We do get pager messages about ticket offices closing out of course (e.g. Castle Cary ticket office closed, please allow passengers to purchase tickets on-train), and what TVM▸ 's are not working. Unfortunately, when it comes to the barriers, I suspect that the staff get very cynical very quickly!
Yes, appreciated. On an insignificantly small sample of my own experiences (two!), this has happened to some degree. They knew the Hungerford machine was out, but not that the train manager's was out also ... so it resulted in quite a long queue at a single "naughty boy" counter, with a chap and then a chapess in front arguing for what seemed like an eternity on a totally different topics (they probably
hadn't followed all the rules, but whether by accident or design I don't know). It wasn't an eternity, of course ... train arrived on time, and just over half an hour later I got through the barriers.
As regards point 3, it's an imperfect world
Indeed, it is ... and in an imperfect world, I mightily resent being put through hoops to stick to the rules because of failings of the system that should be in place. That applies not only to point three.
Personally, I've lost count of the number of passengers from Pad to Reading (and points west) when the train went off an unbarriered platform that "didn't have time to buy a ticket" (while eating a McD), or even more blatently just say "Oh well, you have to try it on". My favourite are the muppets who ask for "A ticket from the last station", and when you say "What name was it?" don't know and are unable to say how they got through the barriers ("The man let me through because I was rushing" - When going Bridgend to Cardiff? I don't think so!)
Yes, you have a conundrum there. Yet I have learned over the years that even the most implausible-sounding reasons can
on occasions turn out to be right. I'll have to tell you sometime about the computer terminal that sometimes went into a loop of misbehaving and jerked back to life with a good thump on the side. And it turned out to be a software problem, fixed by altering code on the mainframe to which it was attached ....
Flamingo, Thanks for your answer. I had really hoped to get back with you much quicker on this one, but I had the busiest of weekends!