[move-on-by]

> One day I woke up to a chart showing browser market share of FF at around 4%, which surprised me - as I thought many people would understand the implications and directions.

Its certainly low, but I also think FF usage is under reported due to built-in tracker blocking. For reference, FF uses the level 1 disconnect.me block list, which blocks StatCounter scrips from loading on 3rd party sites [1].

[1] https://github.com/disconnectme/disconnect-tracking-protecti...

[SXX]

> Its certainly low, but I also think FF usage is under reported due to built-in tracker blocking.

Of course it's not perfect source, but CloudFlare for instance have client detection purely by SSL handshake fingerprint and it's show 6-7% for Firefox:

https://malcolm.cloudflare.com/

Long ago Wikimedia also had own statistics that was pretty much relevant, but I guess it's discontinued.

[K0SM0S]

These numbers do seem in line with the propensity an average user has to switch default apps. I often compare this to how many people out of 100 bother to change their shoelaces on day 1. I'd wager less than 1, due to the friction of effort, configuration and learning curve, whereas the original usually just looks "nice", more "in place", "fitting" the decorum. Big icon is sitting there already.

Thus it's useful to consider market shares for all end-user internet-browsing devices, wherein x86 today is dwarfed by mobile and general ARM. The fact is you can pretty much correlate browser market share¹ with OS market share (which I guess retrospectively validates the anti-trust case against MS in the 90s).

- IE dominance throughout the x86 era (until early 2010s),

- then stratospheric rise of Chrome as Android took over 80-90% market share on mobile and probably 3-4x as many devices as Windows. Chrome took over everything else.

- Safari steady around 5%, about half of Apple's market share in both x86 and mobile (as low as half is surprising to me: since all iOS devices use Safari, the discrepancy between Apple's market share and Safari's must come from x86 MacOS, and I would have expected much more people to use default Safari on that OS, especially given its battery optimizations, compounded by Chrome's relative cost in RAM etc).

- Firefox is at an all-time low (still second though). Given the current trend of OS makers to expand into "ecosystems" (multiple devices/OS + cloud services), I don't find it surprising: mainstream users are more and more entrenched in the more-or-less walled "gardens" of each variety.

Clearly, Google is winning that game as of 2020 in terms of numbers, but Apple is winning too within its own (much smaller but more profitable!) userspace / customer base. MS is but failing so far, TBC, but meh it's Chromium underneath.

[1]: https://www.youtube.com/watch?v=es9DNe0l0Qo

[move-on-by]

Wow, that is a great link, thanks for sharing! Doing a filter based on device, FF doesn't even rank on Phone, and has 12.73% on Desktop. Although the data appears to only be for 24 hours - so not super great, but still an interesting data point.

[dehrmann]

Other takeways from that graph:

- Mobile internet use surpassed desktop.

- Tablets never caught on

- Windows 7 is more popular than MacOS

- Safari (mobile, undoubtedly) is the second most popular browser. IE(!) is third.

[Belphemur]

I'm pretty sure they rely on user agent in the server logs.

Sure you can spoof it, but it would be the same for the stat counter.

[move-on-by]

Do you have any sources to back up this claim? I know for a fact that StatCounter uses JavaScript [1] to collect data, much like Google Analytics. I do not see any reason why they would have a separate collection method for browser statistics. Even if they do have alternative collection methods, if they combine that data source with the javascript collection method - it would still cause FF usage to be under reported.

[1] https://statcounter.com/support/faq/14-how-do-i-install-stat...