1. Nobody tested this until just now, meaning our understanding of the Casimir effect was sufficiently incomplete that nobody should have been writing on the topic with any confidence or authority. A real scientist familiar with the topic probably wouldn't have been; but "real scientists" are sufficiently thin on the ground that you could likely have gone through a doctorate in science and not met one. Teachers in particular, at all levels, seem pretty prone to talking and acting like they're hot shit.
2. Nobody thought to test this until just now, which means that some pretty dumb assumptions were made (they're dumb because they were assumptions and incorrect).
3. Nobody thought to test this until just now, and it's a pretty _obvious_ test too, so either nobody spent any time on it or they were extremely myopic. (Something I've seen in many "scientists"' publications these days; overspecialization to the point of virtual uselessness. They're competent to gather data but not design interesting tests.) I'm only vaguely familiar with our knowledge of the Casimir effect (which is sometimes good!), and I would certainly test all sorts of patterns - on each surface - to figure out how that affects the effect.
Of course this discussion is based on the assumption that what the summary talks about is in the article, which I haven't checked, and that the article faithfully summarizes what's in the paper, which I haven't checked, and that the paper purports that this is new knowledge, which I haven't checked. It's quite possible that what's published in the paper is already well-known.