Hacker News new | past | comments | ask | show | jobs | submit login
Choose Firefox Now, or Later You Won't Get a Choice (2014) (ocallahan.org)
698 points by jakub_g 3 months ago | hide | past | web | favorite | 329 comments

Just so we can avoid rehashing the discussion whenever a post encouraging the use of Firefox is posted.

- Yes, Mozilla is still the better browser when it comes to privacy.

- Yes, Mozilla has made missteps with Pocket and Mr. Robot.

- Yes, it is slow/resource heavy on certain Macs with non-default resolutions. Yes, Mozilla is working on a fix.

- Yes, it has become really fast for most users after Quantum improvements landed, and likely will continue to get better.

- No, you can't just "fork Chromium" if you don't like the way Google is running the project. Web developers will still make their website work well with whatever Google releases, regardless of standards.

- Yes, Firefox doesn't feel native on your platform of choice.

- Yes, neither does Chrome (doesn't support dark mode on macOS)

- Yes, Chrome has better security against malware.

- Yes, Firefox removed the feature that was essential to your workflow, even though most users don't care.

edit, thought of one more

- No, the argument "Chrome will be the new IE if you don't use Firefox" doesn't matter to most users.

You missed Webrender.

I switched over the firefox a month ago from chrome. While I had tried doing that a few times over the past few years, i came back to chrome because honestly there was not much difference and chrome just feels like home. This time around I had switched to try out a new tech called 'webrender'. Webrender takes a completely different approach to rendering and compositing, instead of maintainting layers and repainting certain sections it renders and paints content like a game engine.

Chrome can get jerky on a 4k monitor with lots of tabs open, but not firefox . With webrender everything is silky smooth. It's so good that I used hacks to also enable it on my intel laptop ( webrender is currently not supported on integrated graphics by default ) and the result is a browser which feels and works smoother than anything that I have experienced before. Everytime I open chrome now it just feels laggy and slow in comparison.

Webrender is exactly what I was thinking of when I said

> [Firefox] ... likely will continue to get better.

Any pointers on how to do it?

Currently webrender only works on nightly builds unless you have NVIDIA hardware on Windows 10 (not considered mature enough on other hardware/platforms yet).

But to enable it you turn on gfx.webrender.all in about:config.

For NVidia/Win10 users, is it in Stable, or only the beta channel? I have that configuration available but I still don't see the gfx.webrender.all key in about:config.

The "gfx.webrender.all" about:config preference to enable WebRender is only available in Firefox Nightly (not Beta or Release channels). The WebRender code exists in Beta and Release and might be randomly enabled for A/B experiments.


Going by this bugzilla thread it seems like it should be in stable. But maybe they changed it back.

Though the hardware requirement is actually NVIDIA desktop hardware on Windows 10, i.e. laptops are currently excluded. Which is implemented by checking if your computer has a battery.

Yup, this is an NVidia GTX 1080Ti on a desktop, should've been fine. Based on the other response it looks like the toggle is only on Nightly in Windows too, they're only doing random A/B tests on the other channels. Too bad.

A little OT: How do you use nightly? Do you have a cron job that downloads it? Do you download it every day? Do you only occasionally download it? Or is there some other way? I'd love to use it more but last I checked they didn't offer auto-updates to it.

You download the nightly installer from mozilla's website, and then it bugs you to "Restart to update" everyday. I run nightly on many of my systems, but I don't update every day

OK thanks.

Not yet available on Mac it seems.

It's available on Mac via about:config, on nightly. (All caveats about pre-release software apply.)

I've been using WebRender as my daily driver on Mac for months now.

I wonder if it is possible to subscribe to something to get an update once it's available in any form on Mac?

There should be a FeatureAlert service that notifies you. People should be able to crowdsource feature requests, and devs would implement the most popular ones.

- Yes, Firefox does support Containers, which are a very innovative privacy feature allowing you to maintain separate browser sessions (cookies, site preferences, ...) towards different websites inside a single browser instance.[1][2]

[1] https://support.mozilla.org/en-US/kb/containers#w_what-are-c... [2] https://www.thechiefmeat.com/guides/containers.html

- Yes, Firefox dropped support for remapping keys, an integral UX feature that was present in Super Metroid (SNES, 1992).

(Well, that’s not fair, they still support an API for remapping keys that only takes effect after a page tab has loaded.)

Yes, I'm eagerly waiting for them to support a proper keyboard shortcut API since I'm a former Pentadactyl and present Tridactyl user so I know your pain.

Ditto! I maintained Pentadactyl compatibility for a while when the dev team pushed breaks and the extension's metadata needed to be updated. You'll love my Hitler parody about my reaction to them breaking unsigned addons, which specifically references Pentadactyl.

"If every addon must be personally signed by Mozilla, why not just shrink-wrap the browser and make me get it from Microsoft at Best Buy?"


An almost completely non-discoverable feature which most people won't even understand, let alone use. I think it's awesome as well, but stuff like this doesn't move the needle.

Yes, the discoverability is definitely a problem, though I think that might be on purpose and that Mozilla might be planning to do something user-facing with the feature. But I disagree it's that hard for users to eventually understand and accept, if presented right. Tabs were also a niche feature once.

Interesting. I guess it may be possible to use it to be logged into and use different email accounts, e.g. foo@emailservice.com and bar@emailservice.com from the same browser instance. That would be helpful.

this is what I am doing now.

Plus, opening facebook? BAM instant new container automagically open for me, to keep facebook in its own walled garden :D

Ha. Reminds of a quote, vaguely:

In a world without Windows, something something doors (or Gates, or some other permutations of those).

Though I don't remember the quote, I remember it being a good one.

Maybe: In a world without doors, we won't need Windows or Gates?


Yes, you can use it for this (I do) and it works very well.

Good to know, thanks.

Alternatively, you could use a desktop email client like Thunderbird to do that :)

I guess so. Used them in the past, not for quite a while now. Should give it a try again. Thanks.

In fact, I have friends who use CLI or TUI email clients, like mutt. Never really used those for than a little, somehow, despite being a long-time Unix guy. Should try that out as well. Likely be a lot faster than either desktop or webmail.

Yes this is why

> Mozilla is still the better browser when it comes to privacy

Not just privacy. When Chrome came out it could literally spank Firefox' ass in speed; there was absolutely no contest then. But over time Chrome became bigger and bigger while the folks at Mozilla optimized FF code so that today the latest Firefox can outperform the latest Chrome while still being more trustworthy privacy wise. To me as of today there's not a single reason to use Chrome over Firefox, and users should also avoid using products requiring Chrome; just remember what became of the Internet when web developers started making sites that required IE to work.

Indeed. There are many more reasons than one for why Firefox is better for privacy though, such as Chrome and even Chromium phoning home by default. Containers deserve a special mention in my opinion since it still isn't a widely known feature and is a very nice idea.

I tried this and what I couldn't figure out is: How can I keep my containers separate so that when I click on a link from an external app, it opens in the most recently active container? Similar to multiple profiles in Chrome. I couldn't figure this out and feels like a dealbreaker with the browser behaviors I've adopted.

Tabs open using the container they were opened from

But, for instance, if you click on a link from another app like Slack, they open in the 'default' or highest priority container.

I use both firefox and chrome. Chrome does have user profiles, and I can have multiple open at once. Separate cookies but ALSO separate extensions.

Separate browsers too, for me it's just super unpractical to always have to open a new browser window just to open facebook, personal email, work slack etc.

