For years web development was quite predictable. The resolution of the average screen slowly but steadily increased, bandwidth became less of an issue and everything was good. Then smartphones became mainstream. Suddenly we have to make sure our websites are also accessible on small screens. And bandwidth may also be limited to a few kilobytes per second. In other words: new challenges. But how are we responding to them?
An approach that is quite popular these days is to create a separate
mobile site. In most cases visitors of the ‘desktop’ site of
example.com are redirected to the mobile version on
as soon as the site detects (or thinks) that the client is a mobile device.
What is a mobile device?
And there we already see the first problem with this approach. What exactly is a “mobile device”? We can probably agree on a smartphone. But what about tablets? An iPad, for example, has a resolution of 1024 x 768 pixels. Isn’t that the same resolution a lot of designs are made for?
So is an iPad a mobile device? If your answer is no, what about an iPhone 4? With a resolution of 960 x 640 it gets awfully close. So is every smartphone a mobile device? Perhaps we should focus on the size of the screen. But where to draw the line? 10 inch, 7, smaller?
Perhaps the term “mobile” should not be defined just by the specifications of the device, but by how the device is used. For instance, you would probably use your smartphone while waiting in a queue, but not your laptop. However, using that definition will open another can of worms.
The life of a modern web developer is even more complex. We may want to optimize the content of the mobile site or application for low bandwidth conditions of 3G or EDGE networks. But user agent sniffing doesn’t help here. Smartphones can also be on Wi-Fi. And the other way around, laptops can be tethered and thus have to suck your heavy desktop version through a straw.
In other words, the user agent or even the type of device your are working on, has nothing to do with how much bits per second can be consumed. As a result you should probably try to keep the size of your site to a minimum either way.
There are bad implementations of mobile websites out there. A simple example: you see a link to an interesting article on Twitter. You click on the link and you are directed to the homepage of the mobile site. Good luck finding that interesting article on the mobile version…
Or the other way around: you are reading a nice article on your phone and want to send the link to a friend. Now he’s stuck with the mobile version of the article on his 22 inch monitor.
Another complaint I have is that the mobile sites are often dumbed down. For example a restaurant that only shows the menu and contact information on the mobile site. Perhaps I want to get a feel for the place and like to view the photo gallery. But now I have to switch to the full version. If that option is even available!
Instead of having a separate mobile site, you can use responsive web design. With this approach you use css media queries to change the layout of the page, while serving the same HTML and the same content. In other words: you don’t necessarily care about the type of device your visitor is using, you set the rules on how things should be displayed on certain widths and let the browser handle the rest.
With this approach you do not design separate sites for distinct devices, instead you design for a range of resolutions. So if next month a new device comes on the market, your site will probably be ready for it. (The obvious exception is when the new device does not fall in the ranges you had anticipated, e.g. a television with a really high resolution.)
So is a mobile site always a bad thing?
Don’t get me wrong, having a (separate) site specifically for mobile devices certainly does have its benefits and can be a good solution. But in my opinion it should not be the first option when building a website or web application that should be “mobile aware”.
Actually, I do not have a conclusion… Although I think responsive web design provides a good solution, it is not the holy grail. As I said at the start of this article, we live in exciting times. There is a lot we still have to discover!