January 6th, 2009
When Windows Mobile 1.0 was first released in 1996, Microsoft was saying it had ambitions to not just have the OS in mobile devices but also in anything from phones to cars to petrol pumps. Microsoft tried hard and there was even an automotive version of the OS at one time. In fact, I still occasionally come across Windows Mobile in non-pda and non-phone devices, especially in point of sales terminals.
Conversely, I have commented several times on how Symbian seems too tied to mobile phones and how there aren’t even any rugged devices for use by enterprise customers. With Symbian OS going open source, we will hopefully see a greater range of devices.
The ‘open source’ affect of encouraging a greater range of devices can already be seen with Android. There’s already talk of using Android to power NetBooks. I can foresee a time where we might possibly see Android being used in a wide range of devices.
Related Articles:
Posted in Symbian, Windows Mobile, Mobile, Android
January 5th, 2009
I have been reflecting today on the announcement that UIQ has filed for bankruptcy. With the creation of the Symbian Foundation this was obviously inevitable. However, going further back, what happened? Why was Nokia’s use of Symbian OS so much more successful?
It’s ironic that Sony Ericsson had touchscreen phones long before the current touchscreen frenzy. In fact, as with many mobile companies, UIQ was probably too far ahead of its time. Early Sony Ericsson UIQ touchscreen phones were always expensive, high end devices with a limited market. Conversely, Nokia’s S60 phones reached down into mid-prices and many people purchased (or rather upgraded via their network operator) without even knowing it was Symbian or a smartphone.
The UIQ company itself was treated as a pawn - ownership moved around from company to company because noone really knew who should be responsible for the loss-leading but necessary entity. As with many mobile ventures, it’s just not possible to make money when there are too many partners (Symbian, UIQ and handset OEM) trying to take a slice of the revenue.
Posted in Symbian, UIQ, Mobile
December 28th, 2008
Over the holiday period, I read an article in the paper version of Mobile Today examining why cases of faults are rising as more smartphones and touch-screen devices hit the market.
Here are the parts that I particularly found interesting…
- So endemic is the problem that many consumers almost expect a new phone to have errors
- Some retail staff that Mobile spoke to say the hassle of handling returns make them think twice about selling a high specification smartphone that has just been released
- The clash of the manufacturer’s software with the operator’s settings and services is thought to be a major factor when it comes to the problems reported
- Rather than owing to technical faults, many within the manufacturing fraternity blame handset returns on high consumer expectation and a lack of understanding
I have personally seen that Handset OEMs spend a huge amount of effort testing both the hardware and software associated with their handsets. It’s frustrating to know that many of the problems are actually due to changes performed for (or by) the network operators. I see this as yet another reason why phones might better be sold more like PCs - as open, tested and working devices that can be attached to any network.
The high ‘handset returns on high consumer expectation and a lack of understanding’ is more concerning. It leads the the question whether these people really need such complex devices or whether the complexity can be reduced via a better user (retail and phone) experience.
There’s also an online version of the Mobile Today article.
Posted in Mobile, Phones
December 18th, 2008
Paul Golding presented at last week’s O2 Litmus event. The topic was the ‘Mobile Developer’s Dilemma - What Should I Build?’. I just came across his presentation on slideshare and it brought back how interesting I found this presentation.
The presentation on slideshare includes lots of information for people considering mobile development including thinking about…
- The user experience (the whole product)
- Using Context
- Business architecture
- Co-creation with partners
- Current opportunities
- 2009 Mobile Drivers
Posted in Mobile
December 17th, 2008
It has has been pointed out to me that there’s a Vodafone BlackBerry Storm Developer Competition…
"…the once-in-a-lifetime opportunity to win 50,000 euros – the three best software applications will be provided as downloads for all users in the BlackBerry Application Center"
However, beware. The dark side of this is the Terms & Conditions…
"The winners acknowledge VF’s rights of usage (reproduction, duplication, distribution, publication, transmission, presentation, promotion, etc.) pertaining to the respective applications for all currently known forms of utilisation exclusively, in full, and without any limitation of content, regional restraints or time limits."
…and furthermore…
"VF is authorised to exploit any or all applications in conjunction with other works, and also to process, subsequently modify, enhance or expand the respective applications as they see fit. As the granting of rights detailed above is equally applicable to the relevant object and source codes of the respective applications, the winners are bound over to provide VF, on request, with the object and source codes free of charge."
Reading this, if you were to ‘win’ 1st prize of 50,000 Euros, you are giving your application away to Vodafone for this sum. It’s even worse if you ‘win’ 3rd place - only 5,000 Euros for all your intellectual property. Unbelieveable.
Related Articles:
Posted in Mobile, Network Operators
December 16th, 2008
I came across TagsMe for the first time yesterday. It’s a way of creating Java ME applications without writing any Java. You define the application in a similar way as a web page - using XML and the TagsMe scripting language.
The TagsMe client application is actually a browser that interprets the XML and script. Separating presentation from the behaviour allows the same XML file to be stored locally in the mobile phone, stored on the Internet or generated in real time from a server.
There’s a video demo of how to use TagsMe to create an iPhone style icon list.
Related Articles:
Posted in Mobile, J2ME
December 15th, 2008
On his blog, David Wood, Executive Vice President of Research at Symbian, explains why he thinks there’s a need for signing mobile applications.
David takes the recent example of rogue Android applications racking up hidden charges. Unfortunately, this can happen on any platform, including Symbian - even with 3rd party tested Symbian Signed applications. Symbian Signed doesn’t test for this and many other ‘rogue’ scenarios.
At one time, the Symbian Signed requirements required that a ‘billing’ prompt be given to the user when the application first accessed the Internet. Optionally, a second prompt after this could allow the user to turn off further prompting. It has been removed from the last two iterations of the Symbian Signed requirements presumably because…
a) Most people would just disable the prompts and further ‘rogue’ events can happen unnoticed
b) The prompts really don’t make sense with the majority of people using ‘unlimited’ data tariffs
At the time this ‘billing dialog’ requirement was in place, I had difficulty explaining to my clients why the S60 version had so many extra screens/dialogs compared to the Windows Mobile, RIM and Palm OS versions of the same application. I am pleased the requirement has gone now.
As the above example shows, a rogue scenario can be very complex. Sometimes there are exceptions where the rules don’t make sense. Sometimes there are cases where the rules make more sense (e.g. Data access while roaming).
The problem with the Symbian Signed tests is that there aren’t enough of them and they aren’t detailed enough. Hence, they don’t really cover as much as people might think. Conversely, the existing set of tests are already an administrative and development burden so adding to them wouldn’t be practical. So what should be done?
While I agree we need signing, I am beginning to believe more in the Android self-signing approach where developers can be traced and applications revoked (disabled by Google) as necessary.
As I have said previously, I think more could be done by the phone software itself to police activity rather than solely relying on signing. As an example, in the case of use of data while roaming, the phone OS should be able to detect this and warn the user.
Related Articles:
Posted in Symbian, Series 60, Mobile, Android
