Many digital and software products these days include an API (Application Programming Interface) which allows 3rd parties and/or the customer to extend and adapt the standard functionality. Could femtocells provide a standard API and what would it offer?
Why are API’s popular
Innovation in digital products seems to be accelerating at an ever more rapid pace these days. New products or features which once would have taken years to develop and bring to market seem to appear almost every day.
Software and hardware vendors have recognised that they can increase the pace of innovation by enabling a wide range of small to large 3rd party developers who can provide massive, parallel resources.
- Many niche applications which wouldn’t justify large investments can be put together by one or two man teams
- Domain expertise from a wide variety of subjects can be tapped
- Relatively low investment (although good documentation, software development suite and marketplace/Application Store are really needed to make it work well)
- Increased marketing potential, as 3rd parties sell through to their own niche markets
- Mash-ups, which combine several products using their APIs create previously unthought of (or unviable) solutions
Examples of API's
Here is a small sample of some of the API’s commonly available:
- e-Tailing: Website providers allow 3rd parties to create their own storefronts and shopping websites, including “white labelling” everything from reselling domain names to physical goods .
- Cloud Computing: Amazon’s IT department has opened up its computing “cloud” and offers storage and database capacity which can scale up on demand.
- Telephone Services: Ribbit offers a telephony platform which allows 3rd parties to create their own telephone company.
- Video Services: Netflix, who offer over 100,000 films for instant viewing on your PC, offered a developer API since October 2008.
- Google: Anyone can create their own Google Gadget, for display of information on personalised Google home pages. More recently, the Google Android phone includes a full Software Development Kit (SDK) encouraging innovation through extended applications.
These API’s can be extremely popular with 3rd party developers:
The iPhone software development kit (SDK) was downloaded over 100,000 times within days of being launched in March 2008.
This is partly because of the considerable profits that can made in a short time for popular applications. Steve Demeter made over $250,000 in two months from Trism, a $5 game. It's seen as the latest goldrush, with over 300 million downloads by end 2008 - in fact, iPhone Apps are downloaded more than twice as often as songs.
Apple is also making good profits from reselling 3rd party applications. Not only do they make 75% commission on each sale, but the wide range of applications available increases demand for the iPhone and justifies premium pricing.
So what about a Femtocell API?
I wouldn’t be surprised if the femtocell industry took a similar approach. By opening up femtocells to the wide range of innovation and energy of the army of small developers out there, many more possibilities could arise. Some will undoubtedly be for applications previously undreamt of.
As with early mobile phone API’s, there may be a temptation for individual vendors to create/offer their own flavour. I don’t believe the market size would justify such a splintered approach – there has to be a single, common, shared API which is supported by all femtocells that claim compliance.
The API may start in a simple way, with relatively few capabilities, but would be extensible so that later versions can offer richer and wider ranging features.
Learning from other programs, the API would need to be backed up with excellent documentation, a members forum, a software development kit, some easy form of testing and an online application store.
Future releases of the APIs would need to be fully backward compatible.
Because different femtocells and different operators would support or permit different capabilities and features, the API would need to present the available capabilities to applications on demand. This process of offering the available services (a bit like getting the menu when entering a restaurant) is known as Discovery, and would need to be a basic element of the API.
Could the first version be offered on handsets only?
Some of the applications we’ve seen demonstrated for femtocells are based around the “Connected Home” concept. This is where your handset is aware that it’s in the home femtocell range (the so-called femto-zone) and operates differently. For example, data is uploaded/downloaded and remote control of domestic gadgets becomes possible. ip.access have dreamt up some possible scenarios .
So perhaps the first femtocell aware applications may run on the iPhone or Google Android and simply use the knowledge of being in the femtozone (which can be determined from cell ID) to operate differently.
The dual-mode WiFi/UMA service offered by Orange is associated with Orange’s LiveBox – a home hub which includes broadband modem, WiFi access point and IPTV. There is a developer program for the LiveBox in associated with the French Telecommunications Entrepreneur Club – the LiveBox Lab provides both marketing and R&D support for 3rd party innovation.
I would not be surprised if the Femtocell industry adopts the approach of providing a 3rd party API and eco-system to support the development of innovative applications. Success in this area would improve the perception of the business case for femtocell adoption. A common, standardised API with full support (documentation, testing, AppStore) are essential to gain momentum.