Note: This article has been edited after taking inputs from Brendan Donegan, one of the test engineers in the Symbian Foundation delivery team. (thanks Brendan)
Earlier in the day I headed over to AllAboutSymbian.com and found David Gilson a little worried about Symbian’s move to Open Source. In a rather short article, David outlined his fears of manufacturers forking Symbian and consumers ending up having to choose between various different Symbian OSes, much like what happened with Linux as a desktop OS and Android as a phone OS.
Valid arguments, I would say, about the mindset of consumers who would rather willingly hand over the task of choosing to manufacturers. They don’t want to make decisions; they want a device they can have fun with.
So do David’s fears about Symbian’s move to open source imply that the switch has more negatives than positives? Not at all. And neither did David claim so in his article.
To begin with, the comparison of Symbian OS to Linux for the desktop is pretty unfair. Most desktop machines and laptops come pre-bundled with Microsoft Windows (or Mac OSX in the case of Apple machines). The average consumer is not too inclined to change the OS for the simple reason that the existing one does everything they need. Besides, the history of Linux being an OS only for techies still scares the average consumer in spite of the newer versions being easy to use.
The same cannot be said of Mobile OS. For starters, most people don’t look at Android as Linux. Android, to them, is just Android. Maemo is just Maemo. This automatically drives away any apprehensions about Linux they might have. Android was user friendly from the moment it was launched, which made it easy for people to accept it, unlike Linux for the desktop. And unlike Linux for desktops, Android comes pre-installed on devices.
So let’s just consider Android, then. Built specifically for mobiles, it is the best suited OS for comparison. True, many manufacturers modify it for their own purpose. But it is not necessarily a bad thing. Most modifications are only skin deep, with the underlying functionalities remaining unchanged.
Android has taken the mobile world by storm. The number of devices running Android is increasing exponentially. Would this have been the case had it been a closed, though free, OS? Probably not. On the other hand, Symbian share has been slowly decreasing, even though it is still hovering near the 50% mark. Some manufacturers were moving away from it, while some had never truly embraced it in the first place.
An open source OS gives manufacturers freedom to deliver the best hardware-software combination they can make. It allows them to make minor tweaks to the code to speed up performance, deliver a better UI or just enhance the overall user experience. Given the option, many manufacturers readily embraced Android OS when it came along.
Symbian’s decision to go open source comes across mainly as a way to combat the increasing popularity of Android. Manufacturers now have the dual advantage of an open source code and the thousands of (useful) pre-existing apps and games for Symbian.
Further, with efforts already being made by the Symbian community to improve an already excellent OS, it is unlikely that anyone would feel the need to modify the OS too much. Most tweaks will likely be applied to the UI, which will not affect the applications that are run on the device. And then there’s the fact that mobile OSes are seldom modified and distributed by non-manufacturers.
Most importantly, however, a fact that the article failed to mention is that the EPL is a weak copyleft license. In a GPL license, a person can make any changes he wishes to the OS and release it. Under EPL, however, the changes have to be approved by the foundation that controls the code. In this case, any changes made to the Symbian OS will have to be revealed to the Symbian Foundation which will do a compatibility test on it. If it is found that forking has occurred which renders it incompatible with existing Symbian applications, it will not be allowed to claim itself as Symbian.
All this makes it unlikely that Symbian will be forked enough to leave consumers confused about which one to choose. The move to open source will only make Symbian more favoured by certain manufacturers and will give it a fresh lease of life in the impending battle against Android.
You should also know that Symbian has a compatability program. We require anyone wishing to refer to the OS on their device as Symbian to pass a compatability test. This is similar to, for example, Bluetooth wireless technology. So if a manufacturer does fork than even if 95% of the code is the same as in Symbian as it is now then they won’t be able to use the Symbian name (which the Symbian Foundation owns) if they break even a small number of APIs.
Going open source is definitely a good step. Open source technologies are the demand for the day saving customers a lot of money plus adding an extra value by ways of personalised surface modifications. I definitely agree on the points you brought forward.
Hello there,
I’ve just found this post after all the time after writing my article on All About Symbian.
I’ll say for the record here, as I did in the comments to the article; I don’t think it was wrong for Symbian to go open source, it was a brave move and the right thing to do.
Moving beyond that, they key issue with the EPL is this - Other parties are allowed to modify and/or replace elements of Symbian, and keep those chunks of code to themselves, i.e. proprietary. That is how the licence works, and is were the potential for fragmentation comes in.
(Note, we changed the word ‘fork’ to ‘fragmentation’ at AAS, as this better described my concerns.)
I think the analogies I made are only unfair if you take them too literally. For the desktop Linux for, you are quite right when you say that most people won’t bother to change the OS on their computer. I agree, but that’s nothing to do with the point I was making. I wanted to talk from the point of view of somebody looking into a new platform, any platform which they knew little about, and wondering which options to choose.
Android for example, I still haven’t had a chance to use an Android phone yet, and I want to get one ASAP so that I can start familiarising myself with the platform. Although, I know there are three different flavours of Android (as pointed out in my article). So I feel like I need to ask someone for advice on which is the best flavour for me to get started - I.e. three options leaves me wondering which to pick. Even though Android applications will run on all of them; application compatibility alone does not make a user experience!
Because of how the EPL works, there is the potential for the same to happen to Symbian, and I think that would be bad for the platform. Although, I’m sure the Symbian Foundation will do what they can to mitigate this, but simply passing a compatibility test doesn’t solve the (potential) problem.
Thanks for your comment David. Great to have a reply from the author of the original piece.
I agree that fragmentation better describes what you wanted to put across. About your comment: “I think the analogies I made are only unfair if you take them too literally.”, It’s nice to have clarification on that. I suppose I took them out of context?
Well, let’s see how the move to Open Souce goes for Symbian…