Tree tabs then makes its super convenient to navigate between these worlds.

I know I'm nowhere near the mainstream, but lack of tree tabs is the showstopper for me with Chrome.

You may not have been suggesting otherwise, but in case you're not aware, Firefox also has user profiles and can have multiple open at once.

not so much by default. first you have to launch firefox with a special flag. if firefox is your default browser, and your session started by clicking a link in a different program, youd have to close your browser and click the right shortcut.

unless I am mistaken and -no-remote isnt required anymore.

I love how profiles make separate extensions a feature. I can keep useful-but-rarely used extensions in a separate browser profile instead of installing/uninstalling/deciding re: extensions.

Not by default though. You need to install the "Multi-Account Containers" addon if you want to activate it.

The container functionality is built-in since it's a pretty pervasive mechanism, but yes, there is no default UI for it currently shipped. There are other interfaces to it besides Multi-account Containers, such as Temporary Containers[1][2], Facebook Container[3] and Google Container[4].

[1]: https://addons.mozilla.org/en-US/firefox/addon/temporary-con...

[2]: https://github.com/stoically/temporary-containers

[3]: https://addons.mozilla.org/en-US/firefox/addon/facebook-cont...

[4]: https://addons.mozilla.org/en-US/firefox/addon/google-contai...

> - Yes, Mozilla has made missteps with Pocket and Mr. Robot.

These are forgivable in my opinion. At least this steps are somehow understandable.

What I‘m really miffed about is their partnership with Cliqz. If Google is the devil then Cliqz is Beelzebub. I don‘t understand what they were thinking...

Support for Cliqz integrated functions in Firefox is ending. If you would like to continue using Cliqz, please install the free Cliqz add-on.


Er... You mean Cliqz, the privacy focused fork of Firefox, with a locally hosted search engine and special crowd sourced Anti-Tracking magic, that experiments with zero-knowledge, locally hosted advertising as a revenue model?

There must be another cliqz? Or are there some skeletons in the closet I should know about? Beelzebub to Google's devil is pretty strong for the people that own ghostery...

can you elaborate?

Mozilla has been experimenting for years with ways to lead users to what they want to read without going through a search engine. This is both an attempt to improve user experience, an attempt to workaround Google's search quasi-monopoly and an attempt to find alternate revenue sources that would guarantee Mozilla's independence from Google.

And yes, the difference between this and ads is paper-thin – just like the difference between search and ads is paper-thin – so it's easy to get things wrong.

One of these experiments was through a German start-up called Cliqz, in which Mozilla invested in early stage. For a small fraction of German users, Firefox used the Cliqz engine as an implementation of this recommendation engine: https://support.mozilla.org/en-US/kb/cliqz-recommendations-f...

For these users, lots of private data was sent to Cliqz (pretty much the data that Chrome sends to Google). Cliqz is open-source and there were contracts between Cliqz and Mozilla to legally guarantee user's privacy, but I do not know/remember the details. Also, I seem to remember that the default Cliqz settings in the experiment were set to minimal privacy, which wasn't very good.

The experiment didn't last long, in part because of privacy concerns, but some people were (understandably) unhappy about this, including most Firefox devs.

(I may be missing some details, I haven't followed this closely)

More details here: https://www.reddit.com/r/firefox/comments/74yo19/cliqz_and_m...

> For these users, lots of private data was sent to Cliqz (pretty much the data that Chrome sends to Google).

This. I don't get it why people point out on and on the same three mistakes that Firefox team made in the past. In case of Cliqz, I guess they want to say: "I am not going to use Firefox, because it sends browsing history to Cliqz". Isn't that a bit weird? I agree, that was a mistake and Firefox does not have crystal-clear history, but in case of Google Chrome I am not really sure how Google collect, enrich and manage my data.

Personally, I can only notice two advantages of using Google Chrome over Firefox these days: a developer console and it is "faster" to develop websites targeting only one browser. I think I just got used to dev-console in Chrome, with each release the new developer console in Firefox is way better. About the second, it still doesn't matter much, because we have also mobile browsers, which also takes a lot of our precious time during development. Is there are any other unique selling point of Chrome which I've missed?

Cliqz is a company that primarily provides a Firefox based browser with privacy-oriented changes.

Their slogan is „The no-compromise browser. Cliqz gives you relevant search results and does not leak your private data.”

Sounds good on the surface, but what they really mean with „does not leak” is leak to Google. The Cliqz browser sends every keystroke in real-time to some supposedly Cliqz owned AWS instances. I verified that myself with Wireshark. They outright lie about this in their Transparency Cockpit: ”Telemetry data do not contain any information about queries, search results or visited URLs.”

Now Cliqz is owned by Hubert Burda Media one of Germany‘s largest media groups. Honi soit quit mal y pense.

I wrote a comment about this about a year ago[1] and also one about their relationship with Ghostery[2] more recently.

Hubert Burda Media is also one of the primary drivers behind EU link tax and upload filter legislation.

[1] https://news.ycombinator.com/item?id=15427992#15428104

[2] https://news.ycombinator.com/item?id=17159497#17161135

>Honi soit quit mal y pense

Is that considered well known enough to drop without a translation?

It's notable enough to make it into the English Wikipedia:


For some time last year, the Mozilla Corporation made it so a fraction of Firefox users got their entire browsing activity siphoned to a third party company, Cliqz, so they could bundle ads in the browser: https://blog.mozilla.org/press-uk/2017/10/06/testing-cliqz-i...

What was the misstep with Pocket? I've been loving it since I discovered it thru Firefox.

Pocket used to be 3rd party for a moment, later Mozilla integrated it in Firefox and enabled integration for everyone by default. Some people didn't like it, but others love it. Pocket now belongs to Mozilla and is covered with the same privacy policy as other products and services from Mozilla (GDPR compliant before GDPR was a thing). There are plans to open-source Pocket the same way Firefox-sync is open-source and can be self-hosted. Some people just still think that browser is a stand-alone app that works without 3rd party code or integrations, they just never inspected network traffic of Firefox or any other browser.

It's just some people don't like it's bundled in by default. Seems really as a non-issue, especially since now Mozilla owns the service.

It's still an issue because you can't uninstall it, it's hard to disable completely (and on Mobile the option is buried in various menus and hard to find), and they're removing RSS/Atom feed support which would be fine if it was just to split more things out into addons like they claim, but then they're encouraging that you use Pocket instead. Mozilla encouraging the use of their own proprietary service over an open standard is unacceptable and against their mission, and yet I still have to have Pocket pop up and show me ads every time I reinstall my browser until I can figure out how to stop it from downloading random stories about things.

Mozilla was developing a private Reading List feature based on Firefox Sync when they suddenly replaced it with Pocket. At the time, Pocket was a third party with a business model that included data mining. Mozilla owns Pocket now, but it still requires users to give up some of their privacy.

Mozilla employees denied for months that Pocket paid for the integration. Eventually it came out that there was a referral deal.

Mozilla acquired Pocket in early 2017 and said they would release the source code. That still hasn't happened.

That's cool. And? I use a lot of addons. Should they all be bundled because many people happen to love them too?

To be fair, Mozilla now owns Pocket. It's not just some random extension. Chrome comes preinstalled with YouTube, Gmail etc. too. It's a bit of cross-property promotion and to be fair, Pocket is indeed quite useful.

> To be fair, Mozilla now owns Pocket. It's not just some random extension.

