Putting the menus either at the top or the left (or right or bottom) of a browser window won't do a thing to address the issue here, unless the user always maximizes their browser to fill their screen. In that case, flush left or right would be good from a Fitts' point of view. But most users don't use their browser this way.
You could consider enlarging your links and/or buttons so the hit space is larger. You could add a buffer zone around the hit space so if the user gets 'close enough' then that is equivalent to hitting the target. Space your hit spaces far enough apart so that users don't hit incorrectly. Run usability tests and time your users on common tasks, then iterate your design.
An important aspect of Fitts' law is minimizing the amount of movement the user has to do to get to the place they need to go to complete the next action in their task flow. Therefore you can cluster similar actions together so that the user doesn't have to move around the screen a lot to accomplish a given task.
An interesting part of Bruce's post was his mention of circular menus. Fitts' law tells us they'd be easier to use than linear menus. But usability, like so much else in life, is not so cut and dry. I would predict that the increased cognitive load that the user experienced when they saw a circular list of options rather than a linear one would slow them down so much as to outweigh the benefits of not having to move as much to get to their desired action. All things equal, circular menus may be more efficient than linear ones. However unless the whole world changes to circular menus - or at least, one whole computer environment, applications and OS included - introducing this in your application would slow your users down.
You could consider enlarging your links and/or buttons so the hit space is larger. You could add a buffer zone around the hit space so if the user gets 'close enough' then that is equivalent to hitting the target. Space your hit spaces far enough apart so that users don't hit incorrectly. Run usability tests and time your users on common tasks, then iterate your design.
An important aspect of Fitts' law is minimizing the amount of movement the user has to do to get to the place they need to go to complete the next action in their task flow. Therefore you can cluster similar actions together so that the user doesn't have to move around the screen a lot to accomplish a given task.
An interesting part of Bruce's post was his mention of circular menus. Fitts' law tells us they'd be easier to use than linear menus. But usability, like so much else in life, is not so cut and dry. I would predict that the increased cognitive load that the user experienced when they saw a circular list of options rather than a linear one would slow them down so much as to outweigh the benefits of not having to move as much to get to their desired action. All things equal, circular menus may be more efficient than linear ones. However unless the whole world changes to circular menus - or at least, one whole computer environment, applications and OS included - introducing this in your application would slow your users down.