Unless you’ve spent the last few months trapped in a cave you will undoubtedly have seen, read or heard that “cloud computing” is on its way and is likely to affect the way that many of us use our computers and interact with the internet over the coming years. Much has been written elsewhere about the implications of cloud computing in general and whether it is indeed destined to transform the way we use the web but, at this juncture, I am particularly interested in exploring what the impact of cloud computing will be on mobile applications.
For the uninitiated, it is probably worth reiterating briefly the key characteristics of cloud computing. In layman’s terms, cloud computing simply refers to the use of powerful shared computing resources which are accessed remotely, typically via a web browser over the internet. Users don’t need to know (or care) where the servers are located or where the programs they are using are running – they just need access to a web browser to use the service from anywhere in the world. In practice, the term cloud computing has grown to refer to a number of related capabilities that can broadly be summarised under following categories:
i) Infrastructure-as-service “IaaS” – typically virtual servers (e.g. Amazon EC2, Rackspace Cloud Servers)
ii) Platform-as-a-service “PaaS” – various services for software development and deployment (e.g. Google Checkout, Force.com)
iii) Software-as-a-service “SaaS” – fully hosted applications accessed via a browser (e.g. Webmail, Facebook, Google Apps, Salesforce.com )
It is the last category, SaaS, that I wish to focus on because it represents the visible face of cloud computing that most people have already experienced. The advantages of web-based applications such as Gmail, Twitter, and Facebook are pretty obvious – there’s nothing to download to your PC, you can access them from anywhere, they’re constantly being enhanced and every time updates are made they become instantly available to any user when they next login.
For business applications like Saleforce.com’s online CRM tool there are further advantages – such services are highly configurable and scalable so they can accommodate many different types of client from very small to very large with either simple or highly complex requirements. Furthermore, the commercial model employed by most SaaS suppliers (i.e. monthly subscriptions) is attractive since the cost of entry is low (or zero) and the costs associated with acquiring and maintaining computing infrastructure to host the application are completely eliminated.
So, how will this all affect the use and spread of mobile applications? Over the past couple of years, we’ve witnessed the unprecedented success of the iPhone and Apple’s App Store and this has clearly demonstrated beyond doubt that there is a voracious appetite for mobile applications. Apple has led the way, but we are now seeing a plethora of competing app stores being announced by other players. Notable examples include Google’s Android Marketplace, Nokia’s OVI Store, RIM’s Blackberry App World, Symbian’s Horizon, Microsoft’s Marketplace and the Samsung Application Store.
It seems highly unlikely to me that all these initiatives can succeed, but at the same time, it is a healthy sign that consumers will have more choice, and a wide range of applications will become available to users of many different types of mobile (i.e. not just the iPhone). However, in this new world, where users of all device types (from the humblest Pay-as-you-Go handset to the latest, feature-laden Smartphone) are able to access mobile applications I believe that downloading them from app stores is not the only way forwards. The cloud computing model provides a highly attractive alternative which actually turns out to be ideal for supporting (relatively) low powered computing devices like mobile handsets. Whilst power-users with top-of-the-range Smartphones may be perfectly happy downloading apps, the “average” user with a basic handset is likely to find that using cloud based applications via a browser is both easier and far better suited to the limitations of their phone. Less computing horsepower and less storage is needed and, as mobile network operators continue to increase data speeds, performance can only get better and better. Already today there are some fine examples of cloud based mobile applications such as Gmail’s mobile portal which provides an excellent email experience entirely via a browser.
Another significant factor to consider is that as the mobile application market matures many commercial organisations will recognise the need to mobilise core business applications. Unlike the majority of “apps” that are being downloaded today, most business applications are more complex and sophisticated and they require proper integration with back-end systems. The SaaS model of delivery described above is therefore ideal for this category of application and will work equally well for mobile devices because of the “zero footprint” required on the handset coupled with the flexibility and scalability available when hosting the application in the cloud.
Over the next couple of years, we are also likely to see a number of technology enhancements which will continue to encourage the development of cloud based mobile applications. Open standards such as BONDI, OneAPI and HTML5 ( http://bit.ly/7LxXdn ) will all help, making it easier for developers to build cloud based applications that can be used across a wide range of mobile devices.
In summary therefore, I predict that cloud computing is highly relevant to the world of mobile applications, is particularly well suited to serving the large numbers of mobile users who do not possess a Smartphone and that it is likely to become a parallel medium for delivering mobile applications to rival the app store approach. Welcome to the cloud!