My understanding is that at the time it was foist in users it was very much just some random extension, tho

They did not own it from the start, but they did have a special privacy agreement with Pocket regarding user data. Also, the acquisition was a result of user feedback, something Google would likely ignore.

If you own a Kobo, it is even more useful.

In fact, I switched to Firefox because of the pocket integration, not the other way around.

Chrome has better security unless you want to keep yourself secure from prying by Google.

Would someone who actually _knows_ the answer (I know this will not deter the HN-reckon but worth a try) be able to comment on if/why this security gap exists?

Is it something fundamental to the different design choices, is it just which has had more test/resources thrown at it, and is the closing of the gap practical and/or on a roadmap?

The Chrome multiprocess isolation model is really good. Firefox devs have known this for ages (heck, there were demos of a version of Firefox 3.x with a similar model), but could not actually land this model for two reasons:

- it would break all old-style extensions and users were unhappy about that;

- it takes looooots of memory and users were unhappy about that, too.

Now that Firefox has dropped all old-style extensions, there's large focus on reducing memory footprint, to allow moving to this model without costing as much memory as Chrome.

There are still a few kinks to fix, but this new isolation model should hopefully land within a few months.

Also, all sorts of security improvements are much easier now that Firefox dropped old-style extensions. Also, switching many developments to Rust makes new code much, much safer.

Also, Firefox isolates the browser UI and Web content into separate processes now. So that all the code you download from the internet doesn't run in the browser process. Same is with plugins.

A browser written in Rust (which is probably Firefox's long term goal) should blow Chrome out of the water in terms of security, despite the massive efforts invested in that regard by Google to prop up Chrome's C++ code base.

Would it really though? Sure, there are definitely things about Rust that make coding more secure, but surely that it only one part of security?

Indeed, that's only one part of security. But it's a pretty big one!

Definitely, it prevents the buffer overflow style of exploits but there’s also a lot of leaks in how all the DOM objects interact.

It seems unlikely that FF will ever be completely written in Rust. More likely that new bits that get added are Rust, and many of the old bits stay C++.

If FF gets rewritten in Rust, it will be a decades long effort.

I would just add that Mozilla also targets security critical portions with Rust rewrites. You don't necessarily need 100% Rust to be secure (if you have the right security model).

Google doesn’t have military grade static analysis that rejects anything that could allow buffer overflow attacks?

No, they don't, because such a thing cannot exist for C++. Any attempt would basically recreate Rust, or something equivalently incompatible with existing C++ codebases.

I know of at least one design choice that resulted in Chrome preferring security vs. Firefox preferring privacy: when exactly to persist HSTS information. https://nakedsecurity.sophos.com/2015/02/02/anatomy-of-a-bro...

In some ways. This is the same browser that doesn't (or "didn't"? It's been a while since I've looked) do CLR-based certificate revocation checks because Google didn't like the standard, but will black list specific revoked test certificates to make it look like they do (Verisign had a test site for a while with a revoked cert). To be fair, they are right: CLRs are sometimes huge and downloading them is a pain, but they're still the standard and Chrome decided it didn't want to use them.

So then OCSP comes along and everything will be fine now, right? Nope, Chrome doesn't (or, again, "didn't", they might have fixed this since I last looked, but that doesn't make it any better) let you do any sort strict revocation checking, so if the OCSP response failed and a response wasn't pinned, Chrome would happily let you in without bothering to check revocation (Firefox does this by default too, annoyingly, but it does have a strict mode).

Even if these things have been fixed, Chrome's general attitude towards web standards and security is to run away and not bother when they don't get their way, so even if you're not already skeptical of blanket claims that they're "more secure", which is a meaningless thing to say, there is evidence that their attitude towards web security might make them less so.

Updated to mention "malware"

Ungoogled-chromium exists you know..

"- No, you can't just "fork Chromium" if you don't like the way Google is running the project. Web developers will still make their website work well with whatever Google releases, regardless of standards."

Hah, even compiling Chromium takes a huge amount of effort and a workday or so of time. Don't think I'd want to go spelunking through that codebase

I would imagine that compiling Firefox is just as difficult.

Firefox is much easier and faster to compile but still complicated enough that you need to follow their build guide. Chromium is held back by Google's overzealous build systems that assume everything is built on a unified build server and permanently cached somewhere on their corporate network which results in actual build times being unoptimized.

Firefox is moving to rust, and for example adding webrender to your project is just few dependencies away, so I'd expect building firefox will only get easier, unlike to google projects where everything is built with their custom tool

> No, the argument "Chrome will be the new IE if you don't use Firefox" doesn't matter to most users

But it should and advocates can make it matter to users, just like they did when Phoenix was the cure for the made for IE blues

I have never been able to move over to Chrome. Chrome always synces my bookmarks out of place. My bookmarks are how I get things done and run on muscle memory. Because of this I have been a loyal FF user for decades. I tried Chrome again last year and bookmarks still sync out of order. It boggles my mind that Google never fixed this or just isn't a problem for people (do people care?).

As an FF user for decades, esp on Linux, the experience was really poor before the Quantum upgrade. Yes, we lost a lot of plugins, but some of those plugins existed because of FF's legacy code and we are better for it being gone. FF now runs very smooth and uses so much less memory. I actually had Chrome use up all the memory on my 16 gig laptop... it was first for me since I use it so little.

The one thing Chrome does really well is profiles. I have one profile for work, that reopens all pages that were last open. And other for personal viewing in which I don't care. FF's multi account is a plugin and... it's okay for managing multiple social media accounts from one domain, but not good for separating work and play.

I think you're referring to Containers. If so, you can always fallback to separate Firefox profiles (run `firefox -p` to open the profiles menu iirc) if Containers aren't flexible enough for you.

I've been a 95% happy Containers user since it was a Test Pilot experiment but there are certainly cases where it's not enough. In my case, not being able to use the mobile device emulator in a container tab was a big pain.

Those are terrible compared to the easy profile switching that Chrome offers. Containers also don't keep the pages you lasted work on the last time I checked. (If they do please let me know.)

> FF now runs very smooth and uses so much less memory.

The big memory use fix went out a few years before quantum. Extensions like firebug were leaking memory.

> - Yes, Mozilla has made missteps with Pocket and Mr. Robot.

and the removal of RSS support. So much for the champions of the Free Web!

It was removed because

- pretty much nobody was using it;

- it's pretty easy to reimplement as an extension.

Championing the Free Web is harder if you also need to champion the dead web.

edit "nobody" => "pretty much nobody"

> - nobody was using it;

You'll need a source for that, since most Wordpress sites which constitute a HUGE part of the web have pretty much by default rss feeds. It's not because you are not using it that nobody is.

And even you premise was true (which is debatable), it cost absolutely nothing for Firefox to keep a functional RSS reader without putting much effort into it (or instead, you know, putting efforts in pushing ads to people who never wanted them).

> You'll need a source for that, since most Wordpress sites which constitute a HUGE part of the web have pretty much by default rss feeds. It's not because you are not using it that nobody is.

I'm not making it up. I don't remember the exact details – you'll have to look at the archives of the Mozilla dev-platform mailing list if you want them – but I seem to remember the usage was below 0.001% of users.

I personally read RSS with Thunderbird, which is a much better experience anyway.

> And even you premise was true (which is debatable), it cost absolutely nothing for Firefox to keep a functional RSS reader without putting much effort into it (or instead, you know, putting efforts in pushing ads to people who never wanted them).

