Comment by jchw
7 hours ago
Mozilla making choices specifically to try to "keep up" with Chrome has objectively failed: Firefox's marketshare has plunged so far that it's getting harder to keep using it on the web, as developers simply stop testing or caring much if it works with their site.
I think this is because simply chasing Chrome didn't have the desired effect anyway. By doing so, Mozilla hurt Firefox by gradually reducing what differentiated it from Chrome in the first place, relegating it to be kind of just a worse Chrome that looks even weirder.
A similar statement would be something like, "Firefox may as well pledge to support Web Environment Integrity; otherwise, people will just use Chrome."
In a totally different vertical, Valve has made interesting choices. To my knowledge, they do not provide a way to do attestation on Steam hardware devices, so games that want to support Steam hardware simply have to get rid of their intrusive DRM and anti-cheat, or at least significantly loosen it. Depending on how you look at it, this has either been a massive success or a massive failure. After all, a lot of big games don't run; Fortnite, Valorant, etc. But on the other hand, it remains a very powerful incentive for publishers and developers to reconsider their choices, as these devices still do have a non-trivial marketshare. (And for smaller games this obviously is hard to ignore, probably often moreso than intrusive DRM.)
Mozilla may have been ultimately correct that the choices they made helped them stay relevant; it's hard to know for sure how much their marketshare would be different today if they made bolder choices, but in the long run I think it has doomed them, as it is now unclear what even really differentiates Firefox. The last remaining thing seems to just be that you can run the full version of uBlock Origin, which is pretty valuable and all, but I'm not sure it makes up for the deficiencies. (I also like Container Tabs, but I think that feature is pretty niche, possibly somewhat by virtue of the fact that Firefox is now pretty niche.)
I'm currently using Librewolf. My long term hope was that Ladybird would become the browser that does what Firefox should have, but when I asked Andreas Kling whether or not Ladybird would try to support EME/Widevine in an HN thread, there was no reply, which is pretty close to "probably".
We already have "no compromises" forks / side-versions of Firefox. FF cannot pull a Steam because they don't have a big enough userbase, and the userbase won't grow if they are too purists, sadly. Firefox lost the battle to Chrome over 10 years ago, partly due to its faults (it was heavy and slower to load that back in the day Chrome) and also because Google shoved Chrome down its users' throat (I perfectly remember the Chrome banner in Google's home page, not even SERP).
They can't pull it now because they don't have big enough marketshare.
Gee I wonder why that didn't work out.
Surely not because they weren't pure enough. The browser market was always so big that pure alternatives were never with any advantage: on the contrary, they had more issue because they catered to a loud minority that can always push for a "you are not pure enough" narrative. Because who push for that just doesn't care (rightfully) to be mainstream.
Thinking back at the actual history (and I lived it, having used Netscape Navigator, the Mozilla Suite and Phoenix 0.1), my view is that when Firefox was winning marketshare in Windows-land over IE, the release of Chrome and Google behind it basically killed (slowly) its momentum [1]. Even if the momentum was surely paid for by Google who was the default search engine in Firefox and paid millions to Mozilla for it.
So, to answer your question: Google is the reason it didn't work out.
[1] https://en.wikipedia.org/wiki/Usage_share_of_web_browsers#/m...
1 reply →
>I think this is because simply chasing Chrome didn't have the desired effect anyway. By doing so, Mozilla hurt Firefox by gradually reducing what differentiated it from Chrome in the first place, relegating it to be kind of just a worse Chrome that looks even weirder.
That logic might work for firefox's product direction as a whole, but narrowly looking at the EME issue, what should have they done? If they didn't implement EME, then people will have to install chrome to watch netfilx. The users who are there as a matter of principle (eg. HN readers) might still use firefox as their daily driver, but they'd almost certainly lose some normies who were using firefox on inertia. Meanwhile because they refused to negotiate, they won't be able to influence the direction of the standard, which means no sandboxing of DRM modules and other privacy features. Both firefox and their users lose out in this scenario.
Simple: allow those users to be lost to preserve the product. The entire Internet is not Netflix. Most Netflix users today don't even use their laptops or desktops for Netflix: they use it on their TVs, phones, video game consoles.
Firefox is a bad way to use Netflix anyways. It doesn't support high resolutions because it has limited Widevine support anyways. It doesn't support Widevine l1 and never has. So... It is a worse way to watch Netflix.
I'm sorry but this "we can't possibly take a risk like this" mentality is cancerous. It is comfortable to "play it safe" and all, but nobody takes any risks for anything anymore. It's not calculated when the calculation is "avoid risk at all costs".
> Simple: allow those users to be lost to preserve the product.
Lose the users lose the funding. Firefox can't live out of HN as its sole userbase. Have you checked how much it costs to build and maintain a modern browser?
