f^*_h: What about star class? Don't you have to put limits on the emission spectrum to support life?
Yes, but to my knowledge this is synonymous with star mass and luminosity, because spectral class (largely) depends on these attributes.
You do run into problems with too high a UV flux for large stars, but this gets into the mass range where the lifetime of the star would probably be too short for advanced life to develop. There shouldn't be a problem with a minimum UV limit, some have suggested it as vital for abiogenesis, but this does not fit for certain theories such as prebiotic compounds forming in the outer reaches of solar systems, and life originating in, for example, deep ocean vents.
Fortunately though almost 20% of stars are G and K class stars, if you add M class stars (some 76.45% of all main sequence stars), you get over 96% of stars. Of course not all of those stars will have conditions conducive to complex life (or conditions conducive to life in general), but it does mean that too many stars being too large, should not be much of a problem.
Why? It seems to me that the step from no life to primitive life is an enormous one (so f_v should be small), while the step from primitive to "intelligent" life is nearly inevitable (given the right optimisation strategy, i.e. random variation and natural selection), so f_i should be high. I admit that I don't have any more hard data to underpin my gut feeling than you or Drake.
I wouldn't call abiogenesis improbable; not only does it appear that life formed relatively early in the Earth's history (indicating that it forms readily in suitable environments), but it has been theorised (and in limited cases, observed) that not only are prebiotic compounds common, but the conditions required for abiogenesis might also be quite common.
The step from primative to sapient life wasn't inevitable for most of the Earth's history, and it isn't inevitable for most of the species that do or have existed. Life has no destiny to evolve into sapient organisms; sapience is only one adaptation for survival out of many, it worked for us but we're clearly in the minority.
What about all the other conditions that need to be fullfilled? Presence of a magnetic field to deflect harmful radiation? Presence of an oversized moon to stabilise the planet's spin? What if these are extremely unlikely conditions? Wouldn't that affect your calculations by a random number of magnitudes?
If your atmosphere is stable without a magnetic field, you don't really need a magnetic field. You get more radiation on the surface but life should be able to cope with only the shielding provided by the atmosphere, after all, the magnetosphere has weakened many times in the Earth's history, in periods during which the polarity of the magnetic field has flipped. Nevertheless the presence of a magnetic field would be quite likely if the planet in question had a fluid core and a relatively fast rotation rate (traits that are also more-or-less required for a planet to harbour complex life).
As for a large moon stabilising rotation, is this really needed? Do all planets spin off axis if they are moonless? I know Mars goes through cycles which give it very high axial tilt for part of the cycle, but I thought this was (in part, at least) caused by the asymmetry of the gigantic Tharsis highlands.
Does Venus experience such severe wobbling?
EDIT:
you have the whole historic evolutionary junk in it, that reminds you that your ancestors had been once dinosaurs.
Urwumpe, your knowledge is almost always beyond my own, but unfortunately I have to say that here
You Failed Biology Forever.
Human ancestors were never close to those of dinosaurs probably since the first amniotes, a better example might be a Cynodont, a group of advanced therapsids (which are themselves group of synapsids, which includes mammals and their mammal-like (extinct) relatives):
Synapsids have a single pair of temporal fenestrae (holes behind the orbit that are used to anchor the jaw muscles), whereas dinosaurs (and all living sauropsids (reptiles)- save for chelonians) are diapsids; i.e. they have two pairs of temporal fenestrae.
Turtles and tortoises are anapsids, i.e., they have no temporal fenestrae. A substantial group of extinct reptiles are placed into this group, and primitive amniotes (i.e. the ancestors of synapsids and diapsids) also lacked temporal fenestrae. However it has been theorised that chelonians are diapsids which have lost their temporal fenestrae over time, and are only distantly related to the anapsid reptiles of times past.
