Roughly estimating the diversity of life has for a long time been seen as a challenge. Sequences obtained from the environment by direct amplification are now seen as the sole mean to provide information for at least 99% of the prokaryotes in natural communities. Recent massive parallel sequencing technologies now allow to sequence nearly every sequence resulting from such amplification. Are we ready to deal with such a deluge of data? Using three recent publications (16S rRNA genes amplification in soil and seawater followed by massive parallel sequencing) as case examples, this review analyses the state of the field.