Consider the two products above, the Nokia N9 on the left and the Nokia Lumia 920 on the right. Both employ Nokia’s ‘Fabula’ design language and, at first glance, are so similar as to be indistinguishable. Look closer, however, and which seems the most visually appealing?

The N9, left, resulted from a centralised, uncompromising approach to design – and it shows. Nokia controlled the software, components and external design. There is a notable absence of visual clutter: the contents of the screen float in an unbroken sea of deep black. The cyan polycarbonate case provides the frame and the Nokia logo captions it. The effect is artistic, as if the user is observing a painting hung in a gallery.

In contrast, the Lumia 920 – shown here in the AT&T variant – makes several compromises. It runs the Windows Phone software and, therefore, Microsoft dictates the inclusion of 3 illuminated soft buttons beneath the main screen. It is currently exclusive to AT&T in the US, so the operator mandates the inclusion of its logo in the top left. This adds visual clutter and also necessitates re-positioning the Nokia logo from centre to right, removing the captioning effect seen in the N9.

Small differences, but the result is the N9 exhibits symmetry and communicates a purity of purpose for the user’s first touch interaction, while the Lumia 920 appears asymmetrical and to have multiple functional starting points.

Much of this was achieved only because Nokia had centralised control of the overall experience on the N9. For instance, the absence of soft buttons or a dedicated home button on the front face was made possible by a software innovation allowing the device to be woken by double tapping anywhere on the screen, something which could only be implemented at operating system level.

There are, of course, many reasons why the compromises exhibited by the Lumia 920 were necessary – some of which may benefit customers in other areas of user experience – but it raises questions about the proportional link between size of design committee and the ability to achieve arresting aesthetics. It is a theme we’ve been exploring for some time in MEX Pathway #6, which challenges participants to re-consider form and material usage in digital product design.

Marek Pawlowski

Marek Pawlowski is the founder of MEX. Since 1995, he has focused the MEX business on helping digital industries to develop better, more profitable products through improved understanding of user behaviour with mobile devices and wireless networks. MEX is best known for its events, research and consulting, which balance commercial, technical and user insights to define the future of mobile user experience.

Recent Posts

Design Talk 80. Digitally sustainable; Mirren Fischer, Founder, Brilliant Product Design

Mirren Fischer is someone who finds herself at the confluence of some really big challenges:…

5 years ago

Design Talk 79. Our future towns; Dan Phillips, Royal College of Art

Poetry and place...listening and lockdowns... Dan Phillips, who led the Royal College of Art's (RCA)…

5 years ago

Design Talk 78. Mark Baldino, co-founder, Fuzzy Math

Mark Baldino's path to co-founding a Chicago-based design agency wends its way from a degree…

5 years ago

Design Talk 77. Paul Campbell, Chief Product Officer, Centaur Robotics

Paul Campbell, a former chief designer at Ford, is now re-imagining the wheelchair at Centaur…

6 years ago

Interaction unintended

Opening the oven released a cloud of steam. I should have been expecting it. It's…

6 years ago

Design Talk 76. Placeful computing; Jared Ficklin, Chief Creative Technologist & founding partner, argodesign

Jared Ficklin's approach of 'think by making' has led him to produce remarkable digital design…

6 years ago