4 ms·

I was shocked to see the browser market share graph. Chrome appears to be collapsing on itself, dropping from from an all-time high of 65.4% to 55.4% in just 16

by abtinf 7y ago

I was shocked to see the browser market share graph. Chrome appears to be collapsing on itself, dropping from from an all-time high of 65.4% to 55.4% in just 16 months, effectively losing all share gains since 2016. But then I looked closer at the data, noticing that only net 120 basis points of the change can be explained by share changes in the other browsers. The rest of the data seems to be missing, as the numbers add up to less than 100%. What the heck is going on? Edit: It goes without saying that the article's failure to address this point fundamentally discredits the author. It should be obvious that the first thing an article claiming monopoly would need to do is explain the plummeting market share of the thing that is allegedly a monopoly..

ihuman 7y ago
I think w3counter's graph only shows the major browsers, and doesn't include data from the rest. Wikipedia has a breakdown of browser usage with more browsers (and an "other" section) https://en.wikipedia.org/wiki/Usage_share_of_web_browsers#Summary_tables https://en.wikipedia.org/wiki/Usage_share_of_web_browsers#Su...
kgwxd 7y ago
60% of the time charts are wrong every time.
gilrain 7y ago
Maybe fewer people using web browsers to access Internet content? That could be explained by social networks like Reddit pushing users toward single-purpose apps.
rwbhn 7y ago
Here's a different source that shows no such drop. These stats are highly dependent on how/where one measures. https://gs.statcounter.com/browser-market-share https://gs.statcounter.com/browser-market-share
sebazzz 7y ago
I am also wondering why Opera (=Presto) is in that graph. Since the latest Opera is Chromium-based I would expect it to be counted as Chromium.
- IggleSniggle 7y ago
  Importantly, implemented on top of Chromium does not mean feature parity with Chrome.
  - sebazzz 7y ago
    Neither security parity.
wpietri 7y ago
"Fundamentally discredits" strikes me as extraordinarily dramatic. Yes, there's a dip right at the end. But browser stats are pretty noisy, so it could be nothing. Statcounter, for example, doesn't show that drop: https://gs.statcounter.com/browser-market-share#monthly-200901-201905 https://gs.statcounter.com/browser-market-share#monthly-2009... And Wikipedia's data similarly doesn't show that big falloff: https://analytics.wikimedia.org/dashboards/browsers/#all-sites-by-browser https://analytics.wikimedia.org/dashboards/browsers/#all-sit... Even looking at just the original one graph, the only thing showing a big rise is IE, with a 39% jump in usage. The author makes clear how that fits his thesis, in that Edge is now built on Chrome.

by abtinf 7y ago

ihuman 7y ago

I think w3counter's graph only shows the major browsers, and doesn't include data from the rest. Wikipedia has a breakdown of browser usage with more browsers (and an "other" section) https://en.wikipedia.org/wiki/Usage_share_of_web_browsers#Summary_tables https://en.wikipedia.org/wiki/Usage_share_of_web_browsers#Su...

kgwxd 7y ago

60% of the time charts are wrong every time.

gilrain 7y ago

Maybe fewer people using web browsers to access Internet content? That could be explained by social networks like Reddit pushing users toward single-purpose apps.

rwbhn 7y ago

Here's a different source that shows no such drop. These stats are highly dependent on how/where one measures. https://gs.statcounter.com/browser-market-share https://gs.statcounter.com/browser-market-share

sebazzz 7y ago

I am also wondering why Opera (=Presto) is in that graph. Since the latest Opera is Chromium-based I would expect it to be counted as Chromium.

IggleSniggle 7y ago
Importantly, implemented on top of Chromium does not mean feature parity with Chrome.
- sebazzz 7y ago
  Neither security parity.

wpietri 7y ago

"Fundamentally discredits" strikes me as extraordinarily dramatic. Yes, there's a dip right at the end. But browser stats are pretty noisy, so it could be nothing. Statcounter, for example, doesn't show that drop: https://gs.statcounter.com/browser-market-share#monthly-200901-201905 https://gs.statcounter.com/browser-market-share#monthly-2009... And Wikipedia's data similarly doesn't show that big falloff: https://analytics.wikimedia.org/dashboards/browsers/#all-sites-by-browser https://analytics.wikimedia.org/dashboards/browsers/#all-sit... Even looking at just the original one graph, the only thing showing a big rise is IE, with a 39% jump in usage. The author makes clear how that fits his thesis, in that Edge is now built on Chrome.