Logically, if Firefox is not going to support long term stability and compatibility -- and clearly it doesn't in the case we're discussing -- then the only possible conclusion is that Firefox can't be part of an effectively managed IT infrastructure for these kinds of organisations. That means the correct course of action for those responsible for that infrastructure is to plan to remove any dependencies on Firefox as quickly as possible and to replace it with something more stable, which presumably means IE in this context.
You need to evaluate the ESR release, and/or find a different browser entirely. IE may indeed be the best option if you only support Windows clients. Though Microsoft hasn't exactly been shy about forcing IE along more quickly of late, either.
I'm not sure who mainline Firefox is intended for any more. That's part of the problem, I think.
It seems like Mozilla are chasing Google to the exclusion of almost anything else, and the main goal for both of them seems to be ticking boxes to say they have more bleeding edge features, even though hardly any real projects can actually use most of those features because they aren't stable and portable enough yet. Meanwhile, users get interfaces that subtly shift around every few days, developers are fighting a constant battle just to stand still, and as we've been discussing, organisations can't manage large-scale deployments robustly at all any more.
If you had a Java application, you would not have a problem. If you had a web application, you would not have a problem. If you had a C or C++ application built against either native Windows APIs or a mature cross-platform toolkit, you would not have a problem.
Instead, you rely on browsers continuing to put their users at risk by automatically running code in a constantly-leaking sandbox managed by a company that doesn't give a shit.
Understand this: Java is a hole in the defenses of modern web browsers. It is behind. Way behind. And it will remain so forever. It and its owner are not up to the task of dealing with the modern web.
So yes, you're going to be fighting a constant battle so long as you insist on relying on what everybody already knew was crap in the mid-90s.
Or you could just tell your users to click the magic button.
I work with the NHS in the UK; quite a few of the medical professionals are forced to use IE6 or IE7 because the hospital IT staff are managing their medical devices / desktops / network effectively, just as you say.
When Firefox makes a decision like this, what they apparently did not seem to consider is that they are drawing a cutoff line which has serious costs to some of their users. I didn't see any discussion of this whatsoever in the issue thread.
Imagine a relatively forward-thinking hospital that has been able to allow their staff to use fairly-recent versions of Firefox (once new versions have been vetted by IT). Changes like this may force them to stop deploying new versions completely, until some time possibly a decade from now when they're finally able to replace the Java-based UI.
Would that be a failure of Firefox (or other browser vendors) or a failure of hospital IT staff to manage the medical devices / desktops / network effectively?