Some data on "Re: Gzipping man files"

[Dirk.Eddelbuettel@qed.econ.queensu.ca]

  Richard>  Good point, but the counterargument is that none of those have an
  Richard> additional heavy formatting step to go through.  IMHO groff is
  Richard> already slow enough; running a gzip alongside it would, I imagine,
  Richard> make it even worse.

That seems to the other way 'round. Let's take the man page for bash, I think
that is the biggest beast that we have hanging out here.

edd@miles:~> cp /usr/man/man1/bash.1 .  

edd@miles:~> time groff -man bash.1 > /dev/null
6.91user 0.13system 0:07.34elapsed 95%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (370major+201minor)pagefaults 0swaps  

edd@miles:~> gzip -9v bash.1
bash.1:                  69.8% -- replaced with bash.1.gz 

edd@miles:~> time zcat bash.1.gz | groff -man > /dev/null
0.05user 0.01system 0:06.56elapsed 0%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (75major+45minor)pagefaults 0swaps   

So it is _faster_ to use the compressed man pages.

--
Dirk Eddelb"uttel                              http://qed.econ.queensu.ca/~edd