Altimeter Readbacks

I'm afraid I'm going to demonstrate two areas of where I lack knowledge in
today's post. Both are based in one of Paul Tomblin's replies to my post
from yesterday.

The first one relates to my limited knowledge of blogging. I don't know how
to keep an "Ask the Controller" thread easily accessible from the top of
the page, which would be more likely to be of use. It's probably something
very easy to do, and I just haven't explored it. Suggestions?

The next area is the subject of why some controllers seem to want a
readback of an altimeter setting. I can say that we, in Canada, have a
requirement to issue the altimeter setting on initial contact and as
appropriate while the aircraft transits different areas. I have never seen
a written requirement for a pilot to readback an altimeter, nor have I seen
a written requirement for a controller to obtain a readback of one. IFR
clearances, yes. "Hold short" instructions, yes. Altimeters, no. As a
controller and a pilot, I appreciate the value in having an opportunity to
catch an errant receipt of an altimeter setting. If I hear an errant
readback, then I most certainly have a responsibility to correct the pilot,
and for safety's sake (not my own neck), I will act on that. So if a pilot
reads it back, I'll listen for it and confirm that they received it
correctly. I have caught many errors over the years, even a few that were
horribly outrageous. Any chance to catch a problem is welcome. But if a
pilot doesn't read it back, I won't fish for it. I have no responsibility
to do that.

Having said all of that, I can't offer a reason as to why some controllers
insist on a readback of an altimeter setting, except for maybe personal
experience. Perhaps a pilot set one wrong and if the controller had chased it down, a
crash or incident may not have occurred. This is a normal thing to have
happen in our business: The incident occurs and suddenly, someone
investigating pulls blame out of thin air and assigns it to someone based on
nothing other than a supposition, perhaps one that couldn't even be proven.