Hacker News new | past | comments | ask | show | jobs | submit login
Sponsored Top Sites (support.mozilla.org)
115 points by mmanfrin on Feb 24, 2021 | hide | past | favorite | 114 comments



They've now added these to the search bar, above everything else.

If you already have top sites disabled on your start page, then you have to re-enable this and then disable the checkbox, before re-disabling it all. Terrible UX, arguably a dark pattern and user hostile.

https://twitter.com/OhMySMTP/status/1364274675459764231


...and that's why I use ESR. I get all the security patches but I only have to deal with browser drama every year rather than every month.


Sorry could you expand the acronym ESR? I’m unfamiliar and would like to look into it, thanks



"Extended Support Release (ESR): receives major updates on average every 42 weeks with minor updates such as crash fixes, security fixes and policy updates as needed, but at least every four weeks."


Mozilla: “When you click on a sponsored tile, Firefox sends anonymized technical data to our partner through a Mozilla-owned proxy service. This data does not include any personally identifying information.”

adMarketplace: “We may also receive technical information such as your approximate location, browser type, language settings, user agent, timestamp, cookie ID and IP addresses.”

There's a big discrepancy between Mozilla's statement and adMarketplace's privacy policy. Although there's very little information available, the fact that this technical information is anonymous doesn't appear straightforward. Cookie IDs and IP addresses alone can very easily identify some users uniquely across time and even devices.


> Firefox sends anonymized technical data

Can't we intercept this data and see what it is?


Fair enough, Mozilla desperately needs to develop alternative income streams, their dependence on their biggest competitor seems to be an existential risk


> Revenue (per year)

>2014 $420 million

>2015 $420 million

>2016 $520 million

>2017 $562 million

>2018 $436 million

I'm sorry, it's just laughable that you need $500 million a year to create an 'open source' web browser.

Pathetic


You're leaving out the revenue from the Yahoo lawsuit. When you combine the revenue from actual royalties and the award from the lawsuit, Mozilla's income for the last year we have on record was just under 1 billion USD, and people continue pushing the narrative that Mozilla is more financially insecure than ever. Firefox is certainly at one of the lowest measures of success it has ever seen, but it's also bringing in more money per user than ever.

The picture of Mozilla as frail organization just barely squeaking by has been very lucrative.

The fact is, Mozilla has been in the current position for the last 10 years, which is one that is far less precarious than it's made out to be, hasn't done anything that substantially ameliorates what precarity does exist, and instead continues making awful decisions that push away both people in its user base and its potential set of contributors, with the kicker being that the purported looming threat of poverty is used as the justification for all decisions made. And they never budge on anything, except to move further down on all measures that they point to. So how smart and well-reasoned are those decisions again?

Signed, a lifetime Firefox user (for lack of a better option).


So what do they use the money for if not actual software engineering? I have a feeling if people start digging they'll uncover a situation similar to Wikipedia.


The problem is cost disease married to managerial incompetence, same as what's afflicting universities and hospitals. They're using the money for software engineering, but they're doing it poorly, including doing a really, really shit job handling unpaid contributors (essentially driving everyone away and thus requiring them to hire others to fill the shoes of the people who stop contributing).

They also spend tens of millions of dollars a year on their marketing department -- a bigger budget than Firefox's entire development budget 15 years ago (i.e., when Firefox was actually on an upward trajectory). There's certainly money going into actual marketing, but the returns on that investment are terrible to the point that they'd probably be just as well off cutting the budget in two, keeping one half of it to be allocated for some other purpose, flushing the other half directly down the toilet, and doing no marketing at all (or maybe walking out into the street, handing it to random passersby, and saying, "this is from Mozilla; you should really use Firefox").


How about 500 mil yearly to create an OS from scratch? Is that laughable too?

The level of complexity to create a modern web browser, specially one that has in-house engines like Firefox, is comparable to a modern OS.


> How about 500 mil yearly to create an OS from scratch?

When was Mozilla trying to create an OS from scratch?

