Hacker News Reader

Comment by woodruffw | Hacker News Reader

In a previous thread (these "Mozilla is dead" threads appear perennially) someone pointed out that Firefox's apparent marketshare drop is potentially indistinguishable from their deployment of privacy-improving features, including stubbing out Google Analytics when "Enhanced Tracking Protection" is enabled.

I'm a Firefox user, so I have a vested interest in Mozilla's long term health and financial viability. But "marketshare nosedives" appears to be primarily an editorialization to fit the post's larger narrative.

View on HN

I posted some stats from our website a few weeks ago [0], GA was heavily undercounting FF compared to our server-side stats based on UAs.

One other thing to remember, is to check the falling of desktop usage, because a large part of the modern internet users are mobile-only, and the amount of people who use anything but what Google tells them to (or Apple allows them to) is vanishingly small.

[0]: https://news.ycombinator.com/item?id=38533109

> a large part of the modern internet users are mobile-only, and the amount of people who use anything but what Google tells them to (or Apple allows them to) is vanishingly small.

I'm a Chrome user. Both on desktop and mobile because of the built in syncing.

If I were able to switch to Firefox mobile (Android), I would. But the rendering is often broken or awkwardly different on Firefox mobile. I thought this is a thing of the past...

ffx also has syncing.

Which specific websites?

I hit a few in the past. But, I'm failing to find any of them at the moment... Maybe I should give Firefox mobile another try.

It could be that my baseline was the Chrome rendering, so any discrepancy was classified as "it's broken".

Adblock on a mobile browser is a gamechanger. I wouldn't go back even if a quarter of sites were broken on FF mobile (and they aren't, I've been using it for 2 full years desktop and mobile, and only found a single compatibility issue - a restaurant's take-out ordering page)

Honestly, in this day and age, any discrepancy from Chromium/WebKit is probably broken. But if we're talking about a few pixels' difference here and there, plenty of sites are more broken than that on every browser.

I have encountered a couple of (financial, natch) sites that insist on a Chromium-based browser in order to do vital operations like download statements. Switching browsers for those few exceptions is annoying, but not a deal-breaker (for me).

Have you tried a user-agent switcher? I think you can even set them up to always have a specific agent depending on the domain. Chances are, they don’t need chromium, they are just written by (even more than you’d think) incompetent people.

Found one! Google.com ... Try googling "vti stock" (or any similar rich ui query) on chrome mobile and Firefox mobile.

Works for me. What do you see different?

Chrome gives me the graph, and rich stock data. Firefox gives a minimum amount of data (vanilla install, no addons). See:

- https://ibb.co/gJQhTbK

- https://ibb.co/bJZ7JVR

I get the same behavior for the people queries like "Barack Obama". (Android 14, Pixel 4a)

Because Google intentionally breaks it on non Chrome browsers so that you get a worse experience.

Install the Google Search Fixer from the Firefox add on store and it looks exactly the same.

I see the graph exactly like that in Firefox on Android.

edit: turns out I have the search fixer add-on that is mentioned in a sibling comment installed.

Hm weird I have the inverse experience with Chrome…

Firefox on android also has syncing, you can open up the 'share' and send pages directly to your various firefox browsers. It will sync passwords etc. I haven't had any rendering issues and have been using FFM-Nightly for years now.

They're not saying "I use Chrome because only Chrome has syncing". They're saying "I use the same browser on mobile and desktop because of syncing, and I can't use Firefox on mobile because it has these bugs, so I end up using Chrome on both mobile and desktop."

I exclusively use FF, with Chrome on Android and Desktop only being there for cross browser testing, I never have rendering issues, only issues brought by extensions (mainly uBlock Origin), or FF’s tracking protection, both of which can be disabled.

Is it? I use Firefox on android (and desktop) and I have to drop to chrome about once a year for rendering issues.

Are you using uBlock Origin in Firefox on Android? If not, then try it.

You'd have to prise it from my cold dead hands. By far the best mobile browsing experience.

Not entirely perfect, there's a bug where I'll occasionally get a grey screen on a tab. Hasn't happened enough for me to do anything about tho.

But still orders of magnitude better than my experience with Chrome on Android.

FYI, I've been using ublock origin with Firefox on android for years and never encountered the bug you're talking about, so it might be rooted in something else. I'm using a pixel 6 ( and a pixel 3 before that), in case that's relevant

I get the occasional hung Firefox browser (browser UI works but none of the tabs render any content, or render a single tab that doesn't work). I'm not sure what's causing it, but based on the behaviour I observed, I don't think it's an addon related bug.

I wish I could reproduce it to file a bug report. These stability bugs are the reason I'm hesitant to recommend Firefox to non-technophiles, especially since the web is unusable without at least an ad blocker. While I still use Firefox, Brave seems to be doing a much better job at blocking ads without slowing down.

The person you quoted seems to have just generated this hypothesis out of nothingness, so it isn’t as if they are on super solid ground in the first place. However, I think it is worth pointing out that we’re all vanishingly small percentages of the population; your existence doesn’t tell us much or contradict them, I guess.

I'm a Chrome user

As they say, acceptance is the first step to recovery.

