Just got this notice from my bank: No more Firefox | Hacker News Reader

Just got this notice from my bank: No more Firefox

old.reddit.com

133 pointsvord108074 comments

This is pretty lame BS. I am literally right in the middle of building a public web app (also for a bank) that does not have any sort of arbitrary user agent constraints.

As long as your browser supports reasonably-modern things like flexbox and media queries, my app is just going to work on your machine. Even if it doesn't support these things, I don't see any reason to artificially get in the way of something you might otherwise be able to string along regardless.

I roll 100% vanilla JS/CSS these days so that I can avoid getting entrapped in the kinds of hells that likely entrapped these mbna web developers. It is quite possible they use some JS framework that has a tight dependency on a certain cohort of browser vendors & versions.

Note: MBNA Canada is owned by TD.com (one of the 5 major banks in Canada) and not the US MBNA. In my estimation, it's likely that they've looked at the numbers and are essentially 'firing' FF customers b/c they don't drive enough rev.

CIBC (another bank here) has been doing this for at least 15 years by offering seemingly 'cheaper' banking services using the same systems but different brands (eg PC financial). The downmarket brand seems cheaper on the surface but they take away all the costly things the flagship brand gives away (eg in-person banking, no fee transactions, etc) and charge [more] for them.

Our telcos here do the same thing with MVNOs (eg Rogers, the flagship brand, owns Chatr, for poor ppl).

Anecdotally, if you use firefox on android to sign in to a wells fargo account you will get a shitload more captchas and extra verification steps to actually login than if you use chrome, or their officially published app.

This is not how the web was supposed to be.

I've gotten a ton of captchas when using Brave. One time I was trying to create an account and I just got an unending stream of them. I literally gave up after 15 or so. If it had said I was going to have to do 20, I might have continued. But there was on light at the end of the tunnel, so I gave up.

Sorry, but you aren't human, we have proof thanks to our extensive surveillance network, please renounce your citizenship and human rights today.

You're supposed to use Chrome while logged into a Google account or be sent to captcha hell.

Serious question: I love FF because of many reasons, but "additional voice at the web standard group" is quite far down the list for me. The top reasons I'm using FF right now are the features, namely high customizability on everything, TreeStyleTab, Picture in Picture mode with subtitle support, container tabs, all the privacy stuffs, and so on. How hard is it to have all these features with another engine like Webkit or Blink? Would there be a chance of FF keeping these features but switch out the engine it uses so that it takes less engineering effort to maintain?

Firstly Mozilla is never going to switch to Chromium engine because it'd indicate giving up most of what makes it unique. Second to redo all their unique features would take a number of years, particularly all the privacy aspects that the Tor Browser relies on. Compare it to Brave for example, they've had years to work on their browser and it's not got any of the aforementioned Firefox features and a fraction of the privacy ones.

Brave has even better privacy and anti-fingerprint measures than Firefox

https://privacytests.org/

Brave has more privacy feature out of the box. LibreWolf which is Firefox with changed settings is superior. besides the site only shows if a privacy feature is present not how good it is.

LibreWolf is more slower than even Firefox. Getting 10% better privacy for half the performance and possible incompatibilities on the web is not worth it.

no, also just disable JS and all sites will run faster than they ever could

Please correct me if I am wrong, have not followed LibreWolf for a while. I did try it out for a bit. I believe they are just implementing some facets of the custom user.js [1] and custom policy files and then changing where cache files are stored.

I was not a fan of their cache location changes as I had to write custom rules in bleachbit [2] to vacuum/compress/clean database files created by LibreWolf. In a weird way I think they made their browser less private with that move as not everyone is going to write custom bleachbit rules.

[1] - https://github.com/arkenfox/user.js

[2] - https://www.bleachbit.org/features

it has changed user.js with build in ublockO and removed bloat like pocket.

there is no point in manually deleting cache since it's wiped when closed (you can also change that in Firefox)

it is better Firefox

there is no point in manually deleting cache since it's wiped when closed

I used to think this as well but bleachbit shows me every time I close my browser there is much to be cleaned. I launch firefox using firejail and then when firejail exits it calls bleachbit. Even if I use "History" -> "Clear recent history" and select "Everything" there is still a lot of crud to remove. Try it out if you have some spare time.

