A hundred million reasons for revamping site access

Whatever the outcome of the poll that gives European Windows users the chance to switch from the Internet Explorer default, it is bound to be another nail in the coffin of single-browser development, says Bruce Lawson.

You phone a new restaurant to make a reservation for 3pm. "Sorry," the receptionist says. "We only serve even-numbered parties. It makes it easier for us to organise tables." So you eat somewhere else.

The next day, you visit the supermarket for bread and milk. Your £5 note is rejected by the cashier. "Sorry," he says. "We only take 50p coins. Makes it easier for us to count at the end of the day." You do your shopping elsewhere.

Sounds absurd, doesn't it? The customer is always right, we are told. Yet that is what some companies' websites have been doing for a decade now.

Website restrictions
At best, they give you a message: 'Best viewed with browser X.' At worst, they tell you to get another browser and turn you away. You must use this browser, in this screen resolution with these plug-ins if you want to do business with us, they say.

During the early browser wars that was just about excusable, because the browsers were relatively incompatible. Ten years later, it is not only unnecessary, it is counter-productive — it is highly likely that visitors come to your site today running any one of five major browsers: Internet Explorer, Opera, Firefox, Chrome or Safari.

Very soon, the pot will be further stirred. Before 18 March, everyone in the 30 countries that make up the European Economic Area, who has Internet Explorer as the default browser on Windows XP, Vista or Windows 7, will be asked to choose a new default browser, after an investigation by the European Commission into the inclusion of Internet Explorer in Windows.

The browser-choice screen will also be offered to all new purchasers of Microsoft operating systems, with the main five browsers displayed in random order with their logos, together with another seven also on offer — many of which use the Internet Explorer Trident engine with a different skin.

Poll predictions
It is very difficult to predict what the result of more than a hundred million votes will do to the desktop browser landscape. Research carried out by Google last year shows that the vast majority of people do not know what a browser is, let alone that they have a choice.

Many will choose Internet Explorer again, out of familiarity. But it is likely that, once presented with an opportunity, many of these users will reject their former browser after all the bad publicity surrounding its security flaws, which allowed the recent high-profile Google China hack to take place.

Google is presumably hoping that brand recognition will send some downloads their way, which is why you cannot move for Google Chrome advertisements in newspapers, town centres and underground stations.

If Internet Explorer's dominance were ever an excuse for only coding to a single browser, it ceased to be a viable development strategy some years ago and, after the browser ballot, its dominance will be further eroded.

Mobile devices
That is not the only nail in the coffin of single-browser development. A rapidly increasing number of web visitors use mobile devices. Gartner research suggests that by 2013, mobile phones will overtake PCs as the most common web-access device worldwide. And don't forget...

...visits from devices such as games consoles such as Nintendo Wii and DSi, web-enabled televisions and in-car browsers.

So a web developer is left with a tough dilemma: it is obviously untenable to develop for only one browser, but it is completely impossible to test all browsers, on multiple operating systems and endless mobile phones and devices — particularly when new ones seem to arrive on the market every week.

It is superficially tempting to check your server logs and code only for the browsers that most of your visitors use. But closer scrutiny shows that this approach is a bad strategy. It becomes self-fulfilling. If you only code for browser X and do not care that it fails in browser Y because browser Y is so rarely used by your visitors, then browser Y's users will give up trying to access your website.

Access fallacy
The same argument is often used to justify not amending websites to allow access to people with disabilities. The fallacy of this is demonstrated by the owner of a restaurant only accessible by a flight of stairs, who says: "We don't need to add a wheelchair ramp, as there is no demand: we never get wheelchair uses in here."

The way to ensure your website can be accessed by all browsers, across all devices is to use web standards — the recommendations published by the W3C, the industry consortium of academics, business and individuals. The organisation has guidelines on making sites mobile-friendly, helping your sites conform to the Disability Discrimination Act, as well as writing the specifications for some of the basic building blocks of websites, such as HTML and CSS.

The government offers browser-testing guidelines, and the BBC has a useful browser-support policy.

All engineers and computer programmers know the value of standards, and in the newer discipline of web design, adherence to standards is the way to distinguish real professionals from the 'Your very own website for only £199!' cowboys. Standards documents are not very easy to read, unfortunately — but there is an excellent free course on them from the independent, community-driven organisation Web Standards Project, called Interact.

Standards adherence
These days, all the major web browsers have a good baseline of standards adherence — they even compete on their standards support. In the evolution of the browser market — and potential revolution that the Windows ballot choice will bring about — using web standards is the way to make sure your website is available to any visitor who comes looking for it, rather than rudely demanding they use the technology that you want them to use.

In today's challenging economic climate, this is particularly important: on a worldwide web, someone else is always offering the same service as you.

Disappointed customers of our fictional supermarket that only accepts 50p coins simply go to another shop, and that supermarket will soon go out of business, because sending customers into the arms of the competition does not succeed as business strategy. But no-one would be foolish enough to do such as thing in real life. Would they?

Bruce Lawson works as an open web standards evangelist for development company Opera Software. He has been involved in standards and accessibility since 2002. The views expressed in this column are his own. You can follow him on Twitter.