Maybe it's just me but I can't see much difference between the posted examples on that page. In fact, even though I am conscious of OOF areas in my images, I really can't see anything objectionable unless a lens starts forming double circles or hexagons in the OOF highlights. As long as it's smooth I am fine. Maybe I am not quite a bokeh connoisseur yet.
I find the tests on that page to be absolutely irrelevant because they all involve different light situations, subjects, distance to camera, distance of background to subject, focal length, aperture, film formats, film stock...need I add more problems?
Is a better test series.
They're both 'useful' (the latter one less so, since it deals primarily in lenses that aren't widely used) to some degree. I find informal real-world results/tests, even with a variety of subjects and film stocks, etc. to be more useful to me - they're more representative of what I might conceivably shoot. I guess I'd say my interest in reading lens info has more to do with finding out what kind of imaging characteristics I can expect in average usage, rather than an objective finding.
(Sadly, neither test focuses on lenses I might conceivably buy in the next six months - 50mm Summicron v. ZM Planar vs. Nokton...)
Oh I wasn't meaning direct usefulness, but rather rigour of testing. To someone who hasn't built a huge storehouse of bokehs, the first page is meaningless because it does not afford the reader to identify precisely what are the characteristics of each lens. The other one (BTW, a link from the first one!) gives you a more useful experimental procedure, results, and conclusions. If you won't use the lens tested, you can at least learn more about bokeh and how to test for it.