How do you identify the mobile operating system with the best potential?
One approach is to focus on percentage growth in unit sales figures. For example, James Pearce, writing in today’s Washington Post, identifies Google’s Android OS as “the fastest growing OS in the market”:
Google’s Android OS will ship in eight million handsets in 2009, an increase of 900 percent compared to the year-ago period, according to Strategy Analytics. At that rate, it will be the fastest growing OS in the market. Apple is listed in second place with an expected growth rate of 79 percent.
But hold on. Let’s not be carried away with what looks like large numbers. A 900% increase on a baseline of less than one million last year, is actually a smaller absolute increase than a projected 79% increase on a baseline (in Apple’s case) of around 13.7 million devices sold last year. 900% may sound a lot larger than 79%, but 900% of a comparatively small number is (in this case) less than a 79% of a comparatively large number.
As Chis Thompson writes in his article Could Android Explode? on Reuters.com:
Google has a long way to go before playing with the big boys. Only 8 million Android phones have sold so far [actually, that's the Strategy Analytics forecast for 2009], compared with 75 million Symbian smartphones in 2008. But with more and more companies adopting the Android operating system … Google is clearly on the way.
But nor is engagement from a large number of different manufacturers necessarily evidence of pending commercial success. Famously, although around 50 different mobile device manufacturers have created devices using Windows Mobile, about 80% of the total sales of Windows Mobile is due to just one company – HTC. During my visit to China last week, I heard time and again that manufacturers are stepping back from Windows Mobile, on account of their devices losing money in the marketplace.
In other words, just because many manufacturers are lining up to try to bring Android devices to the market, there’s no need to declare “game over”. That’s far too premature.
What manufacturers will look for is a mobile operating system platform that:
- Allows them to create devices, relatively inexpensively and relatively quickly; AND
- Allows them to sufficiently differentiate their devices from those of competitors using the same platform – so that their devices can attract buyers.
The differentiation between the forthcoming Samsung GT-i8910, Sony Ericsson Idou, and Nokia N97 – all of which use the same core Symbian platform software – is remarkable. (And Symbian devices launched in Japan by Sharp and Fujitsu are decidedly different again.) As I see it, that’s good reason to conclude there’s still plenty of potential in the Symbian platform.
Looking slightly further afield, Symbian Foundation personnel are focused on growing the most productive and valuable software movement on the planet. We believe that, if we get that right, the Symbian Platform will in due course become the world’s most widely used software platform – and the world’s most widely liked software platform.
To reach this goal, we are keeping close attention on:
- Platform delivery: Schedule reliability, quality reliability, asset integrity, and device certification;
- Platform evolution: Vision, roadmaps, scope growth, and compatibility;
- Developer productivity: Tools and kits, documentation, APIs, and timely ecosystem support;
- Open community: Contribution as well as adoption, events and forums, apps publishing, and open brand values.
These tasks take advantage of the best of the historical strengths of the Symbian platform and Symbian ecosystem, but add in a huge burst of new energy and creativity through our increased adoption of openness – open SDKs, open source, and open governance. This combination of “old plus new” is full of growth potential!
Footnote: If “Mobile OS wars 1.0″ was focused on which platform had the best technology, “Mobile OS wars 2.0″ is all about which platform is surrounded and supported by the most productive and valuable ecosystem.
> the world’s most widely liked software platform
Liked by whom? Developers? Handset manufacturers? End customers? Of course it’s a great goal to aim for everyone but it will be interesting to see how this would be measured. For example if a handset manufacturer or network operator passes back the details of a satisfaction survey showing that their product is well like will that be taken at face value? Will it show an indication of Symbian being a good product or the device – or both?
> “Mobile OS wars 2.0″ is all about which platform is surrounded and
> supported by the most productive and valuable ecosystem
I agreed with this 110%. I greatly await the Foundation spending as much public time and effort engaging with the greater open source and wider development communities as it has with brining together the paying founding members. Apple and Google, for all of their relatively quite small market share, are leagues ahead in this area and this is where it’s going to count.
Hi David,
> > the world’s most widely liked software platform
> Liked by whom? Developers? Handset manufacturers? End customers?
Yes, all of these – as indicated in the diagram in the main posting
Of course it’s a great goal to aim for everyone but it will be interesting to see how this would be measured.
I’m open to suggestions for how this could meaningfully be measured
// David W.
Part of me can’t help but genuinely appreciate the figures and headlines in the media regarding Google’s smartphone growth, seeing as head of PR for Symbian Ltd for 4 years, I drafted financial results press releases with similar stats in the headlines, hoping for the same coverage as Google received yesterday. But that was 5 years ago and the stats were global.
As you said David, it’s huge growth but when you have minimal handsets things need to be kept in perspective so thanks for doing that.
