dmd downloads

Sat Jun 27 04:27:03 UTC 2020

On 6/26/20 11:28 PM, Adam D. Ruppe wrote:
> On Saturday, 27 June 2020 at 03:24:42 UTC, Andrei Alexandrescu wrote:
>> FYI I just got around to generate again the plot for the moving 
>> average of daily downloads and the trend is pretty nice:
> 
> Does this filter out repeated IP addresses, or use any other method to 
> try to exclude duplicates?

No filtering except for Travis CI. I'm not sure how IP sharing (common 
in Asia, Africa) influences unicity. A cursory look at the past few days 
suggests a 10:1 ratio of total IPs to unique IPs. So the 5000 downloads 
shown per day would be from 500 unique IPs, which is still good. The 
trend is more interesting to watch than the absolute value.