Cennydd Bowles – Designing the Wider Web
The Cambridge Usability Group was delighted to welcome Cennydd Bowles to talk on Monday 6 June. Cennydd is well known for his UX work at the agency Clearleft in Brighton, his involvement in organising events like dConstruct and UX London and his book Undercover User Experience Design, which he co-wrote with his colleague James Box.
The topic of Cennydd’s talk was ‘Designing for the wider web’. Cennydd felt that the focus on designing for mobile was a bit of a red herring and that the design community now have much more to think about when designing for the web. In Cennydd’s words:
“The wider web will enforce us to embrace diversity”
Cennydd stressed that we need to consider different use cases and whether a website or application will work on multiple devices and screens. He then talked us through some considerations when designing for the wider web including context, display, inputs, connectivity, eco-systems and finally summarised the impact this shifting landscape would have on UX design deliverables.
The focus is on humans, not devices and tasks. Users might be in a static context e.g.using a laptop or PC at home or in the office, for business or personal use. Or they can be in a mobile context. Designers must research the context using ethnographic and other research techniques.
UX designers must consider the physical size, orientation, pixel density and whether the devices are suited to the task at hand e.g. reading or resolving. The fold is important but the pixel range could be anything form 300 – 3000 pixels. And we can’t guarantee that the window will be maximised. The design community needs to start working in liquid proportions/percentage widths, rather than the 960 grid and restrictions of 1024 x 768.
Cennydd explained that this shift will be challenging because our design tools are restricted to fixed sizes e.g. Omnigraffle, Axure. Although Luke Wroblewski advocates designing for mobile first, Cennydd felt that it was important to consider all devices, as well as the task at hand and context in which they are used.
Although the mouse is still a predominant input method, the track pad is starting to be used more and now recognises some gestural interactions. Whilst designers have been familiar with Fitt’s law they now have to consider the size of controls for touch pads. We also need to consider access keys and focus.
Cennydd highlighted the importance of good mappings of from gestures to interactions. Pointing and touching are natural mappings to a touch pad. And touch screens are starting to have some standard gestural interactions such as touch to zoom and swipe. Some gestures are less obvious e.g. shake to undo on the iPhone. And there are new interactions to consider, like free space interaction for gaming. Considerations here are more physical, such as the impact of waving your arms about above head height for any length of time (static loading).
As we are all still painfully aware, bandwidth implications are a big factor when designing for mobile contexts. A good example that Cennydd gave is where CSS is used to hide part of the page display on mobile devices. But the page content that is hidden still has to be loaded when the page first loads.
Apparently Bryan Riegar, a mobile design expert, has advocated the need to communicate data usage to uses so they are able to manage and conserve bandwidth from data hungry apps. Another legitimate use case is also the offline user.
Cennydd outlined three eco-systems that need to be considered:
- Cross-device experiences – it’s tempting to think that consistency is important but instead we need to think about coherency.
- Content shifting – an extension of the separation between content and presentation, designers need to consider that users will read our articles and watch our videos whenever they want on the device which is most convenient or best suited to the job.
- Bridge experiences – the bain of advertisers for example a second screen, use of laptops whilst watching TV for backchannel chat on Twitter, doing research whilst the adverts are on. Cennydd showed an example of a DVD rental service in the US called Redbox which has both a web interface in a virtual sense and a web interface in the physical world. Users can set up an account and browser DVDs. Redbox then tells you where your nearest kiosk is on the street and you go along, swipe your card and watch the film. Andrea Rosmini has heralded this as the advent of digital IA in the physical world.
To round off Cennydd outlined the impact that designing for the wider web would have on UX designers “beloved” design deliverables.
Wireframes do not scale well in this new context and instead we should use page description diagrams (or response description diagrams as Cennydd likes to refer to them) to highlight the modular components that need to appear on each page, their priority or importance.
UX Designers should be delivering libraries of modular components rather than templates. Cennydd used the analogy of providing a client with a set of lego bricks and instructions on how to build something with them. We should drive layout with content, so “the design flows beautifully around the content, not the other way round”.
Swimlane diagrams can be used to demonstrate how different personas will move between different devices and contexts for a seamless experience of your product or service.
Prototypes are probably more useful and important, and we must match the tool to the output, using the tool that is most natural. Cennydd explained how he uses a tool called Air Display on a Mac which enables users to use their iPad or iPhone as an extension of the display.
Processes will also need to change and we must think about inputs, outputs, context etc.
Critique will become an increasingly important part of the UX design process. UX Designers must try to ask the question “so how does this work with..” and challenge each other to find appropriate design solutions.
Usability testing may not be appropriate as you need to test on a variety of devices and in different contexts. So longitudinal studies such as diary studies may be more appropriate.
Above all UX designers need to embrace a new mindset. They must experiment, iterate and break assumptions. They must find out what works and what doesn’t work and then feedback to the design community. Cennydd warned against letting developers be the ones to set standards without empirical evidence!
In this brave new world, Cennydd advised us not to feel downhearted that just when we felt platforms and design patterns and approaches were becoming established and familiar. He likened the web to a difficult teen who is just finding out its true self. He left us with a positive message, to feel liberated and relish in the discovery.
Cennydd is currently writing a book called Designing for the Wider Web which will expand on all the topics covered in this talk. The book is due to be published next winter and I’m really looking forward to reading it.
This content is published under the Attribution-Noncommercial-Share Alike 3.0 Unported license.