The closest they came was FirefoxOS which was using Gecko browser engine (existing) on top of Linux kernel (existing) and a host of other existing open source projects, and FirefoxOS was (unfortunately) a dismal failure.


Maybe I wasn't clear enough.

The level of complexity of a modern web browser is comparable to that of a OS. It is not an OS though.


That complexity is mostly due to Gecko, right? What's stopping Mozilla from switching to Blink? Does using Google's engine conflict with their mission?

Edit: Getting downvotes for this honest question. I didn't realize that browser rendering engines held that much control. Apologies for the silly thought.


The whole point of Firefox existing is too be an alternative to Blink.

This is like asking why Pepsi doesn't just copy Coke's recipe.


This doesn't make any sense. Firefox is an alternative to Chrome, Edge, and Safari. Gecko is an alternative to Blink, Tident, and WebKit.

It's cool to have multiple implementations but this analogy doesn't work since browser rendering engines are trying to produce identical results. You wouldn't say that PyPy exists as an alternative to Python, would you?


> You wouldn't say that PyPy exists as an alternative to Python, would you?

I think you would if Python (proper) was controlled by an anti-competitive information stealing superpower which is becoming increasingly disliked.

Alternative backends to Python would be very important. Ensuring that the language is still usable in the event that the primary contributor goes full-tilt evil.

With that said, i don't think it matters (personally) if Mozilla has their own engine _or_ browser, in the most strict sense. What matters is that both of those things have invested parties capable of forking and/or maintaining free and open source versions of those things in the event that Chrome (ala Google) goes full evil.

It does seem that maintaining your own browser and rendering engine seem effective ways to ensuring you're capable of carrying the torch should Chrome disappear tomorrow. Hypothetically they could just maintain/develop directly into Blink and Chromium.


This sounds very much like a distinction without a difference.

The purpose of Firefox doesn't exist without Gecko, they both are to serve as an "independent" alternative force in the browser space.


that would be a monoculture disaster. monocultures are not resilient.


Not resilient to disease maybe but I'm sure webdevs will be overjoyed at having a much easier time porting sites. The power dynamic that matters is political not technical. Google doesn't really get to exert control over the web with their rendering engine or V8 it's the actual high-level browser policy that matters and Firefox doesn't actually have to follow any of it.

Firefox gutted all of their internals with project Servo. If they had instead integrated Blink would it have been any different?

I always think it's interesting what things people want to have diverse ecosystems with competing implementations (browsers) and what things people consider that an unnecessary duplication of effort (Wayland compositors).


Google routinely implements, fails to implement, and fails to deprecate features that align with their company goals (i.e. namely collecting lots of data). AMP, Browser Extension Manifest v3, cookie privacy features, ad-blocking features, file formats, and many more that most people have never heard of.

Happens all the time. Without Webkit and Gecko, the web would simply be "whatever Google wants to do". Having multiple browser implementations helps balance out whatever Google's agenda is at any given time.


Yes which is why Chrome dominance is a problem. This isn’t true for V8 and Blink any more than openssl is a tool for controlling the direction of browsers.

The reason you have multiple implementations of JS and rendering/layout engines is because they provide a specific technical benefit, not a political one. The browser built on top doesn’t have to be anything like Chrome.


>Google doesn't really get to exert control over the web with their rendering engine or V8

Just because they're failing at it doesn't mean they're not trying (AMP, needless degradation of Google services in non-Chrome browsers, etc). They only have to succeed once.


We're not disagreeing. Google is absolutely trying to exert control over the web but Blink and V8 aren't tools that can realistically help them with that. Especially in the hands of other teams that are free to carry patches.

If Firefox became a Chrome skin then sure, v bad for the web. But if Firefox internally used Bink and V8 it's not really all that doom and gloom.


So you've seen the same person say both of that?


We're on the internet. There's only one other human and the rest are bots.

If you go into threads about Wayland the overwhelming public opinion is that having lots of compositors is silly. And then you go into threads about browsers the overwhelming public opinion is that having multiple browser implementations is sacred.


