The internet is full of articles telling - why browser X is better than browser Y or vice versa. This isn’t one of those. The daily job of web development companies is to create solutions that work on any browser, which is used by large or significant part of our customers' audience.

There is one important thing our customers should be informed of – the more browsers have to be supported – the longer is the development and, therefore, the larger the costs.

Let’s find out why. In theory the W3C provides several standards that govern HTML and CSS that should all be implemented by all browsers. Unfortunately, the standards are large and complex and therefore not easy to implement and as a result the browser makers have implemented them differently.

That’s not all though – since from time to time the browser makers decide to implement new features for their users and developers to play with. And of course the competing browser makers decide to copy features from one another. These new features though, are more often than not proprietary, and not described by any standard, so everybody does them in accordance to their own gut and skills.

Furthermore, let us not forget, that browsers are only programs, and programs are made by programmers, who tend to be human, and, as we all know, humans tend to make mistakes. Accordingly – the browsers are quite simply full of bugs that have to be avoided or worked around by the web developers.

At the end the result is a huge mess that the web development agencies have to deal with. And so they tend to pass on the costs of that mess to their customers. So here is a logical question, especially in the time of economical recession – perhaps some browsers can be ditched as a cost saving measure? The answer, as it usually is in these cases, is – it depends.

Mostly it depends on your audience, and the cost of supporting the browser. There are two things requireded for calculations so you could choose whether a browser should be supported or not:

1. The browsers used by the audience of your web site.
2. The “cost“ of the browser, or relative time of how long does it take to develop and find bugs on a given browser

To find out whether a browser is worth supporting , we have to take the value of the audience of that browser (a browser may be unpopular, but can be used by say the CEO of a VERY IMPORTANT CUSTOMER) as the benefit and divide it by the cost.

The next logical question is – how do I get those numbers? If you already have a web site, you can get the audience (benefit) data using any web statistics tool that your site is using.

To calculate the benefit we just take the visits data for that browser and possibly multiply it by some adjustment constant, if that constant is known (say we know that the Safari users are the big spenders and the Firefox guys never buy anything).

Let’s take a look at the browsers and set some example costs for them:

• Browser: Firefox 3
  Cost: 20.
  The latest version of the most popular browser between the web developers. The costs for supporting it are small as the browser has excellent standards support and wonderful development tools have been written for it. Both development and debugging on this browser is quick and painless.
• Browser: Firefox 2
  Cost: 30.
  The previous version of the developer’s favourite. Is usually found to be used by those customers whose computer geek grandson is overdue for a visit. The same excellent development tools are available, but there are features that the browser does not support, and it has bugs that have been fixed in version 3, therefore additional testing is required.
• Browser: Internet Explorer 8
  Cost: 40.
  The latest browser by the largest software company in the world. Is included in Windows 7. Supports most of the web standards, has solid development tools. Unfortunately though, it also has some minor bugs, and it supports several features in its own little way.
• Browser: Internet Explorer 7
  Cost: 100.
  Previous version of Internet Explorer. Included in Windows Vista. The standards support is incomplete and the development tools are a bit awkward. Most web sites require a special treatment for it.
• Browser: Internet Explorer 6
  Cost: 200.
  The record holder for most headaches in web development companies. Included in Windows XP. Was for a long time the world’s most popular browser. It has weird ideas regarding standards, full of display bugs and sub-par development tools. Always requires special treatment, since even if a web site works perfectly in other browsers, something will look or work wrong in this one.
• Browser: Safari
  Cost: 25.
  From the people who brought you iPod, iMac and iPhone. It has great standards support. The development tools could have been better, but then again – it’s not often you have to use them, as most of the time, if it works in Firefox – it works in Safari.
• Browser: Opera
  Cost: 25.
  The alternative browser from Norway. For some reason it has quite a following in Eastern Europe and Russia. It has great standards support, and good development tools. There are known missing features, but these are usually not required anyway. Most stuff works just like in Firefox.
• Browser: Chrome
  Cost: 25.
  A browser by Google. From the developer’s perspective – it’s a Safari.
  
  

The cost/benefit graph shows data for a particular web site.

The bottom line – if your audience allows for it – ditch IE 6, supporting it is expensive and is not recommended.

Back