Lots of people quote the Miaria study (via "Code Complete"), but no
one ever actually reads it.

der Mouse suggests:
> Or, perhaps, the code was written at four spaces, then mechanically
> reformatted to six - which, if the original coder was any good, will
> probably damage its comprehensibility for reasons unrelated to the
> indentation.

Which, if you read the article, is exactly what they did. (Well, I
don't remember if the original code was the 2- or 4-space indented
version, but that doesn't matter.) The article mentions that this
caused some lines to exceed 80 characters. They're not completely
clear on what they did with those lines, but a few hints and some
reading between the lines imply that they just wrapped them back to
column 0.

This, combined with the fact that they say right at the beginning that
their intention was to prove that small indentation is better, pretty
much invalidates the results as far as I'm concerned.

-- Dan