Two interesting posts by Michael Nilsen about the Hirsch index: the first and the second. Michael notes that the h-index can be calculated to a good approximation from the formula:
h ~ sqrt(T) / 2,
where T is the total number of citations. Therefore it provides no new information beyond the total number of citations, and can not be properly regarded as a new measure of impact at all. Earlier I wrote a little post about Hirsch index (in Russian).