Interestingly one reason I use firefox is because of its builtin synching. My tabs, bookmarks, and passwords are synced. Extensions could also sync their preferences (not sure about data). I also heavily use their "send tab to device" feature, which allows me to send tabs to specific devices to act on them later.

Is GA counting Firefox on iOS as Firefox or Safari?

That’d be interesting to know, but I don’t use anything apple.

The reason I ask is that it would explain the "undercount."

It wouldn’t, we’ve been planning to make the site mobile-friendly every year for 3 years now, maybe 2024 it’ll happen :D Let’s just say the way the site works on mobile does not give us as many mobile visitors as you’d expect ;)

iOS includes the requesting application’s name in the UA string. You could use that to measure Firefox on iOS usage.

There is definitely a downward usage trend from Mozilla own stats.

https://data.firefox.com/dashboard/user-activity

Right; nobody is claiming that Firefox is doing well. But that downward trend is not proportionate with the claimed global downward trend; it's much smaller.

Firefox has lost over 20 million users in 2 years.

Jan 3 2021 210,085,100

Jan 1 2023 193,199,300

Dec 17 2023 185,599,100

I can read. The point is that the number doesn't translate: 7.5 million lost users over 2023 doesn't correspond to a 0.75% decline in global browser presence. 7.5 million is a tiny fraction of that.

7.5M out of 193M is 0.39%, so it's off by about a half. Hardly a "tiny fraction". Presumably the growth in internet users in the period would increase the relative decline in market share, too.

3.9%

You're comparing apples to oranges. The 0.75% decline from the OP is of market share across all browsers. The figure here is a 3.9% (not 0.39%) decline in Firefox users only. This would translate to 3.9% of 3.79% = about 0.14% of market share. It's not about half but more than 5x less.

It does if the global number of active browsers increased. The more people get connected to the internet, and the more of them use Chrome or Edge, the smaller Firefox' market share will be if their user numbers stay about equal.

Worth pointing out Firefox had 310 Million user in 2017. That is a lost of 100 Million user in from 2017 to 2021.

There are also more and more internet users. Marketshare and active users don't necessarily go at the same dimensions.

I'm not sure if we have reliable data on desktop computer usage stats, but I think a large part of the decline is also correlated with that - along with never having been able to significantly gain market share on mobile, which is responsible for an increasing share of people's time online.

> Firefox's apparent marketshare drop is potentially indistinguishable from their deployment of privacy-improving features, including stubbing out Google Analytics when "Enhanced Tracking Protection" is enabled.

That's an interesting angle. I suppose we could compare against Mozilla's stats on download numbers and telemetry? Though between users downloading from distro mirrors and disabling telemetry (I suspect Firefox users are far more likely to be privacy conscious and/or Linux users) those will also be fuzzy.

I would assume that Firefox revenue is proportional to usage and revenue is only down a few percent. This implies that Firefox usage is declining much slower than people say.

Not if you account for the increase in ad revenue from the company paying them.

I don’t understand how to read their “assets” table; but it looks like most of their revenue comes from “royalties,” which I’m sure has some specific meaning for these kinds of reports, but I’m not sure if it scales with numbers of users or not. If that’s the money they get from Google (huge guess alert, I’m probably wrong!), then I’m not sure if it would scale in the short there with the number of users; I guess they must re-negotiate their contract every couple years, right?

Mozilla Corp pays royalties to Mozilla Foundation to use Mozilla Foundation's trademarks. Mozilla Corp makes money by selling the default search engine position to Google, spends some of that money on Firefox development, and sends a lot of the rest to Mozilla Foundation in the form of those royalties.

So if we’re reading this and concerned about the level of support for Firefox, then we should be somewhat consoled by the fact that the Mozilla Corp uses the money for Firefox development before it even hits the Foundation?

Wouldn't that be easy to verify by just looking at the relative market shares of different browsers and checking if there is an 'unknown' browser whose market share is increasing proportionately to Firefox's decrease?

Not if the metrics come solely from analytics like GA. It would appear as an absolute decline in browser numbers instead (which would probably be indistinguishable from noise in most samples).

All I'm saying is you just need a table like this:

    | Year | Firefox | Chrome | Other |
    | 2020 |     20% |    75% |    5% |
    | 2021 |     19% |    75% |    6% |
    | 2022 |     18% |    75% |    7% |

This would make it pretty obvious, no?

No, because there are multiple independent variables there: OEM mobile browsers, for example, and the fact that the overall number and distribution of observed browsers is not static.

Firefox has all sorts of telemetry reporting so they know exactly how much usage the browser is getting.

It's not clear to me that these numbers come from Firefox's own telemetry. I couldn't find the graphs in either of the reports on their own page.

Indeed, and the data is mostly public: see https://data.firefox.com/dashboard/user-activity to check monthly active user evolution.

It's going slowly down - something that is unlikely to improve unless Firefox gets better distribution channels, especially on mobile. Giving up on FirefoxOS cut them from having the "default" browser anywhere.

They are working on a gecko-based iOS browser since the UE is likely to mandate browser diversity, so maybe that will improve things a bit - though this will only make the situation on iOS on par with Android where Firefox is not doing well market share wise.

A lot of those can be turned off. Some are turned off by Linux distributions.

Reply on news.ycombinator.com