Others involved in amateur radio and especially LF might like to express
an opinion about what constitutes a valid qso.

Since the purpose of the QSO is to demonstrate that a communications link
was established between a pair of stations, isn't a QSO defined as the
unambiguous bilateral exchange of callsign and a single piece of unique
information. The important property of the unique information is that it
must not be exchanged on the out-of-band link (ie Internet, HF calling
frequency etc). Normally the unique information is the signal report
in the appropriate format for the mode.

I do not think that any time limit is appropriate, nor do I think that it
needs
to be exchanged in one go. In normal operating (for example VHF contests)
it is quite common to piece together the call over a number of overs, and
to correct locator information from monitoring the other station later
on when there is less QSB. Thus in current VHF practise a contact may
be scored which in effect took 24 hrs. What is important is that the unique
information is unambiguously exchanged on the band in question. I do
not think that it is important that the unique info is a signal report, in
fact
it would be a much better test if it were a set of  random numbers.

73

Stewart G3YSX