Reading through a report from mobiThinking.com, a point caught my eye; the paper suggested that “browser sniffing” is an essential feature to deploy for mobile-orientated websites with two odd but, to me, seemingly contradictory pieces of advice;

Don’t build your site for a minority audience- remember most people don’t have an iPhone.

Don’t build a mobile site and not invest the time to detect handsets and serve the correct version for the phone.

The iPhone might be a minority device, but there are a few caveats that need to go along with that point to put it into context; firstly, proportionally more users of iPhones are accessing online services than any other handset. This means both that mobile Internet users are more likely to get an iPhone because of the browsing experience, and that iPhone users who were attracted to the integrated iTunes, big colour screen etc. are more likely to try out the web access. The rate of growth has been phenomenal and from my own experience, even though I’ve had a Smartphone or PDA for about 5 years, I’ve never had even close to the use that I’ve had with the iPhone.

Secondly, every mobile device is a minority device. The market is highly fragmented, and that’s a trend that’s not going to go away- at least, not in the next 5 years. But the iPhone seems to represent a consistent platform; if Apple were to release a new iPhone with a different sized screen (for example), the existing applications and websites would need to be reworked- meaning it’s an unlikely move for Apple to make. Because the screen is a standard size, and the browser implements web standards well (better than Internet Explorer on the desktop, in fact) in developing an iPhone website, you’re not really developing “just for the iPhone”; you’re developing for the web.

The alternative is developing (or trying to develop) for every single minority device. A huge undertaking…

The point of the Web is that it’s a network of documents which can be read on any device, regardless of what sort of hardware, operating system or software it runs. The web page (in HTML) contains the information, a seperate system called CSS (Cascading Style Sheets) controls the way it’s presented, and a third system (Javascript) controls it’s behaviour (features like animation, drag-and-drop or AJAX.) This means that a web page with lots of graphics in the design can have an alternative CSS file to change the way it’s displayed on a handheld device- either because the visual design wouldn’t work on a small screen, or because the graphics would take too long to load on a slower (possibly metered) mobile connection.

Browser sniffing is something different; it means detecting what sort of device someone is using and then serving a version of the web page that’s optimised for that device. In other words, rather than letting the handset decide how a page would look best, according to an established set of open standards, the page decides on it’s behalf- often with no way to override it.

There’s no doubt that the wide range of mobile phone manufacturers, models and software versions make life difficult for mobile website developers, but the idea of browser sniffing goes against the basic idea of the web. A few years ago, the idea of a full HTML page being usable on a phone was virtually unthinkable. Now there is a new generation of handsets with web browsers that do a genuinely good job of making the web on a mobile a genuinely good experience. It’s up to the handset manufacturers- not the website designers and developers- to make the web a standardized platform, rather than the fragmented, unusable mess that much of it currently is from a mobile perspective. If websites follow the trend that we saw in the days of the desktop browser wars, then everyone loses out- especially the end users. Websites will be harder to develop, and new devices that don’t follow established standards will take time to gain support. We already have a protocol for mobile devices that don’t support web pages; it’s called WAP, and it’s a relic of the past. (But it works…)

Seeing this advice coming from a company who happens to have (and sell) a database of mobile devices- just the thing you would need if you were to implement their advice- leaves me with a bad taste in my mouth.

I did learn something interesting on their website though; Japan, the country with probably the most successful implementation of the mobile web, is number 10 on the list of countries with the most registered .mobi websites. Which, considering the subject of my last post, was something I found quite interesting.

Be Sociable, Share!