I have found that controllers can be overridden. Twice so far in my career I have successfully argued the toss and got my own way, although these were for issues that did not directly affect other train movements. I have also opted not to carry out an instruction immediately which avoided the need to remove a train from service. On each occasion I felt confident that the action I was taking was correct and defensible.
All that said, however, you really are sticking your neck out and taking the responsibility squarely on your own shoulders. If it goes wrong it may not be just an uncomfortable chat with the gaffer that results. There is also a risk that you might just be mucking up whatever contingency control is trying to put into place. It's an issue for sure, and maybe frontline staff should have more input into the situation. But then I think I'd prefer it if the controller was busy getting on with getting something sorted out than having to consult anyone and everyone first.
On the information point, again it's a balancing act. I have found through experience that passengers are more accepting of a situation if you keep them in the loop (as it were), even if you have nothing to tell them. Unbroken silence from the pointy end just makes them fidgety. Yes I agree that we should be told more, but sometimes there really isn't anything to tell them because the one piece of information that passengers really want to know ("How long is this going to take?") cannot always be accurately ascertained. If the problem requires, say, a man in a van to come out and fix some lineside equipment, would you rather he was busy with the repair or dragging the job out for longer due to the need to call-in every 5 minutes?
I agree with my colleagues that something needs to be done to improve the handling of incidents, and I hope that there will be improvements made. However, I would also like to highlight what they have said about how these incidents are not going to go away. There are no "magic bullets", just small incremental changes that can be made to procedures. Therefore the wishes of the travelling public will need to be tempered with a degree of realism.
On the GSM-R battery point, I am a little unclear about this. Is this an issue with all GSM-R sets or is it a consequence of the Electrostar load-shedding process? My memory of driving these units is fading, but I seem to recall that they have a single handset for the PA, PCA and GSM-R with mode selector buttons, as the Turbostars do. This being the case, when the load-shedding finally does for the PA it would also remove the PCA and GSM-R facilities. Surely it must be possible to put these units into a low-battery survival mode whereby the PA, PCA and GSM-R continue to operate, together with an auxiliary compressor to raise the pan and close the VCB, all without the TCMS or any other systems operating. Compressor aside, these are low-draw systems.