> Mobile OS wars 2.0″ is all about which platform is surrounded and supported by the most productive and valuable ecosystem.
Unfortunately I don’t think it sounds as sexy as Mobile OS wars and therefore unconvinced that people would want to read about it in the headlines but analysts do and as we’ve seen, analysts can sell reports.
Growth is what drives the capitalist economy and is what business journalists have to report on. There is not much news in “Bluechip churns out the same number of products year in year out”.
OSes are quite interesting because the investment is indirect.
To play devil’s advocate for a minute:
Google’s strategy is something like “give it away for free, build up a huge subscriber base and then monetise”. Google are big enough to have several loss leader product with HTC and others while the platform gains momentum.
Apple are about control – they roll their own exclusive software so that they can immediately monetise on their products and sales channels. It’s quite an established easy to understand mogul business model.
We can speculate that Symbian furthers Nokia’s ambitions to become a service company. By giving away the platform, Nokia can monetise on the services that run across an ubiquitous platform.
All three strategies have limits when it comes to growth, but if Symbian sustains growth outside of the Finnish company that bankrolled it, then it has the best potential to become ubiquitous.
In my view Symbian’s Non-Nokia growth is a more important metric than total units shipped.
>>compared with 75 million Symbian smartphones in 2008
I don’t understant why people compare non-comparable devices. 8million android Vs 75 million Symbian but note 8million android device falls in to smartphoen category. What is the figure for smartphone out of those 75 million?
>>>world’s most widely liked software platform.
Definetly not liked by developers. If you write application for android/iphone that will run on all devices of that platform. How much is it true for Symbian?
I would be happy to see any paid application that has downloads in range of hundred’s of thousands/millions and just built on symbian SDK **without** any API plugin from Nokia. It will be rare, if any and No one guarantees if API plugin will work or not, infect it can break between firmware upgrade. Absolute nightmare for developers.
Can Symbian foundation change it, I seriously doubt becuase it is controlled by manufacturer e.g. Nokia.
@DW: Technically, the reports describing “growth” for Android as fastest of all OSs are correct – even if the absolute numbers are small. I think these analyst reports are often geared for investors/VCs who want to buy small (or see unrealised opportunity) and sell high (or make money before anyone else does). I agree these reports can mislead though – and investors frequently tank!
@Kim: All Symbian phones are smartphones.
>>Definetely not liked by developers. If you write application for >>android/iphone that will run on all devices of that platform. How much >>is it true for Symbian?
I think, if applications written for a unified UI would be compatible with cross-manufacturer models (e.g. Samsung GT-i8910, Sony Ericsson Idou, and Nokia N97) then that would boost developer incentive.
Superficial UI differentiation could occur between handsets (to a remarkable degree that the mainstream user would not recognise it as a “Nokia” interface – think icons, transitions, fonts) but essentially it would be great if one application would run on all.
Raises the question – do manufacturers want this? Instead, it looks like they want their own application markets. If this is the case, I get a little suspicious as this can only succeed if manufacturer-specific application differentiation occurs, with manufacturer-specific high level APIs. Which means Symbian loses out to the kind of “application fragmentation” that occured between UIQ/S60/intra-S60 – except this time there are more manufacturers than Symbian UIs. In the long term everyone (including manufacturers) could lose out.
Where is the line drawn between the need for manufacturers to differentiate their 3rd party applications and services (if this is the case, or is all in my head), and an operating systems need to have full compatibility across all applications? Can full compatibility still be assured across all Symbian Foundation devices?
> Instead, it looks like they want their own application markets.
Don’t forget the Network Providers who will also want a cut from that pie with their own stores not to mention 3rd party providers like Handigo.
This is the price you pay (1) for having a product that runs on multiple hardware platforms from many providers and (2) being based around the OS not services – the Google Phone could be on as many devices as you like but Google would still control the App Store.
These are issues with the difference between Google (have the services, build the devices), Symbian/Nokia (have the devices, build the services) and Apple (paranoid control over everything).
It will certainly be interesting to see which way things move as, increasingly, the handset becomes a commodity (I’m sure I’m not surprising anyone with that notion).
Also, for the purposes of education, how does QT factor in with the above?(I’ve been away on current affairs!)
Kim, you’ve said: “… Definetly not liked by developers. If you write application for android/iphone that will run on all devices of that platform. How much is it true for Symbian?…”
But again definitely how many devices based on android/iphone do exist? 1, 2 or ‘almost 10′?
And try to compare this amount with tens and tens (!) of symbian-based devises, dedicated for different markets, target audience etc.
Mr DW,
quit stalling with these boring blog posts and launch the website already!
No news since 2 April? Common! Are you guys dead or what?
Where’s the tools? Where’s soruces? Having to be a paying member to write a hello world app is a joke.
Mr mrmr,
>launch the website already!
