Thanks for not taking the flamebait.  For what it's worth, some of the
changes you and your comrades have made that I stumbled into looked
meritous---the fourier scaling and the work to integrate PNGish gamma
profiles rather than the older approach to gamma that if i understand
right causes more quantitization/rounding errors.

I can't contradict anything in your post.  To sum up my argument,

 (1) the current netbpm documentation made netpbm no longer useful for
     my application.  in spite of the code's being there, the package
     itself was not useful to me, and I ended up not using it.  The
     problems for me were both with the availability of the manual,
     and with the text itself.

 (2) the snooty defensive placeholder man pages should not ship with
     NetBSD at all.  It's not TRT.  I believe that incomplete,
     unmaintained man pages are the best answer for now, but other
     defensible resolutions are possible.

 (3) no revision control for documentation is a controversial
     trade-off I absolutely do not agree with.

Thanks for your work on netpbm.  Sorry my momentary frustration with
what I experienced as serious regressions interfered with my being
properly appreciative.

-- 
Le fascisme est la dictature ouverte de la bourgeoisie.
		-- Georg Dimitrov