There was quite a debate in the office with the announcement of Amazon’s HTML5 version of the Kindle Reader at the beginning of the month. Initially, it focused on whether this was just a direct two-fingers to Apple’s rules and regulations about commission on in-app payments for content, and then on the reasons why Chrome along with Safari were the first two browsers targeted (conclusion: offline browser support).
We then, however, got into a deeper discussion about why and when an installed app should be used instead of just targeting the browser (especially if HTML5 is the underpinning technology used). Overall, I’m of the view that the future is undoubtedly online, and most of the time serving live to the browser is the way forward. However, the future isn’t quite here yet, and totally Cloud initiatives like ChromeOS, by example seem both too ahead of their time, and yet vaguely anachronistic (as I described in my post on Chromebook).
So there seem to be some key times when an app installed locally works best – here’s some of the cases that we’ve discussed:
To allow offline access:
Whilst I think that the cases of people using laptops offline these days are increasingly rare, on mobile devices we frequently find ourselves off the network. In those cases, locally cached data accessed by the local app makes perfect sense (if there is one thing that really irritates me it’s a locally-installed app that doesn’t cache data for offline on a mobile).
To cope with big data:
Before taking up my current role, I ran IT for a big marketing agency based in London but with offices across the globe. A constant issue that we had was of shipping video and 3D content from office to office: whilst the Cloud could provide fantastic opportunity for rendering processing grunt, getting the data in and out would be a complete pain in the backside. Local apps are often still the answer to cope with vast data volumes.
To use a device’s hardware:
A local app that takes advantage of gyroscopes, accelerometers, GPS or other hardware increasingly found in the average smartphone seems to make sense. (As an aside, if someone could create an app that would turn a Windows Phone into a customizable touch surface and control for Windows, I’d be eternally grateful).
To provide an audience:
A more publisher-centric one, this. App marketplaces can provide an easier route to get a developer’s work to an audience than a mere website.
I’m sure that there are other cases when the local app is of advantage, and I’m also sure that this is going evolve dramatically over the coming years; however, as Amazon have shown with the HTML5 Kindle, assumptions that the app should always be local need to be questioned…