It may or may not die within few years (I'm placing my bets on the optimistic side), if it delivers value today, and the alternative is a cost prohibitive walle

It may or may not die within few years (I'm placing my bets on the optimistic side), if it delivers value today, and the alternative is a cost prohibitive walled garden unsuitable for sensitive data, then it's well worth it cost in public money already.