Plenty of monocultures exist in the real world and we almost already have a Chrome/WebKit based monoculture. Where's the actual, non-theoretical disaster?

Safari is the only browser that has any significant uptake (~20%) and Chromium was forked from WebKit, so all Chromium browsers already share a good chunk of their foundation with WebKit. You could say that we really almost have a WebKit monoculture since Firefox, the only non-Chromium/WebKit browser only has 3.65% of the market. [0]

Heck, the vast majority of Internet servers and most of the clients are running on the Linux kernel but nobody is complaining about that.

[0] https://gs.statcounter.com/browser-market-share


Why stop there? Why doesn't Mozilla switch to using Chrome? :)


Car manufacturers use parts (like engines) from other competitors all the time. I thought this could be similar with Mozilla, but I guess it was a silly thought.


To follow that analogy, it'd be like BMW using an engine from Ford. Their whole selling point is the engines, and how they are different from other manufacturers.

I'm sure there are many backend libraries, parsers, etc. that are used by both Chrome and Firefox, but they're unlikely to replace significant components like that.


And to keep going...car manufacturers do use engines from other brands, but that's when they are owned by the same company. For example, you'll find Ford engines in Mazdas because Ford owned Mazda (or at least a chunk of it):

https://en.wikipedia.org/wiki/Mazda#Partnership_with_Ford_Mo...


I own a Saab with a Ford engine! It's actually a V-4, two on each side, and it was most widely used to power skidsteers and industrial water pumps and the like. Saab didn't have the engineering capacity to design and launch a engine fast enough to keep up with environmental regulations (having used a two stroke engines up until the 60's!)


The problem is that having most web browser options running on a rendering engine controlled by Google means that the web will simply end up looking and working like how Google wants it too.


It'd be really hard to justify half billion dollars a year to just re-skin Chromium.


Right. They wouldn't need that much money. That's my point.


If every browser uses the same engine, then what's the point of web standards? We can replace HTML with binary blobs and make the web so fast!


How much do you think the Debian non-profit makes per year? What about ArchLinux?

How many full time developers are paid by the Linux Foundation?

It's funny, their budget is much smaller than firefox

https://projects.propublica.org/nonprofits/organizations/460...

Maybe all this complexity and constant changes to the web comes from the fact that there are hundreds and hundreds of employees paid to tinker around.

Mozilla is effectively a $500 Million dollar corporation, not a non-profit.

Money corrupts absolutely.


> How much do you think the Debian non-profit makes per year? What about ArchLinux?

These are distributions. Their main role is to package upstream software. That is different from actually writing an entire OS.

You should compare to however much Microsoft pays all of the developers that work on the different components of Windows.

> How many full time developers are paid by the Linux Foundation? > > It's funny, their budget is much smaller than firefox

That's because the vast majority of Linux development is not done by employees of the Linux Foundation. Only a few top-level maintainers, including Linus Torvalds and Greg K-H, are employed by the Linux Foundation.

Most Linux code these days is written by people employed by companies like Intel, Huawei, Google, Red Hat, NXP, Facebook, etc. See https://lwn.net/Articles/839772/ for example.

You would do good by reading a bit before saying things online.


Exactly. Sounds like a certain outsourced government website...

If Mozilla was smart and didn't spend coin for useless stuff (Pocket?), they'd be able to set up a trust fund with all this money, that would sustain them indefinitely.


They aren't using that for the browser, they are using it to push things like women in code. Which is a dark pattern since if people wanted to fund women in code they would give the money to that charity instead of giving it to a browser and have it diverted.

But given what we're reading here it's no surprise that the corpse of Mozilla uses dark patters at every possible level to obfuscate what they are trying to do.


Well now they have 8% of the browser market. I think you need $500 million a year to barely keep at those 8%.

If they drop those millions of revenue I believe their market share will sharply drop. Opera is operating on around $100 million of revenue and they are in 1% range of market share.