Plenty of users who have no money for TV or consoles, watch it on their laptop on low quality. I am one of those who watches Netflix on Linux Firefox :)
This isn't just any risk. It's an existential one for Firefox. You are asking them to make something that would be unpopular to the majority of its userbase at a point when they are borderline extinct. You are asking Mozilla to jeopardise Firefox for the same of a feature that YOU dislike enough to want to burn Firefox to the ground to get rid of it.
5 replies →
Yes, this is a no brainer. Whenever someone accesses NetFlix on a Firefox with no EME, just display a page which has a short description explaining why Firefox doesn't support EME ("EME is not included to protect your privacy as it allows websites to profile and identify you more easily") and provide a button to open Netflix either through a Netflix app or the pre-installed OS browser (Edge or Safari). Educate your users rather than treating everyone as dumb.
Or, reintroduce browser plugins (with permission based architecture).
1 reply →
> It doesn't support Widevine l1 and never has.
Firefox supports Widevine Level 1 on Windows 11 devices with the requisite hardware and has for a year or so.
Not many providers other than Netflix support it still, but it does.
2 replies →
>Firefox's marketshare has plunged so far that it's getting harder to keep using it on the web
any examples? i only use firefox, including for work, as well as encourage all of my students to use it, and i cant think of any sites that dont work.
if anything, i would say its sort of the opposite (getting easier to use). some microsoft websites were wonky ~5 years ago, but now all seem to work perfectly in firefox.
I fairly regularly come across websites that are broken even in normal Firefox, like, even on Hacker News. I've had to use Chrome for some local government stuff but I reckon they broke Firefox on accident rather than on purpose and hopefully it is fixed now. Google's own Cloud Console was unusable in Firefox for a couple weeks a few years ago :) This is very different from sites not explicitly supporting Firefox, but it does happen. If I think of/find an explicit example of something broken right now I will link it here.
To be completely fair, some of it isn't really by choice. Like, the qmk configurator doesn't support Firefox, but that's because it needs APIs Firefox intentionally doesn't have. Should Firefox have these APIs? The WebUSB thing doesn't personally bug me the way Chrome implements it so I'd say sure. I'd certainly prefer having that over EME.
A lot of websites will "work" because they added Firefox workarounds. But to a Chromium or Safari user, it might as well be called a "degraded" mode.
A lot of video-conferencing websites work that way as Firefox has had a lot of issues keeping up with the WebRTC standard evolution that power them for example.
A lot of video-conferencing websites work that way because Chrome shipped a pre-standardization form ("Plan B") of WebRTC, it got a lot of uptake, and then they refused to unship it when the final standard came out with an alternative way of doing the same thing. It took a decade to finally give in and unship the nonstandard path.
https://groups.google.com/a/chromium.org/g/blink-dev/c/IY2am...
https://webrtchacks.com/is-everyone-switching-to-unified-pla...
Then they did it again with MediaStreamTrackProcessor and MediaStreamTrackGenerator.
Then they did it again with Insertable Streams, which is still an issue for interoperable E2EE (Firefox ships a standardized variant).
If you think of websites as running on a standardized platform, then the workarounds are for Chrome, not Firefox. They include, and default to, code that calls unspecified APIs and then the ones that care autodetect when that doesn't work and fall back to APIs with actual agreed-upon specs.
Describing this as "Firefox has had a lot of issues keeping up with the WebRTC standard evolution" is... inaccurate. My impression is that there have been, and continue to be, areas where this could be leveled as a fair complaint. But refusing to implement rejected proposals because they had too much uptake already? That's the only way to keep the web platform's foundation solid.
1 reply →
Government websites tend to be an issue.
Had to spend an hour on the phone trying to figure out why my dad couldn't access a government portal. Only supported on Chrome.
> it's getting harder to keep using [Firefox] on the web
I've heard this said for the past ten years or so and so far haven't found it to be true.
Firefox should've gone the same route as Internet Explorer and forked Chromium. Firefox could focus its energies on creating features that are different instead of recreating the same ones.
Great, so any minor bug in Chromium's implementation of a spec is going to be forever embedded in the spec as the only allowed way to do it?
Multiple independent implementations are the only practical way to ensure a self-consistent and reasonable platform. Sure, it's not a disaster in an area where people can recompile all their code when something in the platform (eg the compiler) changes, but that's now how the web works.
> But on the other hand, it remains a very powerful incentive for publishers and developers to reconsider their choices, as these devices still do have a non-trivial marketshare.
There's really nothing to suggest this has been the case.
I very much expect Valve to implement a hardware backed attestation and signed driver model for future hardware.
> There's really nothing to suggest this has been the case.
Yes there is, like for example, EAC and BattlEye implementing Linux support (without attestation) for the first time. It is true that not all games enable it, but idk, you're implying there is no reason to believe it, which is not true.
> I very much expect Valve to implement a hardware backed attestation and signed driver model for future hardware.
There's really nothing to suggest that this will be the case.