There's even a few things that bleachbit misses by default such as moz_historyvisits in places.sqlite and files in the datareporting directory that get created even if ff is told not to create them.

those are cache from the browser itself = not relevant for websites. just checked myself and they are always the same 3 files for every user with the same hash.

bugzilla.Mozilla.org/show_bug.cgi?id=1489817

Tor browser uses the same system btw

In today's webbrowser landscape Safari is the one lagging behind.

Firefox is doing very well in regards to implementing web standards while Safari doesn't implement enough to be called up to date or modern.

This is not really true, for what it's worth. The latest releases of all browsers are fairly comparable; Chrome is a bit ahead, and Safari and Firefox mostly trade off against each other: https://caniuse.com/?compare=chrome+103,safari+15.5,firefox+...

Some of those are pretty major, while others are nice to have, and some others are privacy nightmares and Firefox not implementing them is deliberate, like the battery and gyroscope APIs.

Safari doesn't support CSS subgrids, AV1, Push API, HTTP/3, and regex lookbehind.

Except for Push API and video codecs, I see those as major flaws holding back the web.

Firefox can do better at some CSS functionality, with that I can agree.

My bank(name and shame, yeah it's chase) does not work with openbsd. unless I change the user agent then it works just fine. sigh

Some days(usually after a few drinks) I set my user agent to a simple "firefox" and every time I do it is depressing how many things break.

I wonder if people realize how low Firefox marketshare has fallen, or if people are stuck in a 2010s-era idea of what the web browser landscape is like. I also think the prevalence of Firefox among the (overall tiny) desktop Linux demographic tends to warp developer perspectives.

It's under 4% - for all intents and purposes a dead browser. I certainly wouldn't devote any engineering hours towards supporting it.

Not to mention, the core development team has been cut.

Mozilla is rebranding itself into a “Social Justice For Tech” marketing firm.

There are two significant platforms left Chromium and WebKit. I hope that isn’t the future, but browser dominance is such an obscure, indirect problem to the industry that no company would bother touching it.

Not sure where you got the idea that the core development team has been cut, Firefox development remains strong. For example, here's a 1,000+ line patch that just landed two days ago. https://bugzilla.mozilla.org/show_bug.cgi?id=1779952

I shouldn’t of said cut, but they did layoff 250 members 2 years ago and among those were the Servo team.

https://www.protocol.com/amp/mozilla-layoffs-2646950077

That's not the core dev team, at all. Servo was a research group, building an experimental browser engine, parts of which did end up transitioning across to Gecko.

Some senior core developers were definitively let go, though, e.g. David Baron (who is now working for Google on Chrome instead).

Look at the tech Mozilla produces, all of it is non portable, exclusively for FF, and is not particularly great.

Chromium can be extended in many ways. V8 can be used anywhere and is what spawned node and the idea of JavaScript server and client side.

Firefox is more like a product that's source available than an open source project made to benefit the ecosystem.

They would do better to finally pony up, reskin chromium, and stop playing catch up.

They really should rebadge Chromium, then they can frolic with Rust UI changes and focus more on what they really seem to care about.

In most industries increasing your reach by 3% is worth doing.

You are not losing all that 4%, most will just launch another browser to for your site if they have to

Or user agent spoof, then blame the bank if it blows up.

(This is what I do. No problems so far…)

Firefox's marketshare is hardly the point here. As bob1029 said, there's no earthly reason why you wouldn't just build a standards-based site and have it work on any modern browser.

It's a bank. If a customer uses Firefox and loses money because of a Firefox bug or incompatibility, who do you think will be liable? The customer? No way. The bank is on the hook because it's their website that resulted in the loss.

So the bank has to test in Firefox, they can't just put all their hope in web standards that may or may not represent reality.

When Firefox falls to a certain marketshare it makes no sense for the bank to keep testing on Firefox.

"It makes no sense to appeal to 4% of the population" seems like a bizarre claim, especially since that 4% is likely to be programmers in the top 25% of income.

I'm also not sure what error you're expecting? If browser bugs are a serious risk like this, you should surely be able to cite a few example lawsuits from the last year or two?

