Am 03.02.2016 um 09:29 schrieb Ola Fosheim Grøstad:
On Wednesday, 3 February 2016 at 08:23:39 UTC, Sönke Ludwig wrote:
Maybe it's showing different results to you, but the numbers I get are
tiny. Also, picking arbitrary search terms skews the results
considerably.
Not really, you look at trends over time not absolutes.
Okay, I see, it's just percentage of the highest value. To make any kind
of qualitative judgements, it would be necessary to at least have a hint
for the absolute numbers.
Results using Google's categorization instead:
I have found the Google classification to be wrong in the past.
I have no doubt about that!
So I think it is better to find a term that is unique for the language,
I don't know any such term for Rust though.
And how do you verify that that's a better classification? If you look
at the graphs of any of "D", "dlang", "D programming language", "D
language", "D programming", none of them seems to correlate with events
such as the date of first publication, version 1.000, version 2.000, the
conferences etc.
