🧵 Wikipedia.org and sister projects, #browserstats for May 2021:
* 49%: Chrome +Mobile
* 24.7%: Safari +Mobile
* 5.2%: Firefox +Mobile
* 2.8%: Edge
* 2.5%: Samsung Internet
* 2.0%: Chrome iOS
* 1.6%: Google app
* 0.79%: Opera
* 0.72%: IE
* 0.26%: Yandex app
* 0.22%: DuckDuckGo app
100% = 16.4 billion views (desktop+mobile web, no apps, no bots)
Thread special: Browser adoption rates.
references (open data for browserstats)
Explore the data visually:
* Operating system stats: https://analytics.wikimedia.org/dashboards/browsers/#all-sites-by-os
* Browser family stats: https://analytics.wikimedia.org/dashboards/browsers/#all-sites-by-browser
* Page view stats: https://w.wiki/ozJ
Access the raw #opendata:
🧵 Browser adoption graphs, Wikipedia #browserstats
Firefox: 1 week (peak 87%, release every 4w)
Edge: 1 week (<= 97%, every 6w)
Chrome: 2 weeks (<= 91%, every 6w)
Safari: 1-2 months (<= 86%, yearly)
Chrome Mobile: 2 weeks (<= 80%, every 6w)
iOS: 4 months (<= 92%, yearly)
Explore these graphs further:
* Follow our blog: https://techblog.wikimedia.org
* Work with us: https://wikimediafoundation.org/about/jobs/#section-6
* Open source: https://doc.wikimedia.org/
@krinkle Are the different curve directions (some go like e^(-x), others like 1-e^(-x))) artifacts of how the versions are sorted in the legend?
@danielhglus Yeah, exactly. You can play with different sorts at the Grafana link.
I found this the least-worst rendering with minimal effort (where minimal is still an hour or so in total).
If I recall correctly, sorting by version name, means you get either indistinguishable colours (majority new versions at the tail end), or a confusing shape.
For some, this effect was particularly bad and for those I tried to make it better by subtly muting lots of smaller versions.
This Mastodon instance is for people interested in technology. Discussions aren't limited to technology, because tech folks shouldn't be limited to technology either!