In infinite wisdom Joseph Lazio answered:

"R" == Rich  <someone@somewhere.com> writes:

R> One thing not usually mentioned is that what is listed as the
R> planetary mass is the lower bound, assuming that the plane of the
R> planetary orbit cuts the earth. If it does not then the planet's
R> mass is greater. While it may be a good choice to list the lower
R> bound, I tend to think that the odds of being exactly in said
R> planets orbital plane very low. If it were true we could see
R> transits of every planet. And only one planet has been discovered
R> this way. So almost certainly all the planets have a greater mass
R> than the lists show, how much more depends upon just how far away
R> from their orbital plane we are observing at, and there is no way
R> to know from doppler observations (...).

This is both true and relatively unimportant.  It is correct that what
is measured is a lower bound on the mass of the planets.
Specifically, what is measured is M*sin i, where i is the inclination
angle between the line of sight and the planet's orbit.  On average,
though, we would expect that the inclincations would be distributed
randomly between 0 and 90 degrees, meaning that the typical value of
sin i is about 0.7 or that the quoted mass value is within about 50%
of the actual value.

What is true that just because the lower bound for mass is terrestrial,
the planet may not be. WRT SETI, I think  this is important. Can you
tell me why you do not think it important?

That is not true.  From the data we have now, it could be one
intelligent civilization per billion galaxies, or a billion
intelligent civilizations per galaxy.  If we found another
intelligent civilization in our galaxy, then the bottom end of this
range is ruled out, statistically speaking.

R> Statistics if properly done are descriptive of group distributions,
R> they are not predictive.

This seems a bit extreme.

The simple truth is extreme? How so?

I can use the known statistics of a group
to predict the likely properties of a new member that is discovered.

You can 'predict' that the new member is average. But except for Richie
in Happy Days (according to Mork), no one is average. That family with
2.4 kids and half a dog does not exist. Need I explain why? :-)

The only cases where you can make reasonably accurate predictions are
when the thing being predicted is not random. You can reasonably predict
that the next NAACP president will be black, and indeed this probably
applies to the membership. But this is not much of a prediction.

You can reasonably predict that the next president of N.O.W. will be
a woman, and indeed you can make the same reasonable prediction about
it's membership.

The point is that these things are not random, For those things
that are random, you can make no reasonable prediction. Will
the next democrat be black or white? Male or female? Black hair or
brown? You tell me. Seriously, you think statistics are predictive,
so predict it for me (for all three criteria).

Similarly, as Louis says, finding more members of a group improves my
estimates of the group statistics. 

Estimates? It does indeed give you more information about the group.
But the fact remains, stats are descriptive, not predictive. Will more
information allow you to predict the mass of the next planet discovered
(or more properly, the lower boundary)? No, it will not. Except WRT
selection biases (we can only detect massive planets) you can't say
anything. Oh yes, there are upper boundaries as well, anything
over, what is it, 13 Jupiter masses is considered a brown dwarf.

This does not necessarily follow. Even if life exists everywhere,
there is no guarantee that looking will show it.

The two statements above are both true, and do not contradict each
other.

R> I dispute that there is any direct correlation between looking and
R> finding. Many who look find nothing. Some who do not look find many
R> things.

Those who don't look find nothing.

This is not true.

Isn't that a correlation?

In your case, I think we have a definition, not a correlation.

Rich