@Sovereign
Arno's post is mainly dealing with this bug. The side issue he mentions isn't a bug, just bad practice. He seems to have made a significant effort, looking into the bug, going far beyond normal reporting. If they haven't already made a good start on it, the developers may find it useful.
@arno the fabulous
Great job, going through all of that! I have long suspected that Inno neglects proper code development, in favor of rushing out money-making schemes and pretty buildings (which are often the same thing). As you point out, when neglect is allowed to continue, it'll eventually make the code nearly impossible to untangle (I'm sure some refactoring was sensible enough, though not helpful to us). I can only hope that they're planning their move to HTML5 as a complete redesign, eliminating all sloppy, makeshift, hasty, and otherwise poor code. That might explain their weak efforts with the Flash code, and give hope of sturdier code at a future time. Even so, if your assessment is correct, which I'll just assume that it is, partly because I'm to lazy to do anything else, it's still not good.
It also still needs fixing. If they can't find any other way of doing it, they need to simply roll back to v1.121. They certainly need to recognize that it needs proper attention from someone with the proper skill and authority to make real changes. In other words, not a job for the intern.
@Lord B
It's hard to believe that it should be so hard, isn't it?
It was broken in v1.122. If that version also introduced new interdependencies in the relevant code, that would make a simple roll-back or inspect-and-fix difficult, and might explain the trouble. Interdependency is also why, the more updates they release
without fixing this, the harder it will be to fix it by simply rolling back or changing selected pieces of code. Which makes me nervous. If there weren't any changes in dependency, the difficulty in fixing this becomes a bit of a mystery.