It actually does. Every piece of software that you need to maintain is a tax. Paying the tax makes sense if the code is useful, not if it isn't.

Also, frankly, this implementation didn't bring anything to users. It wasn't nice to use. There are much better RSS readers than a browser – starting with RSS WebExtensions.

No. Not literally noone was using it. Just a colloquial 'noone' as in 'noone is using Firefox'.

Additional features that the average user does not want to use or care about creates confusion and a feeling of bloat. An official extension would be nice though.

Did users actually consume those WordPress RSS feeds using Firefox? I imagine most RSS users are going to use an RSS reader like Feedly, NewsBlur, or Google Reader (RIP). Those RSS readers don't require users to copy/paste RSS URLs. Users can search feeds by site or topic names.

Was it hard to maintain? Is RSS antithetical to the open Internet?

If the answer to both is no, they should have left it. I wish I could tell my parents to subscribe to some news sites and my website to see updates, so they dont' have to join some kind of mailing list... but I can't, because I'm not going to explain how to configure addons to my 70 year old.

Pocket is also easy to implement as an extension. In fact, it is an extension. But they bundled it with the browser.

And most people seem to be very happy about it.

Person here.

Very happy indeed. I discovered Pocket as a result of it being integrated and have been a very happy user ever since.

It's their own service.

It wasn't when they first did it.

I am aware. But the difference is Google wouldn't care and Mozilla at least did something, even if it wasn't removing the extension.

And it competed with Pocket.

Not really. People use Pocket. People don't use RSS in their browser.

Market research indicates that most people actually love RSS.

In their browser? Interesting. I would have thought people take the link to a reader/podcast-reader/torrent-app.

An implementation of RSS baked into the browser?

I'd be surprised. Do you have a source?

I don't know what intention you had when you posted that link but it clearly disproves that anyone uses the terrible built-in RSS reader in firefox.

Here, just take a look at the first response:

>Literally 99.99% of Firefox users did not use the built in RSS support[1].

>I love RSS! I also never used the built in Firefox RSS support because it was not very good. Mozilla also compiled a helpful list of alternative RSS readers for the 0.01% of users that used this feature and made it easy to export your feed list[2].

>[1] https://www.gijsk.com/blog/2018/10/firefox-removes-core-prod...

>[2] https://support.mozilla.org/en-US/kb/feed-reader-replacement...

Yoric 3 months ago [flagged]

Ok, so you're just trolling. Too bad.

Is the misstep with Pocket including it in the browser?

Yes. I like Pocket, but I can't see why it shouldn't be an extension that can be removed.

Intellectually, I agree, but

- it actually doesn't hurt;

- market research indicates that most people actually love it.

Literally 99.99% of Firefox users did not use the built in RSS support[1].

I love RSS! I also never used the built in Firefox RSS support because it was not very good. Mozilla also compiled a helpful list of alternative RSS readers for the 0.01% of users that used this feature and made it easy to export your feed list[2].

[1] https://www.gijsk.com/blog/2018/10/firefox-removes-core-prod...

[2] https://support.mozilla.org/en-US/kb/feed-reader-replacement...

Market research, in this particular case adoption rate, also indicates that people love giving all their privacy away to google.

Good point. But Pocket isn't harmful, especially if you don't use it. In the worst case, it makes the Firefox installer a few kb larger, that's about it.

The same could be said of the RSS support.

True. The big difference is that users use Pocket but didn't use RSS.

A lot of people have a really strong negative perception of pocket. That should be enough to be able to remove it.

Meaning the nontechnical set of users that don't go out of their way to disable telemetry, at least...

If you disable telemetry, Mozilla won't know about you. If you want Mozilla to cater for you, don't disable telemetry?

Telemetry should never be enabled by default without consent. Its a dark pattern.

It leaks information about my browsing history.

These "missteps" are the reason I don't use Firefox. With Chrome, I know Google are watching. But I also know only Google are watching. They're not bundling irremovable third party apps in with the browser, or installing extensions without my consent, or sending my entire browsing history to another third party.

Mozilla have done all these things, and when called out on it, they don't seem to realise they've done anything wrong. That crosses the line from naive to actively malicious.

They definitely have a few bad apples in management.

Market research show that noone used RSS/legacy addons/all the other customization options, yet Firefox market share somehow plummeted after they removed all of it.

Literally everything I liked about Firefox is gone, so I have no reason not to use Chrome instead anymore. They are mostly the same, Firefox is just uglier than Chrome now and I can't change that anymore.

Neither does RSS. Forgive me if I'm wrong, but isn't Pocket developed outside of the main Mozilla tree?

I don't remember the current status. It's probably developed outside of the main Mozilla tree, as a number of other features.

But the big difference is that people actually use it.

> Yes, neither does Chrome (doesn't support dark mode on macOS)

Yet, this will change though


Wow. That is a surprisingly complete and useful summary of the usual discussion. Thank you.

> - Yes, it is slow/resource heavy on certain Macs with non-default resolutions. Yes, Mozilla is working on a fix.

On some of them. I use Mac with non-default resolution and I have no issues.

Which is why I said "on certain Macs". From my anecdotal experience, I've never had a problem with Firefox on macOS with any resolution.

I have a problem in this area and it has prevented me from moving back to Firefox. If I run Firefox on my 13" MBPr with the panel resolution (Scaled->More Space) then the battery usage goes through the roof. I'm following https://bugzilla.mozilla.org/show_bug.cgi?id=1404042 and hope this issue gets fixed soon. It seems that Apple changes may have made this issue more apparent to users in general.

I can obviously follow Bugzilla, but does Mozilla have anyone working this from more of a support angle and provide high-level updates on the top issues like this?

lets not for get the political baggage of Mozilla: I would have liked to use firefox, but I'm boycotting it because of the awful smear of Brendan Eich supported and perpetuated by contributors to mozilla. A huge number of people would have to stand up and apologize for participating in that despicable mob action before I'd go back.

> - Yes, Firefox removed the feature that was essential to your workflow, even though most users don't care.

So did Chrome, to be fair.

I can't articulate it but I just don't "like" Firefox.

I used to be a huge fan of FF... I think with Chromium, at least for me, the browser has never changed.

Every time I try to go back to FF it's different and I have to re-learn it.

It might be better for them to clone the FF menu system so that you can switch to it without having to re-learn everything.

I'd use Firefox if I didn't have to enter my master password on MacOS every time I start the program. It integrates with keyrings on Linux, is it that hard to get it to work on MacOS which has more users? Also, does it work on Windows?

Never had this problem. How did you install/how do you launch it? Does it require root access for some reason?

I'm talking about the master password of my Firefox vault or whatever it is called.

As nobody responsible for that Mr. Robot thing got fired, I trust Mozilla about the same as Google/Microsoft.

So you don't trust people to learn from their mistakes without being fired first?

There is a difference between a mistake and intentionally pushing malicious code using channels reserved for security updates.

What the relation b/w Mozilla and Mr.Robot ? Could you please enlighten.

you missed the part where they broke all my extensions which will apparently literally never be ported due to missing API functionality

I think that was covered in "- Yes, Firefox removed the feature that was essential to your workflow, even though most users don't care."

No, you can't just "fork Chromium" if you don't like the way Google is running the project

What, why not? Beaker and others did exactly this.

Because even if you could keep up with the changes (you won't), your fork still won't mean anything for the open web or standards because nobody will care about your total market share of exactly 3 users.

