...I tried my best to understand this but I must be exceeding dumb because I couldn't extract any sense out of it.
Does it mean people can read things better if they are split up and not in one big block?
If so - no shit Sherlock.
I think you've hit the nail on the head there david. If only the researchers had presented their findings in a table, they might have been easier to understand.
Surely CSS is the way forward?
No Sheet Sherlock
Of course if you layout information in a structured and logical fashion it is simpler to understand it.
A table forces column or row identifiers and it groups concepts together.
On a related note, I was looking over at the Haskell docs, and boy a lot has been built up in a short space of time. And it is all well laid out, not W3C style (their docs tend to give headaches), but it doesn't address the basics. A simple table of builtin methods or a simple table showing the differences between ghci and ghc would help.
Same with changelogs, if you don't do a changelog everyone is left wondering, the best has changelogs and people read them, a simple unordered list of bullet points will be read.
A paragraph who has the time, they are for search engines.
Another statement from...
... the Department of the Bleeding Obvious!
Of course it's easier to read something if it's well structured, just as EULAs etc are easier to read if they're not crunched into a tiny box or you're forced to scroll by them in a window that only shows one line at a time...!