I tend to agree with Mike Easley on this one.

In the old days, we navigated with the directional gyro (DG) and used the compass only to set the DG.  The compass was also our backup device in case the vacuum system, or the DG itself, failed.  Compasses essentially never fail -- the earth's magnetic field always works and the only other way to fail the compass is to smash it.

Nowadays, we have EFIS and handheld GPS with dual-redundant electrical systems.  Even if the airplane gets hit by lightning, destroying both electrical systems, the battery powered GPS will probably still work (umm, I hope!).  So the likelihood of needing to use the whiskey compass for navigation is much, much less.

I doubt that the EFIS / handheld GPS "modern" combo gives the statistical reliability of a vacuum DG plus a whiskey compass, but I suspect that it's plenty good enough.  I would consider it extremely unlikely that the "modern" airplane would ever need to be dependent solely on a whiskey compass for heading information.  Therefore, putting it in an out-of-the-way place is probably fine.

By the way, when I was learning IFR in my Cessna 150, I had a fairly old DG that precessed about 10 degrees on each approach.  It became second nature to reset the DG at the conclusion of each missed approach.  I had the compass up on the glareshield and cross-checked the DG every 10 minutes, and that's where it is on my not-yet-finished 360.  (Oh, and the attitude indicator was just as bad, so I bought a new one.)  Good thing I never went into hard IFR in it -- just the occasional marine layer for a few minutes at a time. 

The message here is that if you want to cross-check the DG with the compass in flight, you'd better put the compass in a good spot.  If you're going to be using your electronic systems for a DG reference, hell, use a compass out of a Cracker Jack box and duct tape it to the window and I won't care.  (Well, I will, actually.  I'm just exaggerating to make a point.)

- Rob Wolf