A somewhat obscure guideline for developers of U.S. government websites may be about to accelerate the long, sad decline of Mozilla’s Firefox browser. There already are plenty of large entities, both public and private, whose websites lack proper support for Firefox; and that will get only worse in the near future, because the ’fox’s auburn paws are perilously close to the lip of the proverbial slippery slope.
↫ Bryce Wray
US government guidelines say that US government websites only need to be tested on browsers with more than 2% market share – and Firefox is getting perilously close to that threshold. While it won’t kill Firefox overnight, it would definitely make Firefox progressively more cumbersome to use for American users, and could have ripple effects elsewhere.
Honestly, good.
Firefox is a slow piece of shit anyway.
Brave is secure, fast, ad free, and not encumbered by Google’s bullshit.
Marcus1991,
Well, I find this comment to be unnecessarily rude. Perhaps more importantly though it doesn’t seem to appreciate the value of alternatives and the harms of living under monopolies when the alternatives are gone. Moreover, those who’ve been paying attention to the state of browser wars lately will understand that google hasn’t merely been a benevolent steward of chromium Their vision of the future, should they manage to get it done, is one of DRM, restrictions on 3rd party adblocking, web properties that are coded to work worse under alt-browsers.
Please don’t make the mistake of thinking chromium forks including brave aren’t effected by “Google’s bullshit”! Even assuming they have the resources to throw at maintain forks that deviate from upstream, it doesn’t automatically mean those APIs won’t become defunct if they fail to maintain a critical mass (brave’s market share is in the margins of error).
https://www.reddit.com/r/brave_browser/comments/veb6qw/brave_cant_support_manifest_v2_if_chrome_doesnt/
As just because an alt browser could continue to support manifest v2, it doesn’t mean alt browsers won’t get targeted with regressive payloads that harm the user experience anyway, just as google have been testing and are promising to do going forward.
https://www.neowin.net/news/youtube-seemingly-intentionally-crippled-and-slow-on-firefox-while-google-chrome-works-fine/
Remember that while these things are by-passable for now using simple tricks, look ahead by several years. If we aren’t careful the web may become much less friendly to alternatives. Like it or not, even if you aren’t a firefox user you still need mozilla to help fight the good fight for alternative rights. Cheering their demise just because you don’t like firefox seems very short sighted when the best odds for success for alt-browsers is when we stick together.
Well said, Alfman.
Well said indeed.
@Marcus1991 : I happen to like Firefox. In fact, I even switched to it on my smartphone. It has always been very reliable, and I can ‘t say that about Chrome despite their promises. Nowadays Chrome has proven itself to be stable, but there have passed many years that a lot of people just bought the hype and went with Chrome because it was the new kid on the block.
Brave is total a scam, only uneducated people not able to read use. Just google it.
We can tell everything we need about this person’s capacity for judgement by their choice to use Brave browser. I guess they didn’t lose enough on Dogecoin yet.
For anyone else who comes to the site though, be assured, Firefox is not slow at all! Someone needed to say that, so it might as well be me. I have rarely even found compatibility problems, and I have been daily driving it for literal decades now. There are some issues on some machines with driver support (on Windows), and I really wish they’d fix up some of the basics. They still, in 2023, don’t auto select an appropriate color profile for proper color space management, even though support is there for that feature. (This is worse on multi-monitor setups, even on macos). Mozilla does have a lot of low hanging fruit they just need to prioritize, but slow, Firefox is not.
Brave isn’t a true alternate browser… its merely yet another fork on top of the blink engine. And its not even an “upstream” contributions can go to its a leach fork.
Firefox at least has their own implementation. And I dunno what hardware you are running Firefox on, but they’ve been about the same in practice for the last decade with minor differences. If Firefox is “slow” its because it doesn’t come with an adblocker by default, so I install one… its not slow then.
It’s not slow even without an ad blocker. In fact, from the reports I see around (with profile data attached) most of the slowness reports come from badly behaving ad blockers…
I have seen slowness on occasion, particularly on Linux, when the acceleration features are turned off, or on Windows with certain badly behaving (usually Intel) video drivers (don’t get me started on Windows and display drivers…) – Firefox does turn on its acceleration features too conservatively. IMHO, they should turn them on and let things break, rather than letting the default be “slow” in those cases. Broken because drivers are bad is better than “slow” for reputation. You want the user to blame the correct cause. But it’s not slow without those conditions.
In which case you need to annunciate why acceleration is slow…not just disable it, even though you still must disabling it. Worse performance is better than breaking pages.
“Based on my personal experience, I would say the Raspberry Pi 4 1.5. The performance gap between Brave (or Chromium in general) and Firefox is particularly evident on this platform. Contrary to Mozilla’s assertions, Firefox is not a fast browser, and it is not getting any faster. I have exclusively used ESR editions since version 10, and the decline in speed and overall responsiveness is particularly noticeable between version 52 and 60. Each subsequent iteration of ESRs has only exacerbated this trend.”
Would you and others be extremely clear what you mean when you talk about performance? Is responsiveness page load times? DOM manipulation time or what precisely? I don’t want a generic hand-wavy explanation, but something specific and objective that can be measured.
I’d like to try and measure things more objectively but to do that I need clarification about what the issues allegedly are.
