The point being that while I'm spotting three previous uses of the same ticket (sometimes on the same train, they have been known to try sharing the ticket between themselves) there are certainly more previous uses that didn't get recorded because the passenger gets on an Avanti service (on the odd occasion that it actually runs) or whatever.
I'm presented with a time, date, and service number (or station name for barrier scans) of previous scans. It isn't difficult to work out from this information where precisely they were when the ticket was previously scanned. Pretty much all of the fares I'm speaking of have a BoJ restriction anyway.
Right, but I think we're slightly at crossed purposes. I wasn't saying there's no significant problem of people trying to re-use open return tickets.
I was doubting that there's a significant problem of people trying to re-use open return tickets
that would be fixed if the OP's suggestion were implemented.
A lot of what you're describing would mean, unless I'm very much mistaken, that the used ticket that the person holds when you check it is basically useless, no? Okay, they might have got away with it if, for some reason, you didn't get around to checking their ticket, but that would equally be true if a passenger has no ticket at all. And a requirement to validate before boarding presumably wouldn't much affect people who are already breaking the rules fairly flagrantly.
As I understood the original concern, it was that the return portion of a ticket could be wholly used without ever being scanned, in which case the passenger would know this and may decide to use it again. In which case, if you scanned it when they were using it for a second time, you wouldn't know it had been used before as there would be no previous scans. We can only speculate how common or uncommon this may be, since if it's happening, it is by (almost by definition) not being noticed either by staff or by the system. The OP seemed to think it might be happening a lot, but I thought it seemed unlikely.
People presenting clearly used tickets for inspection isn't really the issue, as far as I can tell, as that's already detectable. Barcode tickets are actually better in this regard as previous use of the ticket can be inferred from scans at ticket barriers, whereas the use of CCSTs at barriers isn't logged, so there is great reliance on ticket checkers marking the ticket (which is actually quite hit and miss in my experience - but maybe I've got an honest looking face).