I was fascinated by popzeus' rankings table
. I notice the total number of published transcriptions he gives is now just over 3 million, and had some time to spare, so for fun have summarized his table for 3 Oct in a smaller table available here
. If you don't like rankings, stats, competitiveness, or (even small) tables of numbers, then please give this a miss.
So far 3,012,389 weather reports have been transcribed for 673,651 pages, giving an overall average of 4 reports per page.
The summary table shows the total reports transcribed, and a summary the number of reports transcribed by various ranks of the top 500 (i.e. those listed on popzeus' table):
the bottom 10%, 25%, 50% (i.e. the median)
, 75% and 90%, as well as the mean (the notional number of reports each of the top 500 have transcribed if you distribute all transcribed reports equally amongst them)
It also makes some deductions about the 232,425 reports transcribed by those outside the 'top 500', based on simple assumptions:
- a guess at the number of reports transcribed by the transcribers outside the top 500* [assumption]
- a guess at the number of pages transcribed by the transcribers outside the top 500 [assuming the overall average of 4 reports per page]Health warning
Remember this is all a bit of fun - in effect, a 'back of a fag packet' calculation!
The above [assumptions] are likely to be extremely generous; one-off transcribers are very unlikely to have transcribed as many reports as those in the top 500!
For definitions & explanations of how the published rankings are made, see popzeus' replies nos. 24
& 125PS thanks, popzeus, for introducing me to datascraping in general, and OutWit Hub in particular! (so far I'm not fluent in macros, so I'm afraid this a one-off effort!)PPS this is a first stab - I'm sure someone will let me know of any goofs!!