I will discuss recent attempts to better constrain the nature of star forming galaxies at cosmic noon (z~1-3), the relation between starformation activity and gas reservoirs, and the role played by galaxy mergers. This will include galaxies across the main sequence, displaying various level of activity: from most active starbursts, to typical star forming galaxies, to quiescent objects.Do all these follow a universal law, and which one ? This requires, critically, an understanding of how to measure gas masses,which is notably a difficult endeavour. New promising results suggest that Carbon lines, both atomic and ionized, might be helpfultracers. On the other hand, at high redshift both quiescent and main sequence galaxies reveal themselves different from what we had been typically expecting.