Microsoft would be the one forking Chromium in this case, so I really don't think that's true. Microsoft is more than capable of doing it.

The whole reason Microsoft gave up on their own engine is because no one was using it, not even inside the company. Why would Edge suddenly become popular just because of a rendering engine change?

... because it is the default on Windows? The marketing effort of Chrome can be chipped away if "they are basically the same", which is a fair comment now that Edge is basically a glorified fork of Chromium

It will be interesting to see what message Google goes for to try to convince ChromiumEdge users to switch over.

A features-based approach seems iffy. They can't push speed or site compatibility anymore, assuming MS doesn't completely mishandle the build. Even tiny Vivaldi seems to remain coupled closely enough to Chromium releases that it will be hard to say "theirs is so out of date and missing the latest and greatest."

If they try to deliberately hobble ChromiumEdge, it seems like it wouldn't look good to regulators. "We didn't test Gapps on Gecko, oops, it breaks" is one thing, but it's hard to keep a straight face saying "it mysteriously works in our version of Chromium and not Microsoft's"

Many of the differences they'd likely differentiate on (the sticky Google login, telemetry) are not features for the consumer, they're the features that benefit Google and justify the cost of developing a browser.

> because it is the default on Windows?

Which clearly didn't mean anything [1] for increasing market share.

[1] https://ferdychristant.com/the-state-of-web-browsers-f5a83a4...

It's not nothing, it just wasn't enough by itself to overcome the competition.

In the web's short history we've already seen several browsers rise and fall. It's possible Microsoft may in the future repeat what Google achieved with Chrome if it can find a new edge to compete, pun intended.

How much success do they have?

You tell me

Since you asked, poorly. I've never heard of Beaker and when I run it through Google I have to scroll down to see anything about Beaker the browser. But I'm sure when they show up to WHATWG or W3C the room goes silent.

- Yes, Firefox's memory leaks are still pretty bad especially compared to Chrome. Quantum helped a lot though.

yes, sir!

Thank you for most of this, except: Chrome is the primary distribution method of malware, and we need to stop pretending it has better security. It doesn't.

The Chrome Web Store is full of active, known-about malware Google does not remove.

* No, Chrome isn't the next IE because it doesn't feature proprietary web features

Chrome does have proprietary web features.

That's weak reasoning. And wrong if you define proprietary as non-standard.

And it didn't work. Some people were aware of this. A few took the time to read/discuss/think about it. A minuscule fraction of passionate people acted on it.

And the rest of the world took the path of least resistance, as usual.

People don't shape society willingly. They mostly let life happen to them. Even when you see them protesting, it's the consequence of an emotional reaction and a group effect, not the reflect of their deep personnal vision of life. They have none.

It's why people keep gaining weight, comming back to a violent spouse or voting again for the politicians that just lied in their face.

I have no idea what to do about that. Education works a bit, but it's so slow and easy to destroy. And meanwhile, it's very unfair for the minority trying hard to make things better. They pay a high price, with no thanks, and little benefits except the hope the others won't screw up things more.

Be the change you want and keep talking about it. I've stuck to Firefox through ups and downs ever since I found out about it back during version 1 point something.

Internet Explorer proved that behemoths can and will fall, given enough time to shoot themselves in the foot. Firefox just needs to retain its 10% somehow.

>Be the change you want and keep talking about it.

The OP already talked about the problem with this approach to things: >"...it's very unfair for the minority trying hard to make things better. They pay a high price, with no thanks, and little benefits except the hope the others won't screw up things more."

The people who stick their necks out trying to make a change usually end up sacrificing a lot and not getting anything in return, even if their efforts are successful. Basically, they're "taking one for the team".

Luckily, with browser choices, it isn't nearly that bad; it's not that hard to switch browsers on a whim, after all, but for bigger issues (like political change, as the OP was referring to to a point, with his mention of protesting), "being the change you want" is usually not a winning strategy at a personal level, and a much more successful strategy is to simply go where the grass is greener.

I've stuck with it and preached it and told people why Google is just going to repeat IE's history with Chrome, and my tech friends still mostly use Chrome.

There is something about A) what everyone else is doing, and B) inertia, that is just too difficult to overcome without some sudden significant change.

The single biggest advantage of Chrome is that it's bundled as the default, non-removable browser on the largest mobile OS in the world: Android. And nearly all those people then expect to use the same browser on laptops and desktops and even here it is often pre-installed by OEM.

If Firefox had somehow had this clinching advantage in an alternate reality, they would be the Chrome now. The actual performance and technical differences between the two browsers are by no stretch of imagination deal-breaking. But Google's marketing and bundling has been decisive.

I just hope Mozilla doesn't abandon Gecko but sticks this fight out.

> The single biggest advantage of Chrome is that it's bundled as the default, non-removable browser on the largest mobile OS in the world: Android.

I'm still upset that Mozilla balked on Firefox OS[0]. It would have made a good ChromeOS competitor on laptops and tablets. I would have opted for a Firefox OS tablet instead of Android as a secondary device. A privacy-focused tablet would have been amazing.

Honestly don't see any reason why Mozilla couldn't restart the project on larger-screen devices - skipping the phone phase for a period. Especially now that they have their rust-powered Quantum Browser showing the world what it can do.

They could always do phones later on when the platform matures. I think they are failing to see that solutions are about ecosystems. Few can deny they have a rich, browser-based, ecosystem. They are still thinking in legacy terms - a single-use application.

I believe they have the monetary means (and the talent) to enter hardware (or via Kickstarter). Their worst enemy... they may be a little old-school, corporate management-heavy on top.

Their leadership let Firefox stagnate, the performance suffered the entire time Chrome was taking everyone's market share. We're talking years of neglect.

> If Firefox had somehow had this clinching advantage in an alternate reality.

The alternate reality... Google invests $22 million in the ported version of Fire OS, KaiOS[1]. They are helping to keep alive (and integrating themselves) into the very OS Mozilla created (and abandoned).

[0] https://en.wikipedia.org/wiki/Firefox_OS

[1] https://techcrunch.com/2018/06/27/google-kaios/

Firefox was losing to chrome long before android became so big, Mozilla dropped the ball here, they were too late on mobile.

Yea, parent comment is definitely a but of revisionist history. I stayed with Firefox longer than most, but the performance gap just became so massive that I really couldn't justify remaining on it. This was back when smartphones themselves were pretty new, let alone during android's dominance

Android phones sold in China don't come with Google Play Services or Chrome. Even my Canadian Samsung Galaxy S7 used Samsung's "Internet" as the default browser app, although Chrome did come pre-installed.

But otherwise, you are correct. For most Western markets, Android phones give Chrome a big advantage.

I would hope Mozilla will eventually retire Gecko and switch to Servo.

Recently finished moving away from most google stuff. Seeing how they are slowly grabbing more and more user data really scared the shit out of me (android location gathering, auto logging in to chrome from gmail, syncing all password data etc).

Swapping the browser was the easy part. Not using google search and moving away from my gmail address that I've had since gmail was released were harder habits to break.

Setting up a google-less android (lineageOS + microG) was a real PITA for a few weeks as I kept running into new problems and things not working properly. I don't see how anyone not into tech could pull it off.

I dumped Chrome, stopping 'Googling' and moved to DuckDuckGo, installed uBlock Matrix so I could block as much of Google as possible (damn you reCaptcha), but it's gmail that's the tie-in. Changing email address is super hard.