Send me your email and I’ll send you a link and login to the beta site.
// David W.
One of the strongest aspects of Android is in its openness that enables operators to create fully integrated service experiences like never before. But, hold on, isn’t this what SF is all about?
For some reason there is a lot of buzz around Android now. Something I would like to see more with Symbian as well.
Hopefully SF will be inspiring enough for the operators to again start the buzz. One of the things desperately needed for Symbian is great devices from more manufacturers.
[...] admin on May.13, 2009, under Android, Symbian Ancora un post graffiante sul blog della Symbian Foundation relativo a quella che David Wood chiama la Mobile OS wars [...]
Those points about growth rates are well made. As we mentioned in ‘The Fonecast‘ this week, a 900% estimated increase isn’t so impressive when the increase is calculated on a brand-new product that was only sold for a fraction of a year (and has an unknown number of models launching in 2009). Even the estimated iPhone growth is worth questioning: is this based on the rumoured TWO new iPhone devices?
> How do you identify the mobile operating system with the best potential?
My 2c,
The questions should be more like: How do you identify the mobile *platform* that has best potential?
To answer that we need to look the picture from few different perspectives:
1) Users
2) Developers
3) Operators
4) Manufacturers
and the contenders now are:
- Symbian
- Apple
- Android
- Windows Mobile
(deliberately leaving out Blackberry and Palm’s Pre)
1) Users like good user experience, continuity and choice.
Symbian offers now best choice, Android is likely to catch up in couple of years. Windows Mobile and Apple are more like single offering. iPhone offers superb user experience, Android is pretty good and Windows Mobile and Symbian are at the bottom. On continuity Apples updatable software platform (supported by limited number of devices) leads, Android shows promise and again, Windows Mobile and Symbian are at the bottom with their “we only fix bugs until next device comes out”.
Apple would be a strong candidate for winner here, but the limited choice puts it second. Points here go to:
Android 4p
Apple 3p
Symbian 2p
Windows Mobile 1p
2) Developers
Developers value good tools and ability to market their products. Clearly the best platform at the moment and probably for few years to come, is Apple’s. Windows Mobile has good tools but distribution is the problem. Android’s toolset is OK with quickly improving distribution. Symbian has horrible tool chain with different SDK’s and fairly weak distribution even with the great number of devices out.
Points here go to:
Apple 4p
Android 3p
Windows Mobile 2p
Symbian 1p
3) Operators
Operators like ARPU, low attrition rate and choice of suppliers. Again, Apple and Android lead the pack even when Symbian and Windows Mobile has had longer history with operators. When operators learn to utilize Android platform better in their own offering, we might have a strong candidate for a winner here. Symbian (Nokia) sales figures clearly outpace Windows Mobile’s giving it an edge over the Windows Mobile.
Points here go to:
Android 4p
Apple 3p
Symbian 2p
Windows Mobile 1p
4) Manufacturers
This is bit more tricky one since manufacturers have bit different position compared to each other. Cost is of course an issue (now more than ever), attractiveness from users perspective and acceptance from operators. Apple sure likes it’s own OS, but since it is not available to others gives it low rank here. Android again steals the victory with it’s cost effective (license, needed technology expertise. Linux & Java ). Symbian has acceptance from operators, but building Symbian devices is far from being cost effective. Windows Mobile has not shown to be any gold mine for manufacturers using it.
Points here go to:
Android 4p
Symbian 3p
Apple 2p
Windows Mobile 1p
Adding the points ends up with final scores:
1. Android 15p
2. Apple 12p
3. Symbian 8p
4. Windows Mobile 5p
So, how could the contenders improve:
- Android. Agreed, much of the points still are promises. But, it simply needs to execute.
- Apple. They have very good package with great user satisfaction, nice service layer (with music, apps, games, other media) and good publicity. However, with their closed model they will not be able to become the mobile platform with “best potential”. Will be good for Apple though.
- Symbian. Much of the problems with Symbian comes from the fact that it is simply too old, based on somewhat difficult technology that is not really compatible with anything. Symbian is the mobile platform that had best potential. Not likely to get it back.
- Windows Mobile. Difficult to see that Windows Mobile would find a business model that would put it ahead of the Android or Symbian. World is moving away from the PC centric thinking weakening the best promise the Windows Mobile has had: Windows compatibility. Will likely suffer.
So, my pick is Android.
Following on from Karl’s post:
The issue of current device upgrades is something that has already come up a number of times. This appears to show a major difference between Google (considers the handset as a commodity and the OS and services as the product) and, say, Nokia (considers the handset the product and the OS a commodity).
Symbian’s toolchain is frequently negatively compared to other OS equivalents. The Foundation needs to rapidly put an “tools guy” in place to start canvasing developers to the exact specifics of these issues. (Caveat – I work for Nokia in OS dev tools so I have a vested interest in that happening).
