With regard to your first paragraph: CBTC is sufficiently far into the future than any reasonable discount rate will still make it sensible to build a standalone countdown clock system that disintegrates when CBTC is accomplished. Put another way: it is easily worth the cost of countdown clock system just to cover the finite span (~20 years) it will take to get CBTC. I know you're playing devil's advocate, but I don't think the MTA can honestly endorse the argument.