As for Android, I took the easy option - switched to iOS. I'd have preferred a third choice, but there's nothing viable at the moment.

It's hard going but worth it for the privacy, and I agree it's not realistic for those not fully into tech.

The world needs more choices for browser, more choices in mobile too. At least there's a few options for computing devices - Windows is a horrible privacy hole IMO, Apple and MacOS is a premium option but seems more private and secure, then there's Google (no comment). Linux can be a huge time sink and hard going but having run it on both desktop and laptop it's a good choice - but for the expensive commerce softwares that only run on Mac or Windows thanks to companies that make 'pro' offerings... for me its Adobe, Capture One, Reason - none of which offer Linux variants or work on Wine properly.

Part of me wishes the EU to legislate more, but that's horrible in other ways.

That’s hard to prove or disprove without having an alternate universe ready for A/B testing.

Although, frankly, I can’t shake the feeling I’ve been living in the B-branch for, say, the last two years. If so, let me just say: if your results are not yet conclusive, please ask your statisticians for what “multi-armed bandit” means to them, then shoot them if the quizzically look at you. You may even skip the question.

Back on point: I know many people who use Firefox for “ideological” reasons. In fact, anybody who still used FF as of 6 months ago arguably fell in that group, save some extreme cases of inertia.

But, even more tittilating: Firefox seems to be once again be competitive wrt performance. That may just be its salvation, and not a moment too early to make it dramatic.

What are you even talking about? You're making it sound like Firefox has ever been a lot worse than Chrome, but that's far from the truth.

I've used both quite a lot and I've always preferred Firefox, even _without_ considering the privacy aspects. I still use Chrome on my laptop as Firefox doesn't support decent pinch zooming (that's more or less my only complaint), but for desktop usage I definitely choose Firefox over Chrome.

Not the OP, but when Chrome came out, it was leaps and bounds ahead of Firefox in terms of performance, and it took Firefox many years (until just recently, with Quantum) to catch up. I personally endured it, but it was definitely hard after having used Chrome for a short while, and I don't blame anyone for making the switch back then.

Most users don't care about features; they just want a browser that works well.

Why would it be "extreme" inertia to still be using Firefox? If it's never felt like it's failing to do the job of being a web browser, then not switching is inertia, sure, but it's not extreme...

I have to say, one reason I've stuck with Firefox for so long is because of memory usage: Chrome/Chromium is a memory hog.

I'm sure a lot of people around here won't be able to hear you over the sound of how Firefox somehow doesn't feel quite right on their macs.

Yeah... I dont quite understand how Firefox does invest into integration with third-party password managers, but not the one built into MacOS.

Arguably Safari usage is comparable to Firefox in terms of staving off a Chromium monoculture. Safari usage on iOS is actually far more important right now to ensurage some semblance of compatibility.

The password manager integrations are extensions made by the respective third-party password managers, not by Mozilla. If you want to use Keychain, it is either up to Apple to make an extension, or someone volunteering.

There used to be an add-on that let users integrate the Keychain in Firefox. I seem to remember that pretty much nobody used it, which doesn't speak well of the usefulness of this integration.

Also, Mozilla is working on its own cross-platform password manager/password sharing between devices. I'm not sure what the intended policy is wrt password sharing (there may be passwords that I only want to have on one device), but this should be more useful than the Keychain, which doesn't even let me move my passwords from one machine to the next one without arcane incantations.

Chrome also removed Keychain support. Apple restricted access to the necessary APIs.

> People don't shape society willing…

Logged in just to say, “what a paragraph, haven't read anything with such clarity lately.”

Thanks for it. Yes, I'm using Firefox since 2.0 and never changed once.

As a long time Firefox user, let me offer a different perspective.

Upgrading any application is a chore, and you never know what will break or whether you lose your data. People don't want to do it, and yet they are forced to do it by companies (and open source is even worse). So if you can avoid it, you avoid it.

I'll leave up to you to decide if Firefox haven't destroyed its own market share by confusing changes to UI and breaking backwards compatibility.

Most people don't even know what an URL is, so I don't think they even think about their browser that hard.

I'm not sure I understand the point you're making here. Wouldn't those people be _more_ beholden to UI stability? If you understand what's under the hood, it takes a little time for your muscle memory to adjust to UI changes but it's not exactly confusing. That's decidedly not the case for the group you're describing.

Taking users away from Google is still a big ask. Google has been thoroughly anti-competitive in this space - they deliberately downgrade the experience in non-Chrome browsers across many properties. For example, typing "GOOG stock" in Chrome/Android brings up a whole lot of details not found with Firefox/Android. There's no technical reason to do this, and they should be pulled up for abuse of monopoly power. Google search itself was better on Chrome till recently, though it's caught up now on Firefox/Android.

Add: I've been a Firefox/Android user for a while now - and with ad-blocking the UX experience and battery life is much better than with Chrome. I encourage everyone to try it (with ad-blocking) for at least a week. Personally for me, Firefox on Android is a bigger deal than on the desktop.

Just downloaded Firefox on Android following your advice. I'm a heavy firefox user on the destkop, but I rarely think about installing firefox on my cellphones since 1. Firefox in Firefox OS was slow and it gave me a feeling that firefox was not for mobile (but I guess it's not true anymore) 2. The fact that Chrome is the default browser and I generally don't install many apps.

The fact that it is the default on Android makes the game very unfair.

> Firefox in Firefox OS was slow and it gave me a feeling that firefox was not for mobile (but I guess it's not true anymore)

I genuinely don't understand how Firefox on mobile is slow. I hear this complaint often (along with the complaint of Firefox on desktop not being able to run certain websites, which also never occurs to me), but for how much I try, I never managed to make Firefox on mobile (android) slow. There were a couple of days not long ago (we are talking weeks) were after an update the browser did get indeed a little slow, but it was still usable, and it got fixed very quickly.

I use the hell out of my Firefox, both on desktop and mobile. I abuse it in all sorts of ways, and I am extremely demanding of it, yet it never breaks a sweat. Right now I have more than 100 tabs open in my Firefox on android - can't tell the exact number because when the 100th tab is opened, the number just becomes an infinity symbol - which is nothing more than average for me, I can easily get to 350, and it still opens and loads up instantly (at the time of the performance problem mentioned above I had ~260 tabs open).

Am I just lucky? Do the performances/usability change so vastly from phone to phone and person to person?

Firefox on Android has a notable scrolling delay which drives me nuts. iOS guarantees that the point where your finger first makes contact remains under that finger as you scroll. Most good Android apps can deliver on this promise 99%, but for some reason, Firefox on Android does not.

Firefox on Android here; Can't reproduce the delay, everything seems fine.

> The fact that it is the default on Android makes the game very unfair.

The EU is already looking into this, and other antitrust issues with Android and even Google Search in general.

An advantage to using Firefox Mobile is google search results suddenly become free of AMP crap

I think I’m now at the point of jumping aboard this movement away from google. Not because their services are bad (they objectively are excellent) nor even because of privacy concerns (though they are legitimate), but because of the argument put forward in the OP: google are taking over everything. There is far too much power concentrated into the hands of a single company.

As an individual there is not much that I can do about this, but what I can do is to use google products and services as little as possible, and to support smaller, independent alternatives.

I think this has been posted on hn in the recent past, seems like a reasonable jumping-off point: https://medium.com/@ricst/de-google-your-life-now-82fad3ec0f...

