I was reading a great post by Russell Beattie recently on handheld stylesheets and the great implementation that Opera have launched on their community portal recently. It took me back to my (not so distant) days finishing up my Masters degree with our digital project. As you may guess, my group did a mobile project.

It was a great little project and it really allowed those of us who’d been working in web for so many years to apply our skills to a different platform and develop some really interesting learnings. For me, I was pretty amazed by what we found with regards to Information Architecture and Interaction Design.

Being a university project, of course there was a lot more research involved than you’d usually have the budget of the time to do for a commercial build. This allowed time for me to definitively show that there were very, very few ID conventions when it comes to interface design for mobile web content.

Of course, in developing the ID, many of the conventions we applied were derived from our experience in web, then pared back and pared back until they’d fit in that tiny little screen format. In some instances, this was v. successful. In others… predictably, not such a good strategy.

For example – one of the conventions that it seemed natural to borrow from the web was the ‘click minimisation’ approach. To minimise the number of user ‘movements’ required to access content.

As you can see from the photograph of our ‘lo fidelity’ (aka paper) prototype testing above, we had the opportunity to do a lot of user testing in this project (another artefact of the academic environment, God love it). We found that, in general, click minimisation was unnecessary in mobile web.

Repetitive clicking, required for scrolling down lists etc. with many handsets, is ergonomically less intensive for mobile users. These people are used to text messaging – which is ergonomically more stressful – repetitive hitting the *same* key is easy. Users reported no problems with a list scroll requiring more than a dozen clicks.

Problems arose if users lost their sense of ‘place’ within the mobile application, and this is a failing of several implementations that we reviewed in our research phase. Due to the real estate limitations, often times the ‘locating devices’ were sacrificed in the interest of content or functionality – to the detriment of the usability of the site.

On the flip side, some ‘web’ strategies were effective. We did some interesting research looking at the best implementation of ‘paging’ interaction devices. Here an old ‘web’ model held fast, preferred by users over range of options that we’d presented as being more ‘native’ to the mobile device and its constraints.
We built our little community driven application in a xHTML site for people to access using the internet on their mobile phones. We wanted to create an environment where, ultimately, we would be able to devise a range of stylesheets that would optimise the delivery of the site to different handheld handset types, as well desktop web environments.

It was my opinion that everyone wanted to build things this way so I was surprised to read this from Russell:

I have to say I’m getting more and more convinced that this is the direction that the mobile web is going to go. No recoding markup, no “separate and always unequal” access to content, etc. Just include a different style sheet on your server and you can re-arrange your web-standard markup as you see fit. I’ve been trying to find “good” examples of XHTML-MP (WAP2) sites out there lately, and there really aren’t a whole hell of a lot I can find. Yet, if you look at the handheld-style on the Opera pages, there’s a lot to learn there in terms of designing for mobile size screens. The pages look clean and usable, yet they still have a compelling design to them as well. Imagine if the whole of the mobile web was like this instead of hanging on to the legacy of WML and the “page of links” format style? It’d be a much different mobile world, I can say that.

He’s right though. There’s not really that much going on out there in mobile web land. No one is really building good sites for web… they’re just building stuff they can sell to the telcos.

So, at least in Australia, let’s blame the telcos (although, its easy to understand their motivation – mobile is so easy to monetise). At the moment they get to take advantage of users ignorance and create these imaginary walled gardens through their ‘homepages’… most people don’t know how to navigate away from that and to content of their own personal choosing.

Is there an opportunity for someone to create an alternative to the telco’s homepages? Something big enough and easy enough and attractive enough to encourage users to get away from the telco homepage and into the wild blue yonder?

How much would the telco’s fight to stop this from happening? Probably lots. Could make for a v. rapid flip opportunity… not that I’m encouraging that kind of behaviour.

Argh. This is probably about three posts in one. But its my first in the mobile category, so, perhaps I’ll come back and tease out some of these issues into something less rambly and more meaty soon.

Anyone out there got any experience with UI design for mobile web? What’ve you learned lately?

Technorati Tags: mobile, handheld, stylesheets, Russell+Beattie, Mobile+Interface+Design, Mobile+User+Testing