I bet Microsoft can spend $100 million or less just to be around 7% with Edge since they have Windows. I also don't think Google is spending much more to keep Chrome at around 70% when they own search and YouTube and Android.

They need $500 million just to keep head above the water and remember they are swimming with sharks.


Right, it's a corporation worried about market share, not some open source project that random hackers contribute to.


While it's still a lot, software development is only about half of the Mozilla Foundation's expensive (and not all of that is the browser). They spend around $50 million/year on marketing, ~$75 million/year on administration, etc. They publish annual audits and reports (which ironically probably eats up a million bucks on its own).


This made me smile :-)


I think the lack of pressure on Mozilla has prevented them from innovating at the level necessary to sustain themselves if Google pulled the rug from under them. It might be good for them if they put 50% or so of their revenue in an endowment for sustaining Mozilla and forced themselves to exist on a fixed budget.

There are excellent engineers working at Mozilla - the fact that there hasn't been a substantial new product from Mozilla in the past decade is really a shame.


Unfortunately it seems like I was one of "small percentage of Firefox users" this feature is being tested on.

Contrary to what the article claims, there was no way to disable this feature from the menu: it wasn't possible to uncheck the 'Sponsored Top Sites' button.

I had search for 'sponsor' settings in 'about:config' to disable it. Sadly, I don't think most users will be able to figure this out.

I'm seriously reconsidering moving away from Firefox, or at the very least no longer recommending it to non-technical users


I had Yandex as a sponsored website when I installed Firefox. The process of removal did seem unnecessarily obfuscated, but I only had to spend like five minutes on it.


I found the setting by following FF's instructions and turned it off. Took about 5 seconds, I guess a minute total including searching for how to do it.


Do you know if your data was shared with advertisers before you were able to turn it off?


Of course I know that. What a strange question.


GP is not literally asking if you know. They are asking you if your data was shared or not.


Well, that makes it an even stranger question. Why ask one thing if what you want to know is something else?


I'm sorry my question wasn't clear, but I was not trying to ask one thing and know something else. I will try and restate my question.

Were you given an opportunity to disable the sponsored top sites before they were first loaded and your data sent to advertisers?


That's ok but I don't know if I can help much here - I really have no idea of knowing when I was given that opportunity.


Yeah, let's block ads on websites and put them in the browser UI instead. Only ours of course. - someone at mozilla corp


Firefox does not "block ads on websites", unless you install a third-party ad blocker.


The company that has a major browser, a major mobile operating system, and a few major websites is scribbling notes.


I really hope the Mozilla team sees this thread.

I'm not opted-in, but if this goes forward to all users, you better believe I'm withdrawing my annual donation to the foundation, and I'm switching to a alternative.

This is not acceptable.


That "anonymized technical data" link points to a github repo with no documentation. It's not clear what data you're actually sending. What do you consider personally identifying?


That looks like a big red flag for me too. They somehow have to check for ad abuse after all.


There is a legal definition for personally identifying data, for example in GDPR


Maybe there are multiple legal definitions


We really need "corporate bullshit removed" forks of both Chrome and Firefox.

IceWeasel and Chromium were effectively Firefox and Chrome without the (surface level) corporate branding. Chromium is soon having many of its Google tie-ins severed. Debian and Mozilla resolved their trademark dispute, so IceWeasel mostly went away.

These browsers are open source, yet they both suffer an unpalatable amount of corporate control. Google can do whatever it wants, because it's the king. Mozilla is able to get away with this because they are the only real alternative for people who don't like the king.


How do you expect these projects to afford its development? Free software does not come for free.


Maybe stopping to live with the browsers that we have for a while wouldn't be the end of the world?

You're right that free software development isn't free, but these browsers already exist, and are already free, and for the most part, they work. CVEs can be found and fixed without corporate backing.

Instead, we're rushing to pour more and more stuff into the cesspool. WebUSB and WebBluetooth are not things that I particularly need or want. Part of this is Google's fault. They're currently on top, and if they unilaterally decide to add something, everyone else needs to either implement it or die. I mean, I still think they should fucking stop, but obviously they won't, and I don't have a good answer to this.


Old man yells at (internet) cloud

> for the most part, they work.

What is stopping you to just keep using an older version? If you are a developer, what is stopping you to target only the subset of features that you seem necessary?

> WebUSB and WebBluetooth are not things that I particularly need or want

I assume you are thinking about the browser on your desktop OS. What about platforms like FirefoxOS - KaiOS?


> What is stopping you to just keep using an older version?

CVEs, CA issues.

Firefox ESR kind of addresses this, but it's a band-aid. The new mis-features eventually show up in ESR as well.

I want an actual fork, where dumb features like "ads baked into the browser" are kept out permanently.

Look, I said that we should do this because I think it would be good for the community... I am well aware that maintaining a fork like this would be a ton of work. I was never trying to insinuate that it would be easy, just that corporate backing is both not required and an active hindrance.


> just that corporate backing is both not required

Sure, no one needs the backing of the corporation. But whoever gets to be the one doing the "ton of work" that you are asking for needs to be paid, do you agree?

Now, the question is: who is going to pay for that? Something more specific than "the community" would help. Are you willing to pay for it? How much? Can you get more people to commit to it? For how long?


> But whoever gets to be the one doing the "ton of work" that you are asking for needs to be paid, do you agree?

I realize that this may be a contrarian position these days, but no, I don't.

As an example, the vast majority of Debian maintainers don't get paid, yet as a community project, it has worked exceedingly well for nearly 30 years.

Yes, there are certain notable exceptions like Linus Torvalds getting paid by the Linux Foundation, but this is really not the norm. A lot of big community driven open source projects have managed to work well without serious funding for a long time.

Would money help? Sure. But no, I don't think a project like this (even a large one) is doomed without it.


Oh, yes... the decades-old I can benefit for free from that totally unrelated tech project, why can't I get this one as well? fallacy.

Yeah, Debian has volunteers and a lot of people working on it for goodwill. But a project the size of Debian is something where you can get many different contributors who do all sorts of different things and become maintainers of these different subsystems. Couple that with a little fundraising and professionals that work on it because it is something that will benefit them economically (e.g, IT companies that provide services where having a free operating system allows them to commoditize their complements[0]) and it's easy to make it a sustainable operation.

None of that applies to what you are asking. There is no business to be had on a "crippled-but-leaner" browser, so no one will have any incentive to work on it for free. And if you are not willing to offer money nor volunteer your time to do it, why do you think others will? What you are wishing is just for someone to work for you to free. It reeks of entitlement.

[0]: https://www.joelonsoftware.com/2002/06/12/strategy-letter-v/


You're putting words into my mouth.

I stated that I believe the community would be better off if this existed. I have not asked anyone to go and do this.

You asked if I thought it was possible to do this without funding. I do. That seems obvious -- it's been done before without funding, it can be done again. The fact that I think it's possible without funding says nothing about whether I think it's the ideal way for something like this to happen.

Debian is not unrelated, I specifically used them as my example because Debian maintainers forked and maintained IceWeasel during their dispute with Mozilla. They are the ones who have done this before.

---

> There is no business to be had on a "crippled-but-leaner" browser, so no one will have any incentive to work on it for free.

You and I seem to have dramatically different takes on what makes software valuable and what motivates people to work on it. I don't care if there's no business use-case for a "crippled-but-leaner" browser. There's a principled case for it -- Many people would prefer to use a web browser that doesn't have ads baked into the shell. That preference gives the leaner browser value.

The fact that some people prefer software like this is what makes that software valuable, regardless of whether anybody can make any money from it. This is a statement of inherent value, not a statement about economic viability.

And secondly,

> And if you are not willing to offer money nor volunteer your time to do it, why do you think others will?

I think that others are likely to do this based on observation. There are people who choose to work on projects like these because they personally prefer software that works a certain way, out of principle. Again, the only statement I'm making here is about the likely existence of people who are willing to do this sort of thing for free, out of principle. I am saying nothing about whether or not they deserve to be paid, and I'm not making any moral argument about who is and isn't entitled to use their work or make demands from them.


> it's been done before without funding, it can be done again.

What has been done, specifically? Do not compare the effort of IceWeasel (and the many mozilla-less forks of Firefox that came at that time: basically just keep a folder with different set of resources and some automated build with a few different variables) with the work of keeping a fork with code that needs be "frozen in time, except for backported CVE fixes". Not only it's a different type of job (the first can be mostly automated, the second requires an ongoing process and effort from highly specialized professionals) it was also required for legal reasons, so it was at best a hurdle and no actual benefit.

> I believe that the community would be better off if this existed.

Yet, you are not willing to contribute anything to make this happen. You'd be willing to reap the benefits from it, but not to actually put in any effort. Show some willingness to put skin in the game, otherwise all your "beliefs" are nothing but hot air.


> Yet, you are not willing to contribute anything to make this happen. You'd be willing to reap the benefits from it, but not to actually put in any effort. Show some willingness to put skin in the game, otherwise all your "beliefs" are nothing but hot air.

Once again, stop putting words in my mouth.

I never said I wouldn't contribute something to make this happen, but this is the second time you've insinuated as much. Enough.

I never claimed that this should happen without funding.

I stated that I believe it's possible for this to happen without funding.

None of those statements are equivalent.


> I never said I wouldn't contribute something to make this happen,

I asked if you are willing to pay, you said no. So let me ask: is there any way that you are willing to contribute? Will you volunteer? Will you review the fork? What is going to be your share of sacrifice to the "community"? Are you going to do this continuously or find a successor? It's not just "funding" that I am talking about.

Moreover: if this is so important for you and if you believe that more people share your values, why don't you lead by example and start doing this instead of waiting for others?


That is not what you asked, and for the third time, that is not what I said.

You asked: "Does the person doing the work need to be paid?"

Which I interpreted to mean: "Is this project doomed to fail if the developers are not monetarily compensated?"

I answered: "No, I don't believe this project will necessarily fail if the developers aren't paid."

I did not say that I wouldn't fund, work on, promote, organize, or otherwise support such a project.

I made no statements about whether I would or wouldn't support such a project, other than to say that "I think this would be a good idea," yet you've repeatedly brought these elements into the conversation as a bludgeon so that you can claim that either my principles are dumb, or that I'm not principled enough.


Read again:

> Are you willing to pay for it? How much? Can you get more people to commit to it? For how long?


That is not the question I explicitly quoted when I answered "no."

I deliberately did not answer this second question.


Yeah, I know. Can you answer it, then?


So that you can turn my answer into another ad hominem attack? No, I don't think I will.


What ad hominem? It's not about you or your character. I am not trying to argue that your "belief" is right or wrong. This is up to you.

What I am asking is "how strongly are you willing to stand behind your belief?" It's a simple filter to check if you will continue to hold your belief if it costs you something to defend it. It's just a test of Skin in the Game.


> Show some willingness to put skin in the game, otherwise all your "beliefs" are nothing but hot air.

This felt like an ad hominem. I was making a case for the existence of people who choose to write software for principled reasons rather than for money. How much extra money I have to personally throw at a project doesn't change this number of people, since they're (by definition) not doing it for the money.

But you seem to think that me not having enough skin in the game is justification for dismissing my argument as hot air. I don't think that justification holds.

---

I'm sorry, I'm not really mad, I'm mostly just frustrated with where this conversation has gone.

Would I support such a project? Yeah, I'd probably throw a few bucks at them, like I've occasionally done in the past with Mozilla, with other projects, and with individual software developers who wrote tools that I like. Would my meager contributions make or break the project? Obviously not. Are you going to argue that the sum total contributions from people like me will likely amount to less than one engineer's salary, and therefore the project is doomed? You can make that argument if you want -- I don't agree with it.

