"Boogers were already semi-standardized. User interfaces should follow the principle of "least surprise" — if people are used to a certain metaphor, icon, or behavior, you should honor that so people understand your product immediately. No one else had a mini-viewer."
It's amazing how many people misinterpret and/or completely make up their own definitions for the principle of least surprise.
"6. The mini-viewer doesn't convey more information than boogers do."
This is wrong, and seems to be the key point of what the 'mini-viewer' has over 'boogers' -- the miniviewer shows the shape of the text where the text has been changed. If you're familiar with the text, you have a decent chance of being able to look at the shape of the text in the mini-viewer and possibly know what that text says, vaguely if not precisely. The boogers just show the vertical position of the changes in the document. Thus, the mini-viewer does convey more information than the boogers.
The fact that it is a miniature representation of the document is also a plus -- the boogers are an abstraction where as the mini-viewer is a representation.
The main thing I get is that this guy is thinking about UI optimization in terms of how can the UI be optimized for the guy implementing it, rather than how the UI can be optimized for the person using it, which is exactly the wrong way to think about optimization.
> The main thing I get is that this guy is thinking about UI optimization in terms of how can the UI be optimized for the guy implementing it, rather than how the UI can be optimized for the person using it, which is exactly the wrong way to think about optimization.
I read the article as more of "this is the problem with interpreting 'do incremental development' as 'be a genetic algorithm.'" If you told a computer to come up with things that make the UX slightly better, it would come up with boogers before it came up with the miniviewer, because boogers are a local minimum, whereas the miniviewer requires "climbing the hill" of development-without-promise-of-reward. To put it another way, a half-implemented miniviewer would be much worse than boogers—so in order to come up with the miniviewer, you can't rely on "evolutionary" customer feedback, but rather have to use some "revolutionary" R&D as well.
Maybe I missed something, but I got the sense that the article was addressing how humans should design for humans -- computers designing for humans is a whole different story and doesn't seem to fit into the scope of what that article was addressing.
As for evolutionary vs revolutionary, I think a main factor in how UI design has progressed over the years has been how capable computers have been at rendering and processing UIs. But I think we're at the point (and have been) where computers are capable of rendering whatever UI is best, but a huge portion of our collective UI design knowledge is based on the time when it wasn't capable of that. So we really need to rethink the core of the visual computer interaction experience. There's a whole lot of useless cruft in every major computer UI (win/max/*nix) that needs to go.
"The main thing I get is that this guy is thinking about UI optimization in terms of how can the UI be optimized for the guy implementing it, rather than how the UI can be optimized for the person using it, which is exactly the wrong way to think about optimization."
Agree 100%. The boogers don't give you any context at all, this was a big improvement.
It's amazing how many people misinterpret and/or completely make up their own definitions for the principle of least surprise.
"6. The mini-viewer doesn't convey more information than boogers do."
This is wrong, and seems to be the key point of what the 'mini-viewer' has over 'boogers' -- the miniviewer shows the shape of the text where the text has been changed. If you're familiar with the text, you have a decent chance of being able to look at the shape of the text in the mini-viewer and possibly know what that text says, vaguely if not precisely. The boogers just show the vertical position of the changes in the document. Thus, the mini-viewer does convey more information than the boogers.
The fact that it is a miniature representation of the document is also a plus -- the boogers are an abstraction where as the mini-viewer is a representation.
The main thing I get is that this guy is thinking about UI optimization in terms of how can the UI be optimized for the guy implementing it, rather than how the UI can be optimized for the person using it, which is exactly the wrong way to think about optimization.