At the time of writing the review, I hadn't completed the game and thus didn't give it a score. The Dtoid tips line got an angy email from some guy who hilariously ended with "email me back if you want my full review," basically saying that I sucked as a gamer and reviewer because I couldn't beat it.
I'm not going to argue that specific point so much as I'm going to say that I have since beaten JUMP!, and consider it just as bad as I did the first go-around.
The problem with JUMP! isn't that it's overly hard -- it's that it assumes being "retro" when it comes to difficulty is a reasonable substitute for legitimately interesting level design. You'll die not because the game is throwing a bunch of really interesting, really difficult shit at you (a la, let's say, N+), but because the platforming challenges are so not-demanding that you'll more or less just sleepwalk through most of the boring parts, and occasionally die just because you weren't paying all that much attention. I think I addressed this in the original review through constantly blabbing about "uninspired platforming" whenever possible.
When I finally hunkered down and tried to keep my attention as fixed on the game as possible, I beat it with a half-dozen lives left over; I also didn't feel any better than I had when I lost all my lives on the second-to-last level. Yeah, I'd put enough effort into the game to beat it in one sitting, but I wasn't really given anything in return.
The game wants it both ways: it wants you to get that retro feeling of accomplishment and triumphing over punishing-but-fair level design, but it also wants to make the game easy (and therefore, in many cases, boring) enough so that an average person can make it all the way through.