Note this advice is true for email services as well. Pay for and use an alternative now, or later you won't get the choice.

Google is hellbent on locking people into their ecosystem. One such move is the promotion of AMP: https://techcrunch.com/2018/02/13/amp-for-email-is-a-terribl...

Speaking of email and browsers, the Gmail app on top of iOS presents the user with a choice between two browsers:

1. Safari

2. Chrome, which on iOS is a reskinned Safari, but with Sync and controlled by Google obviously

But when Safari is selected they actually open links in an app Web View. Which does not share sessions and cookies with Safari. And their reskinned Chrome doesn't do Safari's content blocking either. Firefox on iOS doesn't do that either, but at least Firefox has its own tracking protection.

Speaking of Chrome on iOS, they give the user the choice to switch search engines, supposedly, however they only list Yahoo and Bing as alternatives. No DuckDuckGo or Qwant or the ability to add your own.

Android still allows users to switch their web browser or to use an alternative app store, but I wonder for how long. I mean they disallow users to uninstall their apps. My guess is that they are afraid of locking Android down more due to fines from the EU.

Sadly what made Google great several years back was openness. But times are changing.

> Firefox on iOS doesn't do that either, but at least Firefox has its own tracking protection.

Firefox on iOS can't do this, unfortunately :(

Which is why iOS should be avoided too. They may appear to be playing kind of nice now, but they won't forever. That blatant disregard for openness is a sure sign they're going to end up the same way, no matter what they say now.

> iOS should be avoided too

What would you suggest using?

Android with as little Google as possible, which is none if you're willing to go far enough. Nothing is also an option if you're extra dedicated.

I'm curious about this. I get that firefox for ios has to use a webview, but why can't it allow extensions that can manipulate what's in the webview? The main reason I'm currently sticking with Android is that I can install firefox and use ublock origin and https everywhere.

You can run your own JavaScript in WKWebView, which is what Firefox uses, so you can run "extensions" to the extent of what the App Store will allow you to do. What you cannot do in this case is load content blockers, which a newer API that uses a list of selectors to block elements without JavaScript at all. Currently this API is limited to Safari itself and SFSafariViewController, which is untenable for Firefox because it is essentially "Safari in a box" in that the component provides its own UI and you essentially cannot interact with it at all.

Not sure what the situation is nowadays, but extensions were not supported due to Apple’s App Store policies.

I would be happy just with Safari’s content blockers, but apparently they use a type of web view that doesn’t support that either.

>I mean they disallow users to uninstall their apps

All default apps on Android live on the system partition, meaning even if you uninstalled them, regular, non-rooted users wouldn't have any additional space to use.

This policy also means you can't really mess up (for average consumers) an android phone by uninstalling something important, like webview or the Play Store.

Any updates to the default apps are installed on the user-accessible data (IIRC) partition, and so can be uninstalled by users, actually freeing up space.

Copying links out of Hangouts on iOS results in getting a massive Google tracking url.

Easy path to success for Firefox is to continue to beef up security and privacy tooling for their users and have sane defaults. They've been winning a lot of users who feel abandoned by the big companies in this space.

A problem and conflict of interest for them regarding this is that they are funded mainly by Google. Firefox has a conflict of interest here because they are not doing all they can here because that would cut off their main source of revenue.

So, for example, they have sort of a mute option for sound (just like other browsers) but they don't allow you to mute video by default (without blanket turning it off for all websites). I don't know anyone who loves being bombarded with video content when they want to read an article. I find the moving stuff to be distracting; even if the sound is off. Also the bandwidth is kind of not nice, especially on mobile. The BBC has started putting multiple video previews in their articles. This kind of attention whoring is highly annoying. This is not news to anyone of course.

The obvious fix allow people to whitelist sites to allow autoplaying video (youtube, netflix, and other websites where people go specifically to enjoy video content) and switch that off for everything else. All the extenions for that claim to do this are basically a combination of broken or simply setting the config property that blanket bans all autoplay video for every website (just preempting the inevitable comment to try extension X here).

Similarly, the containers extension is great. This should be baked into the browser and improved to the point where the default container is a private container unless you whitelist the domain. Right now private browsing happens in a separate window. It's an obvious thing to do: every site by default is private until you say otherwise.

Both kind of are in direct conflict with how Google, and by extension, how Mozilla monetizes. This is the main reason this is made hard to do. I'm sure if this was easier, there'd be countless extensions doing exactly that. Instead people sort of emulate private browsing with stuff like noscript and other extensions that throw away cookies. Also nice but not user friendly.

Doing all this in fool proof, user friendly way would be a great way for Firefox to differentiate. It would unfortunately cut them off from Google funding probably.

Just FYI, audio autoplay is being actively worked on in the current Firefox development version, precisely to solve the problem you describe. There are some issues with coming up with a UX that doesn't privilege a small set of sites out of the box over everyone else [1] and also doesn't prompt the user on every single site that tries to play video (the current Firefox Nightly behavior; it's rather annoying).

So for this specific case, the reason it hasn't happened yet is not some nefarious conflict of interest but actual hard problems that need to be solved (and are being solved), coupled with resource constraints.

As far as containers go, I agree they are great if you understand the model. I use them. Most users find them terribly confusing in their current form. Again, a UX issue. That said, the idea of having things in a private container by default is somewhat interesting. The hard part in some way would be the discoverability of the whitelisting flow and user confusion about why their customization information is not being remembered like in every other browser. I'll try to shop this around to people a bit and see if they can figure out a way to make this work sanely.

Disclaimer: I work on Firefox, but not in UX.

[1] Google's solution in Chrome; it suppresses new startups by making the experience suck there, which is good for incumbents like YouTube but maybe less good for the web as a whole in the long term.

I have an extension that blocks audio per site already (mute sites by default). Awesome stuff, but blocks only the sound. It shows a little mute icon on the tab if it mutes. I simply unmute when I actually want sound and the setting seems sticky.

I was talking about video auto playing (with the sound off) by default. I don't want that ever, unless I allowed it before. I want to block html5 content from activating at all unless specifically allowed. I'm on the beta channel. As far as I know there are no plans to address this and extensions that have tried to implement this are running into some limitations that make this hard apparently. Hence a lot of variations of flashblock extensions that simply set the about:config setting instead of having per site policies with whitelists and blacklists.

Private browser containers would be nice to have even if they are not yet default out of the box. I'd simply make this an option on all containers (including the default one and default it to off). I agree the ux for this would need a bit work to be discoverable/intuitive.

Ah, I had misunderstood the issue; thank you for clarifying.

It looks like https://bugzilla.mozilla.org/show_bug.cgi?id=1405827 tracks WebExternsion API that would make it easy to create extensions that address this use case.

Most people don't care.

Firefox became popular by being fast and offering tabbed browsing when IE didn't. I think being fast and offering radical quality of life feature is a good strategy to follow even now.

> tabbed browsing

And popup blocking. I remember how much of the Internet was utterly infested with popups in the early-to-mid-00s. Firefox was the only way to not get inundated, unless you were a really hardcore geek and ran a proxy server like Junkbuster (the ancestor of Privoxy), and most of the population wasn't going to do that.

Popup blocking changed how the Internet works, and you can thank Firefox and Opera for that.

Firefox became popular by copying Opera to the extent that right now one of the reasons I use it, is because it feels more like Opera 12 than the modern Opera ever could.

Back in 2014 I was using g Firefox, for precisely this reason. I am still using it today, for the same reason.