This argument seems bizarre to me considering the volumes of money banks work with and the lack of programmers who only use banks that take Firefox. I'm not sure what you're expecting, that banks are just doing it to spite Mozilla?

I really don't believe the return-on-investment for a major bank to support Firefox is negative. Quick Google napkin math says a customer is worth $500/year, and a decent programmer is easily available for $50/hour. You only need to stop one (1) customer from leaving for every day of programming this costs. This thread is at +89 and has 32 comments, so it's probably worth at least a programmer-month.

If you can't make your website Firefox compliant in a month, you have some very deep design issues. Firefox compliance is so trivial that most websites achieve it without even trying, because Firefox follows basically the same standards Chrome does.

Yep, but the people making the 'technical' decision to support Firefox only see the gains from reducing the labor.

The loss from not supporting Firefox (which I agree, is probably more than the savings) will show up on somebody else's balance sheet. IT saves a bit of money by siphoning indirectly out of Customer Retention, and the details get lost in the noise. It's the curse of large organizations.

4% of desktop web traffic is Firefox is not at all the same thing as “4% of desktop users also don’t have access to or refuse to use a Chromium based browser and will switch banks because it”.

I really can't think of how you can lose money because of a browser bug unless you put so much intentional complexity in what is essentially a CRUD app that a browser bug could trigger disastrous failure, in which case the fault is on the developer for overengineering rather than any browser bug.

Yeah, this seems like a red herring at best.

I understand an argument against continuing to test, even though I find it insufficient, but liability doesn’t make sense here.

> If a customer uses Firefox and loses money because of a Firefox bug or incompatibility, who do you think will be liable? The customer? No way. The bank is on the hook because it's their website that resulted in the loss.

> > So the bank has to test in Firefox, they can't just put all their hope in web standards that may or may not represent reality.

Thanks to the open nature of the web, i can modify the request/response, JS that executes, etc right in the browser. So for your argument to be valid, the bank would literally have to fully trust client side directives and do next to zero validation of the users input. Even banks aren’t that stupid…

And if you know of a bank doing that, please share! I’d love to print myself some free money!

Security issues somewhat withstanding here, but even that argument is bullshit because Chromium/Chrome has just as many bad security bugs as FF, sometimes even more and/or worse.

So no, this is bullshit. The reason they’re doing this is so they can reduce development costs while more likely than not engaging in surreptitious activity surveillance, given FF has some of the strongest protections against that crap that have ever existed. Coupled with recent-ish reports that credit bureaus want to let your browsing data impact your credit score, they’ve already got several big fish on the line willing to buy. And with little to no isolation for client side storage per domain in most of those browsers, i can see no other incentive for them to do this.

not at the bank i work for. they have some sites that only work in chrome, some that only work in edge, some that only work in internet explorer, and absolutely nobody cares except for some pedantic nerds like me. fixing the problems would cost more than any potential errors they might have.

In my country it is the same or worse. Not only banks but the public institutions in general, force you to use outdated browsers (IE, old Firefox) and Windows, coupled with installing sketchy plugins and unsigned .exes transmitted over unencrypted HTTP

Seems like you forgot what it was like when Internet Explorer was the dominating browser.

Sounds like they're actively working towards recreating that world.

I really have no sympathy for a $100/h web developer that has no time to check with Firefox every once in a while that their Chromeism works over there as well. It's not like Firefox is a buggy browser that requires tons of work.

It's pure laziness.

Falling market share in a growing market doesn't necessarily mean falling total usage.

Additionally, the market FF serves this days is educated, rich and ahead of the curve. Not a market you necessarily want to ignore.

> the market FF serves this days is educated, rich and ahead of the curve.

That’s some weird projection that made me laugh. I used to use firefox everyday until it became too annoying and I moved on to another browser and never looked back. Firefox used to be great but those days are long gone.

I don't think there's anything I'd exchange for having to suffer more ads, but you do you.

Generally demographic makeup of FF traffic is more affluent than Chrome, similar to Safari. IE is closer to Chrome.

It can vary by the field or nature of the site, of course.

By this logic, no one should bother supporting Linux.

