>>Hi to all,
>>
>>Here is the controversy:

12.345   -> 12.3
>>         -> 12.4

In my opinion, the first one (12.3) is the correct 'mathematical rounding' (to 1 decimal here). However, some colleagues say the second one is correct.
>>
>>Please, tell me that I'm correct! ;)
>
>Normally you want to round to the closest. 12.345 is closer to 12.3 than to 12.4 - calculate the difference to see whether this is true.

Agree.

>The only place where there can be controversy, and where there is really no "correct" method, but only agreements, is when the number to be rounded is exactly in the middle. 15.85 is just as close to 15.8 than to 15.9. The principle usually used in computing is to always round up in this case, but several other agreements exist.

I agree here that other agreements exist. For example to round .x5 up in case x is uneven and down if x is even, something like that. (I just read about this in one of my son's textbooks on floating point measurements, 1st year physics.)

But on purpose I referred to mathematical rounding. This afternoon (ET) I searched on the internet for the definition of this method, that we all probably learned in school. Google is phantastic, but nevertheless I had trouble finding an official definition of this 'obvious' thing.