For (3) and (4) I think you strongly underestimate the importance of the long history between Symbian and the operators / manufacturers. These folks tend to be risk averse and I don’t think will be mass-jumping to Android any time soon even if it was a markedly better offering (which to most people it isn’t).
A whole section you missed out would be about brand. At the moment the only people I know with G1s are hard-core geeks who actually care more about the politics of the OS than its functionality. Most people have Symbian phones as they are the most common and have a good functionality / cost balance but they in no-way show off or sell benefits of those devices to their friends. On the other hand people *love* their iPhones and even now show them off at an opportunity.
Symbian has some way to go to make either the OS itself or the development environment as *exciting* as the iPhone / Android to those two respective groups.
Replying to David D,
I do agree with David that different approach of Google/Apple vs Nokia may affect the marketshare. The more user have content on the phone (music, applications, games), the bigger the need is to have a device platform instead of throwaway devices. This may hurt traditional phone manufacturers who tend to forget the device once it is out of their hands.
Challenge with Symbian toolset is not only the lack of tools. The Symbian developer environment is not that developer friendly and it will be a difficult task to change the fundamentals. Competitors have done very good work in this arena.
About the history of Symbian and operators. It’s more of an question about the history of Nokia and operators nowadays. Symbian as it was, exists no more, and only manufacturer in addition to Nokia making any meaningful amount of Symbian devices is Samsung.
Android should not be judged on the G1 performance alone. Android simply puts out an offering that should make sense to a manufacturer, operator and the end user. It is not perfect and the biggest challenge for it will be avoiding the fragmentation.
Original topic in David W’s post the the “best potential” OS. Unless more manufacturers joining in to use Symbian is bigger that the ones who are leaving it, I’m sorry to say that I don’t see Symbian holding the best cards in the game.
> Challenge with Symbian toolset is not only the lack of tools. The Symbian
> developer environment is not that developer friendly and it will be a
> difficult task to change the fundamentals.
I completely agree. I think that Foundation should, at the beginning at least, spend more time concentrating on building a relationship with, in particular application level, developers and work on better APIs, signing systems and tools. Personally I don’t think this has anything like enough focus going on at the moment.
> Unless more manufacturers joining in to use Symbian is bigger that
> the ones who are leaving it
I’m sure that will be the case. Others may disagree.
The problem with analysts is that they analyse things without taking people and feelings into account. Android, well that’s in the position Symbian used to hold. It has the operating system power, the user friendliness (at least on the surface) and the household name backing it to turn feature phone users into smartphone users. By not concentrating on telling people they’re using a smartphone (and therefore not scaring them off with the still prevalent thought of a gigantic brick phone that’s way too expensive to even consider) Google has delivered an experience to consumers that they thought was only possible on a PC. They gently lured them in and opened the market a bit.
For Symbian Foundation devices to get back to the top they’ve got to be easy for users to use, powerful and open enough for developers to do anything they can dream up, compatible in the idea that the neither the core systems nor any manufacturer add ons get in the way of applications, and the marketing has to be aggressive as hell (the only real lesson I believe the iPhone taught us is that people are idiots for a shiny toy that they’re told they need).
David Durant, above me, has a point about working on the signing system. I’m a regular user of the beta labs and recently sent a Samsung S60 phone owner over there to check out some of the nice little apps they’ve been working on. He spent ages trying to install something only to find out they were signed to only work on Nokia phones. I’m not a developer like some of these guys but that’s counter-intuitive at the user level. Manufacturer shouldn’t matter at all in the signing process if the OS is the same.
Sorry but I am just a dumb consumer here. The most important thing to me right now and i know it is very important to other people especially over at howardforums S60 N97 discussions is, will i be able to put the next symbian foundation OS onto my N97?
I don’t understand the difference between N97 firmware and Symbian OS now.
If the answer is “it depends on Nokia”
Then will we be able to turn to some hacker in protest to stick the new Symbian OS on our phones because let me tell you Nokia is infamous for neglecting their NAM firmware updates and that is exactly what will happen will hack the new symbian os on to the phone if we can.
If you could please give your answer not only here but in the howardforums official N97 thread thousands of N97 owners would appreciate it. Thank you.
Hi Aaron,
>I am just a dumb consumer here
Consumers are very welcome here! But let’s scrub that adjective “dumb”. Come to think of it, let’s scrub the “just” too!
>If you could please give your answer not only here but in the howardforums official N97 thread…
I’ve replied on the howardsforum.
// David W.
Hi David, all,
I wonder if you’ve heard this is piece of news: 18(!) new Android devices are expected to ship this year. See this: http://tinyurl.com/qws4n3 Okay, expectations and announcements are different from pure facts, however, this number is still impressive.
Gabor