As a 30y FF user I wish Mozilla would focus on innovation and building the very best browser to get back market share.

Not possible. Google owns the network and wins hands down for being able to advertise Chrome wherever and whenever they want, while Mozilla is forced to buy ads. Being better than Chrome is not enough anymore.

My bank (B of A) started complaining about Debian's firefox-esr being too old and out of date, so I loaded the bleeding edge version just to access my bank account, but use the esr version for general browsing.

I for one am glad to run a version that Eric S. Raymond has officially blessed.

Hopefully you're joking... :-), if not, it means Extended Support Release.

(The first time I saw it, I thought the same)

Joking or not, enjoy your upvote for givin' me a good giggle when I sorely needed it. :)

What they really mean: "We're no longer developing for/testing Firefox, and if you run into a technical issue we will not offer any support". Firefox will probably continue to work fine.

This is likely in response to some internal product/development team being told to go faster.

I’m guessing that their feedback was that cross-platform UI testing is occupying a significant chunk of their time.

Leading to some MBA type thinking they could cut the testing time in half if they eliminate browsers that don’t consist of 99% of the user population.

It's Internet Explorer all over again.

Wonder what will be the next javascript of today ?

javascript, just like it was before today.

wasm

Few weeks ago my bank force me to use password and SMS code for transactions because of EU regulations. But if I would use app (which I won't), I don't have to use any passwords.

Which EU regulation tell banks to use SMS and password to make transactions without app?

Strong Customer Authentication (SCA), "a new set of rules that will change how you confirm your identity when making purchases online".

A Google search will surface help pages from most European financial regulators and payment processors and banks. For instance, [1-3].

The rollout was delayed a couple of times so cutoff dates mentioned on older pages may well have shifted. In the UK, the cutoff date was 14 March 2022.

[1] https://www.fca.org.uk/firms/strong-customer-authentication

[2] https://www.ukfinance.org.uk/our-expertise/cards/card-paymen...

[3] https://www.visa.co.uk/partner-with-us/payment-technology/st...

My bank offered SMS code authentication as alternative before, but since new EU regulations came out 1-2 years ago, Im forced to either use a physical card-reader OTP generator (which, to be fair, is free, but clunky, so I can't easily put it in my pocket) or use their app that doesn't even work on my device (i use a hardened custom android distro, but have no root and bootloader is locked)

i stopped using firefox as it was not fixing some performance issues. also there are plugins which on chrome, but not ff. and also they never made bookmarks feature complete. i have donated not less than 100 usd to mozilla in total i guess. part because of rust. and brave is more innovative in p2p and web3 area. it has tor. so brave was my final reason to switch.

Chromium-based engine : Chromium, Chrome, Brave, Opera, Vivaldi, Edge, Naver Whale.

Firefox-based engine : Firefox, GNU Icecat.

Firefox is slowly becoming the pariah of web browser.

I'm building a web app, and tested the app against all those browser. The behavior is really different, especially regarding the implementation a newly developed web api and the depreciation plan of ancient javascript api. For example, xhr depreciation [1].

As for chromium, they are quite responsive in responding the developer inquiries regarding a browser-specific API implementation. [2][3]

Sadly, it's IE vs Netscape all over again.

[1] : https://xhr.spec.whatwg.org/#the-open()-method

[2] :https://groups.google.com/a/chromium.org/g/chromium-discuss

[3] : https://www.researchgate.net/publication/353527090_Rancang_B... (it contains my communication excerpts to the Chromium team, regarding xhr deprecation plan)

The yellow warning box on [1] is referring specifically to one flag, not to XHR generally.

> Firefox-based engine : Firefox, GNU Icecat.

Plus many others: https://en.wikipedia.org/wiki/Category:Web_browsers_based_on...

Chrome is horrible, it is so unstable and crashes all the time.

Classic Canadian mediocrity spotted again.

In the time it took the OP to screenshot this and post it across several social media channels, they could’ve closed their account and joined a bank without such BS restrictions. But gotta get those upvotes.

And of course I realize the OP could have done both, but I am trying to balance rolling on coke with drinking whiskey so please accept my acknowledgment

I always welcome public shaming of such companies.

Reply on news.ycombinator.com