As expected, the "mystery bird" is actually pretty common and easily identified once I actually SAW the bird doing the singing -- I described the song as a series of half a dozen high-pitched notes rising on a chromatic scale with a downward glissando at the end, as I recall, and it turns out to be golden-crowned kinglets. This also fits in with the "I only sing from truly annoyingly high tops of trees" bit.

The Oregon recording at the Cornell site is a good approximation of the first part; the bit at the tail end seems to be rather different, but I put that down to geographic variation. The high-pitched LOUD whistle I heard a week or so ago and could not identify was not heard again after that day, which makes me think it was probably an escapee from someone's house, sadly. I fear that puzzle will not get solved, but it's good to have the common bird ID-ed!

Cheers,

Josh in Licton Springs (North Seattle)