"app development on the mobile web sucks less than the alternative"
And Michael adds a very witty obituary:
"In loving memory of the mobile applications business. Adoring child of Java, Psion, Palm OS and Windows Mobile; doting parent of Symbian, Access Linux Platform, and S60; constant companion of Handango and Motricity. Scared the crap out of Microsoft in 2000. Passed away from strangulation at the hands of the mobile industry in 2008. Awaiting resurrection as a web service in 2009. In lieu of flowers, the family asks that you make a donation to the Yahoo takeover defense fund."
In general, I agree. Barcelona was full of widgets & web services, and I've been telling my handset software customers for some years that they should be working on the best browser implementations they can.
I've also often pointed out that future operator/vendor "next gen walled garden" initiatives like IMS are driving the most creative developers towards AJAX and Web 2.0. And I certainly feel that the concept of "aftermarket" mobile applications (ie downloaded by end users) is almost an irrelevance to most users, apart from maybe games. Normal people don't download software to their phones, even if Nokia does try to rebrand them "multimedia computers". Come to that, normal people don't care if their phone is "smart" or not - most smartphone owners neither know nor care what's powering their phone or what it's theoretical capabilities might be.
There's also been no mechanism for popular mobile apps to spread virally - you can't just tell all your friends how cool something is, and suggest that they download it now. You need to know which phone, which OS/firmware, which operator, which data tariff etc etc. There's friction. The web is easier, at least once everyone has a decent-ish browser and a flat-ish data plan (give it 3 years in developed markets, I'd say).
I don't think the situation is quite that clear-cut though, and that there will be plenty of reasons to continue using native apps on smartphones, together with other virtual machines and on-device portals like Java, Flash, BREW and SurfKitchen.
In particular, the following use cases remain for native (or virtual machine) device applications:
- Pre-installed applications at the factory.
- Pre-installed applications by the operator or other service-provider (eg RIM)
- Pre-installed applications by the retailer or distributor
- Certain markets are a bit more application-savvy (eg the US, with its history of PDA users), although other markets still view installing (or even thinking about) handset software as a geek-only activity.
- Applications installed by enterprises for their end users
- Applications like VoIP that need access to underlyig device APIs and capabilities like codecs.
- Applications (maybe IMS apps) for which carriers are able to design & enforce a complex over-the-air automated download & install process. Likely to only work in situations where the user has a deeply-customised phone, rather than a 'vanilla' device.
- Games, and even then only by certain demographics.
- End-to-end services coupled to specific devices or a limited range, rather than generic handsets (eg BlackBerry, Amazon Kindle)
Bottom line - I totally agree with Michael that web-based applications are becoming much more important relative to "installed" mobile apps. But I think it's a little early for the obituary, deeply amusing though it is.