Look, this is the kind of thing we're talking about:

https://twitter.com/CohanRobinson/status/1364172683118866433

I think it's bullshit and don't want that in my browser. I bet you probably don't like it very much either.

The only thing I was trying to argue in response to your questions essentially boils down to "people have gotten angry and forked some pretty major projects for far less than this."

ffmpeg and libav

Debian and Devuan over systemd

...etc.

It just doesn't seem too far fetched to me to think that some person / people might feel incensed and motivated enough about this change to create a de-crap-ified Firefox fork, even if they weren't getting paid. That's all.


The point that I'm trying to make here is that even the projects ran by "people who are incensed and motivated enough about this change" need to have the upstream going and be continuously supported.

If you say "we can get rid of crappy technology" but you still need the crappy technology to be used by the absolute majority of users, are "we" really getting rid of it?

To get "rid" of the crappy technology, "we" must go beyond this piggybacking. "We" need to put resources into an alternative. "We" must ensure that are enough people working on it to satisfy the needs of the majority, not just our own. And unless you are really invested into defending this ideal to the point of sacrificing yourself (or your money/time/resources), you will have to make compromises. For Mozilla, this is coming as sponsored links. For Brave, this is coming as a business model that still deals with ad distribution (although in a model that the users also benefit from it and that are not tracked).

You can complain and wish it was different all you want, but at the end of the day the truth is that these compromises do not bother "the community" to the point where they would pay to keep an alternative business around.


> We really need "corporate bullshit removed" forks of both Chrome and Firefox.

And since Google's recent announcement, unofficial Chromium builds will be unable to sync passwords with Google's servers.


Good! I hope that maintenance of a de-googlified Chromium continues, and I hope they strip out even more of the undocumented telemetry and forgotten whitelisting of miscellaneous Google-controlled endpoints.

The result will be a better web browser.


firefox: user.js repos (eg. https://github.com/arkenfox/user.js/ or https://github.com/ghacksuserjs/ghacks-user.js/)

chrome: ungoogled-chromium


This happened to me yesterday. First thing I did after verifying that it wasn't a malware infection was Help → Submit Feedback, only to be met with:

> We've paused submissions to this form so that we can improve how we collect feedback.

Stonewalled à la Google; insult to injury. “Unfck” indeed.


It seems like you're not able to opt out until the feature has been installed and is running which is a bit frustrating. I had a similar issue recently with Heroku not allowing me to delete my account without accepting their new terms of service.


I'm starting to regret switching back to Firefox with this move. Wisdom of the crowd: what's the next decent browser?

My initial reaction was to search the Amazon tag and found reports of rogue plugins. I thought I'd been done security wise at first..


I'd recommend you a Chromium fork but I'm not brave enough to do it here.


I've never liked the built-in new tab pages in browsers. I made my own years ago never looked back. Since 2017 I’ve run my new tab page as a small SaaS.


Hey that's not a bad idea. I frequently check my website's dashboard anyway, could just add my 'top sites' there.


imo, for companies which are forced to compromise their product to monetize, the only way to maintain goodwill with your users to be candid about it. In this case looks like: "why are we doing this? Because we need the money", more or less.


They will just point at others and pretend to "unfck the internet" as a diversion while slowly making the users the product. "Hey watch The social dilemma on Netflix guuuys".


In 2019 Mozilla got 789 million dollars in revenue. That's half the estimated cost of the Linux kernel.

Do they really need more money? Do they?


I'm glad I ditched Firefox half a year ago. Safari under Big Sur is actually not that bad as a daily driver, and with the upcoming webm support it will do everything I want it to. Firefox is just not the same anymore. It used to be tech focused. Now they pretend to sell you "privacy" and virtue signal every chance they get, but it's clear Mozilla Corp is just after more money and behind your back they use every chance to monetize you. Technology doesn't even come second. In hindsight I'm glad they failed at making a successful mobile browser, it's time for them to bite the dust.