Please stop creating web applications that only work with Chrome (or Chrome and Safari).

not only that but stop testing your frontend code only on chrome. i'm the only one on my team that uses a non-chrome browser and every once in a while i find a bug or a css issue that is non-chrome only because nobody tests outside chrome.

Unfortunately the choice isn't as easy now, personally. I'm stuck having to use an older, non-Quantum version for my daily browsing following Mozilla's decision to obsolete their previous addons as of v57 (and having since deleted them from their servers).

Previously Firefox was a no-brainer choice for its deep customizibility and functionality it afforded addons, which primarily distinguished it from Chromium, however since the move to Quantum I have too many addons without an equivalent to switch, with a couple favorites that did migrate to WebExtensions offering an inferior experience.

Vivaldi would be my second choice should I be forced to switch since its integrated features more closely match Opera's original features, albeit less extensive.

Mozilla felt the need to do what they did in that regard and I understand the various reasoning for the decision however the way it panned out didn't make for a browser I'm currently interested in upgrading to, sadly.

Okay, I'll switch to Firefox now and give this a try.

Just fyi this article is from 2014 and it was mentioned here in HN several times. But yes you are not too late :)

The future is here for at least one K-12 school district that I work with. The majority of the organizations computing devices are iPad, Chromebook or Windows S. None of those devices can run Firefox.

Recentish Chromebooks can trivially run Firefox Mobile.

I actually spent last night getting a full-blown desktop FF running on my chromebook (by way of Crostini, their Linux virtualization). It's not something my mom or even my wife would ever be able to do, but it is possible. And if Mozilla really cared, they could package it in a way that would make it a whole lot easier to install.

I hadn't considered the android version of Firefox on a CB, good point

Seeing as browsers on iOS can’t have their own rendering engine, no, you can’t run Firefox on iPad in any meaningful way.

Ah gotcha. I'm not super familiar with iOS, and was vaguely aware of such limitations but not that specific one. BTW, I didn't intend my question as disagreement; it was a sincere question.

It’s great to see people swapping to Firefox from chrome, but if you really want to support them, consider donating to Mozilla. As others have said, there is a conflict of interest as they getting funding from google. I donate £30/$40 a year to Mozilla for this reason, and put my money where my mouth is.

How will avid Apple users on this website choose Firefox as a browser on their iPhones and iPad computers?

EDIT: To clarify, mean the rendering engine (so we keep the ecosystem diversity), not the reskinned Safari. Reskinning Safari with Firefox branding does nothing for WebKit web monoculture.

We can, just not the Firefox rendering engine. As frustrating as that is, perhaps if more iOS users picked firefox as their "browser" then maybe Apple would take a hint and open up allowing other rendering engines. Although I'm not holding my breath.

I'm an Apple user in as far as I have an iPad Pro and a Macbook Pro, I certainly don't hold Apple in as high regard as I used to do but those two products are superb even with the limitations of the iPad/iOS.

> As frustrating as that is, perhaps if more iOS users picked firefox as their "browser" then maybe Apple would take a hint and open up allowing other rendering engines. Although I'm not holding my breath.

The problem with this is that allowing other rendering engines would open up a whole can of worms that Apple would rather not have to deal with, and there's no easy way, even if Apple wanted it, to allow JIT execution for third-party browser engines without utterly destroying the entire codesigning system.

What design choices mean third party Android browsers can do JIT but third party iPhone ones can't?

Android doesn't do codesigning, at least in the way that iOS does: on iOS, all code that executes must be signed and approved by Apple before it is allowed to be loaded by the OS into memory (with some exceptions e.g. Safari). JIT gets around this because applications can now generate and execute arbitrary code, which is obviously something that Apple doesn't want. So this basically means that you are not allowed to put apps like this on the App Store.

It is a result of Apple's broken security model. On Android, generated code has no more provoleges than the code that generated it, just like interpreted code.

> On Android, generated code has no more provoleges than the code that generated it, just like interpreted code.

This is true on iOS as well, just it is on almost every other operating system: you don't get to expand your privileges by being able to generate code because said code still runs in your process. There's no "broken security model" here; in a sense iOS has a much stronger security model than Android does because all code that gets executed can be statically verified beforehand.

> in a sense iOS has a much stronger security model than Android does because all code that gets executed can be statically verified beforehand.

The code generator can be statically analyzed in the same way that an interpreter can. The security model is broken in the sense that it applies a wholly unnecessary restriction on apps, which means not only can you not have your own fast JavaScript engine on iOS but you can't have fast emulators or run other languages fast on the device either.

> The code generator can be statically analyzed in the same way that an interpreter can.

Unless the state of static analysis has somehow changed significantly without me realizing it, no, this does not help at all. While I can (to some extent) verify the code in the code generator itself, I cannot verify that the code generator will not generate arbitrary code unless it lists out all the code it can possibly generate, in which case it's, well, an interpreter. The task of verifying program behavior is now shifted from static analysis on a compiled binary to dynamic instrumentation of a running program, which is much harder to do.

It is trivial to prove that anything that a code generator can emit can be translated into something that an interpreter can interpret. Run the generated code through an interpreter. QED.

> It is trivial to prove that anything that a code generator can emit can be translated into something that an interpreter can interpret.

That's not what we are trying to prove, though; we're trying to show that the generated code does not perform certain undesirable operations. this is much harder to do with dynamically generated code than static code.

Not using the Firefox rendering engine means you're effectively using Safari for rendering sites and not helping ecosystem diversity. The renderer is still WebKit controlled by Apple, not a Firefox browser controlled and built by Mozilla.

I don't have an iOS device, but presumably Firefox on iOS still integrates with Firefox Sync, so it may reinforce using 'real' Firefox on a laptop or desktop.

It does, although doesn't support Firefox extensions as the Android version does. Again due to browser engine restrictions.

Damn you, Apple.

Actually, since Firefox uses WKWebView, it should be able to support extensions to some extent. Whether this would fly on the App Store, with the rules against alternative marketplaces for software and all, is indeterminate, but I don't think there is a technical reason why this couldn't work.

That is definitely a fair point, and that's the frustating bit for me. I had hoped that eventually Apple would bow to pressure to open up allowing other rendering engines but as I say I'm not holding my breath.

They haven't done this in the last 10 years of iOS/iPhoneOS, can't imagine why they'd change their mind now. It is a shame.

>Apple would take a hint

never gonna happen

~10 years ago, Apple killed the original Firefox for iOS, which was using Gecko. I wouldn't hold my breath.

I'm using Firefox on my iPhone too, even if it's just a reskinned Safari.

It has a more convenient UI, plus I get shared history and bookmarks with my desktop. So basically I use it for Firefox Sync.

They can use Firefox, just not the rendering engine. I expect that would still be considered an adequate move. Nobody I know (bias) has an iOS device and no laptop or PC, so they can also adopt it on their PC.

How is using reskinned Safari with WebKit rendering engine considered adequate move?

That would be like putting a Firefox skin over IE6 engine a few years back - it would not help the ecosystem in the way this article asks for.

It would still use any bundled Firefox services. Mozilla makes (most?) of their money from search engine referrals so this alone is likely to be a direct support of them.

They can run a WASM-compiled version of it.

Does Wasm run comfortably on iOS devices? If so, that might be an idea.


(but yes, I know it's still using webkit behind the scenes)

I don't believe one really can. That's a good argument against Apple's walled garden IMHO.

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact