Halting Problem == Russell’s Paradox?

Is a (set-theory based) program which predicts its own halting outcome (and then does the opposite) able to predict its own halting outcome correctly?

Well that was short and quite interesting, I’m looking forward to their follow up posts.