I assume Firefox users over proportionally use privacy extensions.Thus they overproportionally won't appear on Google Analytics and similar places, which for some statistics reduces the numbers even more than reality.
Even back then, Chrome was commonly being significantly overcounted due to user-agent string shenanigans. And these days I’m confident (without any figures, or even relevant recent commercial experience) that will have increased sharply. I expect that it is now massively overcounted, at the same time as Firefox is significantly undercounted.
Statcounter is particularly commonly used, and honestly one of the worst. Its mobile figures, for example, are completely useless because they don’t report browser versions. CanIUse figures (which lean heavily but not solely on Statcounter) are lousy and unrealistic due to some of these sorts of issues, and just generally being out of date. (I examined the matter closely on 2023-05-27 and the figures corresponded with being about six weeks behind.)
"What, do you not live your life based on 99th percentiles?"
I tend to think of work as upside-based on downside-based. Most feature work is upside. 10% lift on conversions is great, 40% adoption is winning, and you're playing for the moonshot of 10x. Infra work is downside-based. 98% secure, 98% available, 98% acceptable performance -- that'll all failure. Winning means the thing works as expected and nobody notices.
Not everything sorts cleanly into upside vs. downside, but a lot does. Allocate your risk accordingly.
Arguably, if you only have a website, that won't work for anyone without access to the Internet. So then you should have a physical presence in each of those people's location, and arguably you shouldn't provide any improvements that give me more than physical presence does, so you should not have the website in the first place, since people without the Internet can't use it or you have to keep your website without any improvements over the physical office.
If you only have a website, arguably 2+ billion people currently wouldn't be able to access it.
And it seems odd to bring 150 mil people as an example, when the baseline should be at least 2+ billion with website only.
Most websites are glorified rich text or forms. And most of the rest should be that. This is even more true for the kinds of websites people need to use rather than some designers art experiments. They don't actually need all these fancy features except to make their developer's work slightly easier.
With CUDA you try to keep all threads doing the same thing. Sometimes that’s very difficult, but if each thread does the same thing 98% of the time, is that enough? Well since there are warps of 32 connected threads, you might expect the probability that any thread in the warp diverges to be .98^32, or 50% of the time spent with one thread in the diverged code. 50% still doesn’t sound that bad unless threads diverge at different times, and then 50% warp divergence might mean a 16x slowdown overall. 98% isn’t enough in this case.
On the flipside, if a client enters enough venues that refuse entry to them because of something the client can fix on their end, eventually the client will probably change themselves -- "If you meet one asshole during your day" and all that.
To bring the analogy back to browsers, if a website works fine for a client, they'll have no pressure to change anything on their end -- why upgrade from Windows XP when the site looks fine in IE6? Eventually the client is forced to upgrade -- normally by their operating system. That works, but what if the operating system adds another 2 years to their end of life -- do you just hang on and hope the shim / hacks you added hold?
(Of course, this assumes that each decision is independent, which, when you're talking about browser support for CSS, is certainly not the case.)
So, 10% of android users dont have web features beyond, at best, June 2025.
caniuse.com does not track this - they lump all Chrome for android together in the latest version.
This is painful as someone who wants to make use of some very useful, powerful new features, but is targeting people who are most likely to have old, slow, not-updated devices...
Which seems to indicate about 4.8% are below Android 9.
But also, Firefox for Android still supports Android 8, of which there are 1.7% below.
There's a discussion to be made here about who is dropping support for these users, is it Google (and especially Apple, who doesn't allow other browsers on iOS) or the site owner? Especially given how insecure it is to use outdated browsers.
If you add up the distribution inclusive of Android 9 (which is what I was trying to refer to, perhaps unsuccessfully), it is 9.2%. That corresponds with the 90.7% Cumulative Distribution for Android 10
If you're arguing that it is Google who is dropping support/making people have insecure browsers, we're in agreement. As with Safari (or at least those at Apple who control/fund Safari), the Android team is very anti-Web/Chrome. Lots has been written about all of that at https://infrequently.org.
Also, since this article/post is about 98%, Android 7 and below account for 2% of usage still, and its max Chrome version is 124, which was released in April 2024.
Close, but the other way round. Don't avoid a feature because it lacks good enough support. Write code to progressively enhance the experience if the feature is supported in the user's browser. If you're not willing to do that, then don't use the feature.
Progressive enhancement today means you can use pretty much any browser feature you want. You just have to do a bit of legwork with some @supports or JS prototype checking after doing the basic version first. It's not really much extra work.
The web has been doing this since, when, the 90's?
I use Firefox. In the last year, suddenly quite a few web sites have just stopped working on it for me. Firefox is over 2% market share.