productivity references & data

[agate!spool.mu.edu!uwm.edu!linac!pacific.mps.ohio-state.edu!cis.ohio-stat e.edu!news.sei.cmu.edu!ajpo.sei.cmu.edu!progers@ucbvax.Berkeley.EDU (Pat Roger]

We need to supply "proof" that the standard 10 lines per day for debugged,
tested and *documented* code is real -- or incorrect, and what the
number really should be, if incorrect.  Anybody have good data or
references?  For that matter, where did I get the idea that 10/day isthe
so-called standard?  (We do better, but that isn't the issue.)

Thanks in advance,

Pat 
progers@ajpo.sei.cmu.edu