It's funny how we're all so brainwashed by advertising and corporate logo placement, we don't even see it as brainwashing any more... I hate the everpresence of these images. Is anyone maintaining a sensible autoconfig.js and firefox.cfg on GitHub?


How long until there is an Unmozilla'd Firefox distribution?


This thread seems like a good opportunity for everyone to share other browsers they're aware of which are at least somewhat usable.

Setting a very low baseline of "works for browsing Hacker News", I'll get the ball rolling. Each of these browsers is already pretty good, and could be even better with your support.

I use primarily GNU desktop and Android, others can fill in Mac, Windows, iOS, and other platforms.

DESKTOP:

Pale Moon

Opera

Vivaldi

Waterfox

qutebrowser

Links

lynx

w3m

Midori

ANDROID:

Bromite

FOSS Browser

Brave

Smooz

Ghostery

Via

These are all browsers I regularly test with and they work quite well. Heavy sites like Instagram, Reddit, Facebook, YouTube, and Gmail do not work with them, and I generally avoid those sites, resort to using alternative front-ends, or open Chromium for a minute or two.


It's worth noting that Pale Moon, Waterfox, and Ghostery are variably outdated versions of Firefox, and Brave, Opera, Vivaldi, Bromite, and Smooz are just Chrome.


Bromite is de-Googled Chrome. It's the reason why notifications are not supported; push notifications rely on Firebase Cloud Messaging / Google Cloud Messaging.

It's actually less tied to Google services than Firefox.


Bromite is the only Android browser I've ever seen which has a "View Source" command, which is both incredible and quite sad. Kudos to Bromite team, I am grateful to them every day.

Yes, I'm aware that in other browsers I can scroll up to the address bar, tap the address bar, stumble-drag my way to the beginning of the text field, and type "view" (switch to symbols view) dash (switch to letters view) source (switch to symbols view) colon.

I've yet to see a single iOS browser which has a View Source command.


The older Firefox browser (up to version 68) for Android had the "View Source" feature. Then it got replaced with Fenix, and we all know how that turned out.


The only viable non-Mozilla alternative to Blink-based browsers are WebKit-based browsers (or the minimal ones with custom engines that are developed to run in the terminal, if you can stomach it).


"Mozilla" is a large umbrella which spans several different rendering engines and a whole host of browsers.


Are there any browsers that are not tied to a large for-profit corporation (Google, Apple, MS) other than Firefox?


I mean Firefox is inexorably tied to Google so that doesn't really work either.


True, but I meant on the technology side of things. Chromium based browsers like Brave have had issues in the past where they pulled in upstream code that, say, connected to Google servers. Firefox is the only browser I know of the doesn't have that problem.


I think your use of the word "outdated" is incorrect.

They are more like "LTS" versions, which maintain a stable user interface while keeping up with security patches and continuously adding improvements and polish.


I'm referring less to active maintenance and more to the tech involved. Pre-57 Firefox, which is what Pale Moon and Waterfox Classic fork, had several design choices which all major browsers, including Firefox, moved away from and are, IMO, inherently insecure (also slow). It used to be a playground for security researchers. I count building shinier versions of 2000's browser tech as outdated.


I consider the point of view that only using bleeding edge browsers is valid as outdated.

I also think that saying "several design choices" but not actually listing them is disingenuous.


Ha, ultimately it is just semantics. That said, I personally wouldn't trust these browsers for online banking or similar uses. The main choices I'm talking about are no multiprocess sandboxing and keeping XUL addons, although the speed and rendering improvements in Quantum are significant too.


I don't need to use online banking, I need to view pages with text and images, occasionally JavaScript, and not get owned by some random 0-day or exploit.

For this purpose, it can be argued that the lesser-known and slower-developed browsers are in many ways more secure than the whales:

They are developed more slowly, with much less code being introduced day-to-day. They are also less targeted, since hardly anyone uses them.

For sandboxing, there are many other solutions, which I think are also more secure if they are externalized.

As far as XUL, yes, it is certainly more secure to disallow add-ons which can do anything useful, like modify the UI.




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

Search: