Current mood:	blah
Current music:	The Flowers of Bermuda - Stan Rogers

blrg* and FKO Bits & Pieces
Came down with a nasty cold a day or two after FKO. Yesterday (Wednesday) I went to the dentist for the semiannual torture known as teeth cleaning. I'd warned the people there that I had a code, and they took appropriate precautions: they said that there has been a cold running about in the last week or so. Also, that the incubation period is between 7-10 days ... so I cannot blame my brother (or anyone else) for infecting me at FKO. Most likely the infection came from family members during Easter holiday or the birthday party for my brother, Bob.

---

At FKO oreouk had a taste test on various cookies (three packages were Oreo** cookies sold in the UK, Canada and the US; and one was a 'ringer' to see if the testers were paying attention to the logograph on the cookies. oreouk carefully kept records. The test was to see whether the results invalidated the null hypothesis H₀, that is, people at FKO could not tell which Oreo cookie was sold in which country any better than random chance.

Ultimately the total number of responses, n_t = 32. Of these, two responses mistook the 'ringer' cookie for a legitimate Oreo cookie. Although a deeper analysis would probably be required, we discarded the two entries of those who did not pay attention, making n=30. Of the 30 responses, 4 responders got the assignment exactly correct.

Where k items may be placed in k categories (be they letters in envelopes, or Oreo cookies with countries), the total number of ways is k! (k-factorial). In our test, k=3, and so the number of possible ways to assign the 3 cookie-types into 3 country-categories is p(k) = 3 x 2 x 1 = 6. Simple probablity theory shows that the mean correct responses (if the responses are made purely by chance) is n/p(k) = 30/6 = 5.

Seeing that the actual number of correct responses was 4, I do not believe that the null hypothesis, H₀, can be rejected with any degree of confidence. Until a larger dataset is obtained, I remain suspicious of any claim that Oreo cookies can be assigned to nationality-categories. Rather, I believe that the variation among Oreo cookie batches amongst manufacturing plant(s) in a single country is as great as the variation between plants in the various countries.

---

Finally, at FKO Heather Dale mentioned that her new CD The Road to Santiago was coming out this Spring. Indeed, she previewed the title song at FKO. I tell you this because I played Stan Rogers' Between the Breaks ... Live! on the way back home, for my brother, peteralway. Pete then opined that Dave Clement performed Stan Rogers tunes better than Stan Rogers. That is certainly a debatable assertion. But one of the songs on Between the Breaks ... Live! is "The Flowers of Bermuda" which — with its quirky changes in meter — indeed adds to its level of difficulty. I rarely hear someone attempt it in the filkroom, although I personally think the tale is as uplifting as that of the "Mary Ellen Carter."

So. After returning home, I look for the new Dale release on her website. And, to my surprise, it says that Heather has recorded "The Flowers of Bermuda" on it, which, she says, has a "crazy melody." Unfortunately, there's is no MP3 sampler, so I shall simply have to buy the CD and find out how well she did on this non-trivial Stan Rogers piece.

---

* "blrg" is a trademark of pocketnaomi — used by permission.
** "Oreo" is a registered trademark of Nabisco.