It's difficult, but not futile at all IMO. If you were picking a side to win test matches today, there would be few players from much more than ten years ago. However, if you pick players based on how they fared against their peers it's possible to come up with a side that's representative of all of rugby if you know your history well enough. Take Lomu for example, some people argue that he wouldn't be as effective in today's game in which players are more used to coping with giant outsides. Whether it is true or not is irrelevant to this discussion, the point is that you need to assess how he got on against the opposition he faced. Arguably he was an irresistible force (unless you're South African).
It's much tougher in rugby than other sports as there are few, if any effective statistical measures to assess players like many other sports. Don Bradman is a classic example, get a time machine and transport him to today and there's every chance he would struggle against current bowling attacks, but he is statistically head and shoulders above anyone before or after him.
It's funny that you mention boxing. It's not a sport that I know much about, but it seems to be one that's obsessed with this type of comparison despite the additional complication of weight categories. In the context of the original question in this thread, the assessment of Rocky Marciano is relevant, despite his unbeaten record in the blue ribbon division, he isn't rated that highly because of the poor competition he faced.