I recently recovered a data-set my father (a biology professor) recorded at UC Berkeley in the 90s on Stickleback mating behaviors. He says the data was recorded using the "open-source BEAST event recorder software, on DOS, on a Compact computer." The files have names like R721F211.CPR
. According to file
, they are "ASCII text, with CRLF line terminators", and the contents look like (for example) this:
"right male, tank 2, 1st trial, 6-4-93, 255-325, precourtship"
TIME= 10
30.00 20
2084 -2089 2099 -2117 3905 -3908 3932 -3933 6890 -6904 7945
-7947 11177 -11178 13193 -13197 15580 -15581 15664 -15665 0 0
43.00 22
1459 -1460 3561 -3562 4651 -4652 5661 -5662 5834 -5835 6779
-6781 9162 -9163 9169 -9170 10795 -10796 11892 -11893 13052 -13054
What is the grammar for this file?
My dad and I looked at this and deduced the following:
- the first line is a note describing the contents of the file, which my dad must have typed out;
- we are not sure what the
TIME= 10
means - we think the "20" in
30.00 20
means "event with ID 20" but we are not sure what the30.00
means; likewise we think the 22 later means event "22" but we are not sure what the 43.00 means; - we think the items like
3932 -3933
are time intervals, as in, event with id 20 occurred from second 3,932 to second 3,933.
Dad says he recalls the software was open-source, accessible via an academic precursor to the modern Internet, ran on DOS, and was developed by a professor in Hawaii. This is all we know.
Either the grammar for these files, or software with which to interpret these files, constitutes an answer to this question.