In article <ctv256$tf6$2@gnus01.u.washington.edu>,
"Max Power" <mikehack@u.washington.edu> wrote:

SETI Software's ability to detect [very weak] pulsars, or pulsars in 
general -- has it been fully quantified?

If you mean Berkeley Space Science's SETI@Home client software, it is
designed to not detect pulsars (pulsars would have a relatively large
bandwidth which would cause them to be suppressed by the baseline
smoothing (and probably the initial quantisation) process)!  Rejection of
natural sources in general is a design aim.  There is another BOINC
project, AstroPulse, that is designed for this and uses more of the
bandwidth from the original SERENDIP receiver channel.

I understand that the 'pulse' search function only works for a very narrow 
time window, and has other flaws in its design. 

The time window is limited to around 24 seconds by the economic
considerations that make SERENDIP and SETI@Home possible.  Berkeley have
no control over the pointing direction of the telescope, which means
that they can make almost continuous observations.  If they controlled
the direction, they would have to fight for time with all the other
astronomy and atmospheric research projects that wanted to use Arecibo,
and couldn't hope for more than a few days each year.