Fyrd / caniuse

Raw browser/feature support data from caniuse.com
https://caniuse.com
Creative Commons Attribution 4.0 International
5.66k stars 1.38k forks source link

Big changes in browser usage stats in regional stats #6819

Open sashachabin opened 1 year ago

sashachabin commented 1 year ago

Hi!

My same Issue exactly one year ago: https://github.com/Fyrd/caniuse/issues/6426

This affects the browserslist website and browserslist-lint.

Demo

1. Sum of usage in regions > 100%

I created demo with version 1.0.30001525 from CDN. Large values ​​are shown in bold https://codepen.io/nibach/pen/KKbgZxx

image

2. Big changes in China stats

Regional changes from version to version in CN.json https://docs.google.com/spreadsheets/d/13XUk2aMpE9Qe_SWng2-FwZqT9_cPOrhWsikrP9TSzWA/edit#gid=0

sashachabin commented 1 year ago

And I found big changes in IE 11 usage statistics in China 1.9% → 6.4%

caniuse-db@1.0.30001523

caniuse-db@1.0.30001525

sashachabin commented 1 year ago

I did some more research. And I found big changes in regional stats in China (CN.json region).

For example, IE 11 stats in CN:

I create Google Spreadsheet with changes over the last months: https://docs.google.com/spreadsheets/d/13XUk2aMpE9Qe_SWng2-FwZqT9_cPOrhWsikrP9TSzWA/edit#gid=0

IE 11 stats in China
Fyrd commented 1 year ago

Thanks for pointing these out, indeed a number of regions this month had bad data due to being processed before the previous month's statcounter data was ready. Have just updated these regions and the numbers for August should be correct now.

sashachabin commented 1 year ago

Thank you!

I checked new caniuse-db 1.0.30001532 version. This problem with value greater than 100% is fixed!

sashachabin commented 1 year ago

I compared statcounter data in China in August and latest CN.json in caniuse-db.

statcounter chart

CN.json in caniuse-db image

Some browsers with stats differences

statcounter caniuse
Chrome For Android (and_chr 0) 37.94% 51.5564%
Edge 115 4.93% 1.7372%
Chrome 115 1.37% 0.4816%
Chrome 98 1.03% 0.35432%
sashachabin commented 1 year ago

@Fyrd Why do you think there might be such a difference in regional stats?