As for historical analysis, Mozilla archives contain all their old releases.
https://archive.mozilla.org/pub/firefox/releases/
Apparently it is much harder to find a historical archive for chrome & chromium…
superuser.com/questions/920523/where-can-i-download-old-stable-builds-of-chromium-from
stackoverflow.com/questions/54927496/how-to-download-older-versions-of-chrome-from-a-google-official-site
commondatastorage.googleapis.com/chromium-browser-continuous/index.html
“Precisely” the latest Firefox run slower than the latest Chromium on the same OS instance and on the same hardware. It’s overall impression. Pages load slower. DRM’ed pages load slower. (I do not use Netflix but I watch polish TV called TVN24 and polish edition of Eurosport channels) Framerate of videos is lower on Firefox. The interface is more laggy on Firefox. Recent introduction of hardware acceleration of Firefox on Raspberry Pi 4 changed very little.
No need to compare the performance of older Chromium builds. Just to compare current versions of both browsers seem to be enough to notice this very clearly.
I mentioned the tendency of Firefox becoming slower and slower in time because I used it (almost) since the beginning. Version 0.9 AFAIR.
autumnlover,
That’s not very precise, actually. You’re forcing me to come up with my own tests rather than use yours. But it’s fine, I tested a couple websites.
I took the average of three and timed them using screen recording timestamps. These results show chrome may have a slight advantage, though it’s not very dramatic. I noticed that the ads cause a lot of variation (in both browsers), this is likely due to external factors (ie the ad server) rather than client side performance. For this reason, I think it makes sense to exclude them for performance considerations.
> though it’s not very dramatic.
Now embark on this experience on Raspberry Pi 4 version 1.5 and the latest Raspberry Pi OS.
The results will be undeniably “dramatic,” rendering them undeniably noticeable. Why sacrifice your daily browsing experience merely for the sake of some intangible “freedom” or “ideology”? If Mozilla is unwilling to enhance its product or even heed its users’ feedback (consider the recent imposition of the “extensions” icon on the taskbar) despite a dwindling user base, perhaps it should cease to exist?
Chromium is NOT “the next Internet Explorer.” It excels as a multiplatform and open-source browser. If someone finds the standard Chrome/Chromium experience unappealing, alternative UIs like Brave or Vivaldi await their attention. Intriguingly, Vivaldi currently surpasses Firefox in terms of customization capabilities. You can easily hide that irksome “extensions” icon in Vivaldi using standard settings 🙂
niebuszewo,
You do you, but for the record this is exactly the sort of handwaiving I wanted to avoid. Is there a reason that the percentages would be drastically different on a raspberry PI? Maybe or maybe not, but those assertions without the use of empirical data is not objectively helpful. Too many people just say things that reflect how they feel rather than how it is.
Do you mind actually timing it for real?
>. Is there a reason that the percentages would be drastically different on a raspberry P
> Is there a reason that the percentages would be drastically different on a raspberry Pi
Maybe because it falls into category of “less powerful hardware” where differences are obvious without any “timing” ?
And it made me a little bit curious. How can I perform such a test?
autumnlover,
Well, the problems I have with subjective experiences are that they can be affected by conscious and/or unconscious bias (ie, audiophiles claiming one thing, while A/B testing suggests another). And even without bias, subjective experiences are hard to calibrate & convey between strangers on the internet. In short, I choose to put much more emphasis on empirical data.
It really depends on what one wants to measure, which is why I asked originally. Javascript benchmarks can be done entirely in browser. FF has a page performance tool, but in order compare apples to apples without relying on browser specific tooling I recorded the screen and looked at the timestamps to measure how long it took to render the websites.
https://www.maartenbaert.be/simplescreenrecorder/
I don’t know if this works with wayland, might have to find another tool for that.
https://www.labcorner.de/screen-capturing-in-wayland-on-ubuntu-21-04/
mpv media player lets you see the frame time down to the millisecond.
I ordered a RPI v4, so maybe I’ll be able to perform a test on there soon.
The knock on of this is suddenly no government employee will be allowed to run an “unsupported browser” so those users will get migrated away, further reducing the user base in America. Vicious cycle continues.
What are you talking about? I will need a source that federal employees are allowed to install their browser of choice today. LOL Most organizations only allow supported software and that probably doesn’t include Firefox while restricting the installing of software in general. This post is about the possibility of federal WEBSITES not officially supporting Firefox in the future.
There are multiple options of browsers that can be used. Below is a link to the US DoD as an example;
https://public.cyber.mil/pki-pke/end-users/web-browsers/
US government machines and US contractors are not a single homogenous entity. There are thousands of distinct teams across departments that manage installs and approved software (and hardware supplier) lists. And that’s without taking into account the various tiers like state and local levels.
UK government is more open on the information, but Chrome, Firefox and Safari are all supported and installable on governmental machines (although to be clear that doesn’t always mean end-user installable).
https://docs.publishing.service.gov.uk/manual/browser-support.html
As an adjunct, in the services I personally set up you normally had a UEM service/portal which provided the option to download only from the approved apps list. So for example, an iPhone user could download Firefox or Chrome, but not Brave or Opera (as unsupported browsers)
Honestly…
Firefox needs to invest in being more compatible with Chromium. There are web standards for a reason, and most of the stuff Firefox doesn’t support is related to implementation of standards. Just admit that where the standards are vague, there’s no point in “going their own way”, and prioritize compatibility more. My two cents as a loyal Firefox user who rarely has issues, but has witnessed some egregious compatibility issues on certain websites (e.g. complete failure to load a page).
I don’t understand why Firefox usage is so low, it’s really my favorite browser… Those who have been on Chrome for long should really give another try to Firefox. The windows version is very nice too.