What Makes A QSO? Is Recording OK?

I hit this question while thinking about my mobile HF operating these days. I put an FT-857D in my truck so I could run HF any time I wanted on the go - as opposed to VHF/UHF only. Before I got the rig I used my TS-2000 on trips and when I really wanted to do mobile HF, and before the TS-2000 I used my IC-718.

Recently I have started using a voice recorder for logging, rather than writing notes. It is a much more efficient and safe method while on the road. Not only that - I can record the QSO, which is what I do, so I don't have to remember anything. That boosts the safety that much more - I don't have to concentrate as hard. Once I get the call sign of the other station and their name the rest is in the recorder.

The biggest place this helps out is CW - and working weak CW at that. This could even be applied to a home station. If you are trying to pick out a weak station down in the noise you may only pick out bits and pieces of the exchange, but not the whole thing. Usually we just ask for repeats. However, if the QSO is recorded it can be played back as many times as needed to get the information.

This leads me to my question.

Do you think recorded QSO's count if the purpose of the recording is so the operator does not have to copy in their head the message or information and transcribe that to another form immediately as the signal is being received?

I think that since the signal was indeed there, and was recorded, the QSO happened. However, I think that the operator should use his/her best judgement as to whether or not the signal being received is intelligible though. If the signal is truly below the noise floor and in normal circumstances would not be detectable I don't think the QSO should count - which I believe any honest ham would know.

Recorded QSO information, to me, falls in the same category as the WSJT weak signal modes do. I will use JT6M on EME for example. The stations in QSO alternate sequences. Only when one station starts it's transmitting sequence does that station's computer switch to data analysis and eventually display of the received signal. The signal is not, and can not, be decoded as it is being received. The signal MUST be recorded and then computer analyzed. There are no two ways about it.

However, running voice modes (SSB, FM, AM, DSB) and CW (Morse code) the signal is entirely capable of being, and has been since before computers, decoded as the signal is being received. I think there may be some bias towards these modes staying as as-received modes, in other words if information is not copied by the operator at the time it was sent the operator must ask for a repeat.

I guess the determining factor is that the information must get through for it to count. If an operator does record, lets say a weak CW QSO, and confirms receipt of the exchange thereby completing the "legal" QSO requirements but in reality the exchange, even upon numerous playbacks of the recording, is still not discernible the QSO can not count. The operator would have to have the other station repeat the exchange and fill in the missing information, otherwise the previous "QSO" would be void.

Wow. Maybe I am thinking about this one too hard!