I've got about 500 hours in both SR20s and SR22s, so I'll throw out
some real world experience (not that it's worth anything in a
newsgroup, but here goes.)

The folks claiming that they stall without warning and are then
difficult to control have obviously never flown one. You get plenty
of warning (as required in Part 23) by buffeting, mushy controls, and
the stall warning. If you keep going, the inner part of the wing
stalls first (it is at a higher angle of attack than the outer part;
take a look at the wing cuffs.) When that happens you still have
aileron control if you're sloppy and try to use it. It feels like a
washboard road. If you are sufficiently uncoordinated you can drop a
wing, and if you try hard enough you can spin one, but you have to be
asleep at the yoke to do this.

They are not particularly difficult to slow down because of the fixed
gear. You can deploy half flaps at a relatively generous 120 KIAS.
Much easier to deal with than the Baron I fly.

I'm not sure where the "too small flaps" claim comes from; they seem
to work just fine, and making them bigger would reduce the speed at which
they can be deployed, and make the power-off descent angle with full
flaps even more impressive than it is now.

Handling is responsive and predictable. They're way fun to yank and
bank. The side yoke does not require much force and comes naturally
very quickly.

The trim is a bit of a pain (it runs too fast) but you get used to it
pretty quickly.

The paranoia about spins and spin recovery seems overblown. The test
pilots have spun them and recovered, and nowhere in the POH does it
say that spins are "unrecoverable" (contrary to another posting here.)
Until recently the POH suggested normal spin recovery technique,
though this was removed in a recent revision, presumably as a CYA
move. I have heard, though cannot confirm, that some kind of spin
certification will be required for JAA certification, so hopefully
that will put this issue to rest.

Having said that, part 23 only requires recovery from a one turn spin,
which isn't really a spin at all. The drama about the "death zone"
below 900' is seriously overblown; I suspect that the majority of
pilots have never had spin training, and even those that have are
unlikely to recover from the usual base-to-final spin in *any*
aircraft.

The 4000-ish hour life limit on the SR22 is a certification artifact;
Cirrus chose to use a very conservative formula based on the SR20
airframe life tests in order to speed the SR22 certification process,
but will be extending the life based on testing.

I think you'll find the *vast* majority of Cirrus owners are very
happy with their purchases (with the exception of ArtP, who seems
to have gotten a lemon.) I've never had a maintenance problem that
cancelled (or ended) a flight, and the only failures I've had have
been with OEM parts, and this is true of most owners.

The only design characteristic that I think can cause handling
problems for low-time transitioning pilots is the high wing loading,
which requires higher takeoff and landing speeds (rotate at 70 KIAS,
final approach at 70-75 in an SR20 or 75-80 in an SR22) and causes
serious sink rates if you get too slow. You have to land them like
heavy airplanes--hold the approach attitude all the way into the
flare, and no 50 AGL roundouts like in 172s. A number of the landing
accidents were due to this (IMHO), coupled with insufficient training
(which I understand has been fixed, though it's been three years since
I last had the factory training.) Early on the trainers were
recommending coming in five knots faster than the POH numbers in order
to accommodate sloppy 172 technique, which results in excessive float,
and bouncing, and things go badly after that if you don't immediately
go around. If you have the discipline to fly the numbers, they are
pussycats to land, and have more than enough energy to flare and land
smoothly even with the power at idle.

As far as the accidents go, simply pointing at statistics and calling
the plane a death trap and saying that they are "falling out of the
sky" isn't supportable by the facts. Of the eight fatal accidents
(not counting the flight test accident) five (and possibly a sixth,
though there isn't much data on the crash in Spain) were CFIT. Hard
to blame these on the plane per se.

Ultimately it comes down to whether people do more stupid things in
Cirrus aircraft than in other brands. Statistically it's too early to
tell, and the time-in-type average is very low. Basically, you can
cook the numbers to support your position, regardless. I think it's
probably true that someone who is going to be stupid enough to scud
run at night or in mountainous terrain is probably more likely to die
in a Cirrus than a Cessna because of the speed. It may well be that
pilots feel safer in a Cirrus than in a 25 year old 172 (I know I do,
and it's arguably true, particularly IFR) and perhaps that leads the
marginal ones to take bigger risks. But there is no shortage of
pilots doing dumb things in all manner of aircraft, and dying on a
regular basis. Time will tell.