Jeff - Great article. I never liked STR() in any language. Hope that it just gets deprecated since there are many better ways to do what it does.

I would like to point out that contrary to the wording in your article STR is not actually rounding incorrectly. It is using the simplest and least accurate mode of rounding a Float that IEEE754 allows; round to zero. This mode gives us the common behavior experienced with float-to-integer conversions in mathematics. This is from the Wikipedia (not the best source on earth but one everybody can get to) article on the subject.

round toward zero (truncation; it is similar to the common behavior of float-to-integer conversions, which convert -3.9 to -3 and 3.9 to 3)

http://en.wikipedia.org/wiki/Floating_point#Rounding_modes

I am not sure why anyone would have a function use this rounding mode for float values without documenting it very well. Unless the of course the person that wrote this TSQL function had no control over its documentation:-P