We’ve been iterating quickly over the last couple of weeks to define a potential new tab screen for Firefox.
The feedback for the last major iteration was largely positive, and it seems like we are on the right track. But we ran into a paradox.
We believe that the new tab screen should have two main functions: (A) To show you the sites you are most likely to be interested in going to, and (B) to not distract you. That’s the paradox: by design success is when the pages we show are maximally interesting/distracting, but an explicit goal is to not interrupt your flow.
This iteration focuses on solving that paradox by proposing a solution that we’ve dubbed “the cognitive shield”.
The Cognitive Shield
No matter where we put the links to your most visited sites (and their latest news), it always seemed to be a distraction, based upon our own perception and the feedback from thousands of testers. Given that the bulk of those testers are multi-tasking-adept early adopters, we’d expect that feedback to be even stronger from more mainstream users. Our original thought was to place the links along the bottom of the page — outside your foveal vision. In practice, the peripheral vision proved too strong, and the links still drew your eye and interrupted your cognitive flow.
The cognitive shield hides the distractions until you move the mouse. Then the links fade in quickly.
Here’s the thinking:
If you are typing a destination into the navigation bar, then your locus of attention is on the place you are trying to go — so we should stay politely out of your cognitive way. On the other hand, if you are using the mouse than you will probably benefit from the mouse-based navigation aid, so show it.
Whether you are using the mouse is a good indicator of whether you are in a cognitive flow or not. That realization resolves the paradox: the links are there when you need then, and not when you don’t.
The design of the cognitive shield is a ring of 8 circles, each containing one of your top-visited sites. We think of it as a personal watermark.
We went through a number of metaphors for the cognitive shield.
Instead of a metaphor which always strains — what symbol represents frequently used sites? — we went for an abstract glyph.
There are a couple more features we’re in the process of adding. In particular: The ability to manually add a site is entirely broken. There’s no way to change the total number of frequently visited sites shown. There may still be encoding problems for non-roman scripts. Middle-clicks may be broken. The visual style needs a refresh. And the sites don’t remember their new positions post restart.