phones with some mobile oriented apps, and that's what Maemo/N900 almost delivered. If Sailfish is anywhere close to that experience then I'll be a happy hacker. Admittedly commercial (mainstream) success probably requires bit more, but on the other hand I know non-linuxy people who were/are quite satisfied with their N9/N900.
Correct, I meant that FirefoxOS would be the open-source mainstream OS, not the linux mainstream OS. And I don't think that Ubuntu will be mainstream, because they don't have any partners while Mozilla have tons of partners and a schedule.
Will there be any competition between Sailfish and Tizen? Will it be a war among brothers?
I miss Meego.
I wonder how this compares to established Free Software mobile operating systems such as Cyanogenmod http://www.cyanogenmod.org/
Cyanogenmod is not an separate OS. It is just Android.
It's hard to see how yet another mobile OS will make traction. Perhaps the Android runtime will make it compatible enough that developer-types will use it.