It should be easier to "see" data in complex systems.
Back in the 20th century, I got fed up with poor presentation of IT product tests and specifications.
Scanning through tables of numbers, looking down a column for each product or across the row for each of the several criteriait took too long to figure out which product was best overall.
For that matter, it also took too much effort to figure out which tests actually revealed substantial differences, enough to make those tests worth the effort of performing or interpreting.
In search of a better way to look at those numbers, I wrote a spreadsheet script.
The script computed averages for each row, then colored the background of each cell in a table on a spectrum that ranged from red (half of the row average or worse) to green (twice the row average or better).
I made the colors light enough that the number in each cell could still be read, but the overall import of the data suddenly jumped out at a glance.
A row with little color meant a test that was not differentiating products; a column all in various shades of green highlighted a product that was superior across the board. It was a matter of using the right sense to convey the right kind of information.
I thought about that exercise when I saw this months item in the journal Nature concerning a musician who has a specific sensation of taste when she hears a particular musical interval.
For example, the sound of a minor second (a half step or semitone) tastes sour; a major second (full step) tastes bitter.
I didnt put quotation marks around the word "tastes" in that previous sentence because, to that musician, the sensation is just as real as if it were coming from something being eaten instead of heard.
The smell of user interface design.