What we see are the public statements, for public consumption, they are designed to "sell" the changes to the wider public. What we do not see is what is being said in private meetings by officers who get to decide how to optimize the business. So we have to judge not by what is said for public consumption purpose, but by what in effect is being done, or what they plan to do.
This is how personally I see the deprecation of the blocking ability of the webRequest API in manifest v3:
In order for Google Chrome to reach its current user base, it had to support content blockers -- these are the top most popular extensions for any browser. Google strategy has been to find the optimal point between the two goals of growing the user base of Google Chrome and preventing content blockers from harming its business.
The blocking ability of the webRequest API caused Google to yield control of content blocking to content blockers. Now that Google Chrome is the dominant browser, it is in a better position to shift the optimal point between the two goals which benefits Google's primary business.
The deprecation of the blocking ability of the webRequest API is to gain back this control, and to further now instrument and report how web pages are filtered since now the exact filters which are applied to web page is information which will be collectable by Google Chrome.
Indeed, almost the entire ad-blocking market is controlled by the company behind Adblock Plus (eyeo GmbH), who has contracts with Google. It appears they also own AdBlock, and uBlock (not confused with uBO), so during the last years they basically tried to capture the entire market. The fact that Eyeo has >150 employees tells us something about the amount of money to be made from ad blocking. Although they have only published the numbers from 2016, it seems they are quickly approaching around €50 million yearly revenue, with almost 50% of pure profit. For Google this Acceptable Ads Program may be more than a 100 million dollar business.
The only real nuisance is uBO and the future possibility that someone comes along and uses Google's own software to eliminate their core business model.
Basically in this entire environment if an extension does not take part in extracting money out of people, it becomes a problem for most parties involved.
Someone at Google in the higher ups probably realized at one point that giving the user so much freedom and control could theoretically backfire enourmously.
Google indirectly controls ABP, but they want the ABP model to apply to all blockers, so that they both get money from non-blocking users as well as from blocking-users.
In the perfect world of Google content-blocking does not exist beyond mere visual ad-blocking of the most annoying ads.
ABP already allows cookies and network connections, so google still knows everything about those users.
Personally I use a combination of pi-hole, third-party cookie blocking and uBO, which takes care of basically all cross-site tracking. But when I recently had a look at another system of someone who uses ABP I noticed that the blocking really is only visual, theres still a profile that is being sold to data brokers, you just don't see the stuff they recommend to you.
The default settings of ABP are also extremely anti-user.
ABP/Eyeo is a wolf in sheep's clothing.
uBO users on the other hand are basically invisible to the survaillance capitalists.
This is why it’s so unfortunate and frustrating that people are depending so much on Chromium/Blink. Electron, various browsers that are just skins over the embedded framework...
We were still dealing with legacy IE6 (and IE 7/8 by that point also) in 2013! Almost a decade since we got Firefox and the first version of Chrome.
The web has empirically suffered through homogeneity and the lack of strong competitors in the browser space, not helped by the underlying HTML/JS spec becoming exponentially more convoluted to the point where building your own renderer is nigh impossible.
And since everyone is standardising on a Google, their decisions on ad blocking and supporting APIs are automatically going to flow down into every dependent product.
It’s frustrating that MS copped out and went Chromium. Like we never learned this lesson the first time.
We still have WebKit and Firefox and (sooner or later), more of Servo at least. These need to be protected if we don’t want Google and the ad network to totally control the browsing experience.
Although, controversially, we as a web dev collective have learned nothing if we continue to only develop for a single browser. Not for the web itself, but for one particular rendering engine. “Only works in Chrome,” is poor engineering for anything more than a prototype or proof of concept.
Why do people act like this is all some great big accident, the result of eternal human fallibility?
It's not. This was planned. It was the objective all along. Why else do you think Google Maps is free? Gmail? Why did they drive around 1,000s of cities with cameras on a car taking photos for Street View? Was it really just to become the world's most comprehensive search engine? Hell no, it was to get people into the ecosystem and to stay there, and be content with doing so. Just think how much money and manpower went into each of those free products we take for granted, not the least of which includes the Google search engine; now think about how they make so much money off your attention that they can offer it free of charge, because sucking you into the ecosystem is just that profitable.
Chrome is the same thing. Why would a for-profit company develop a web browser more or less unprompted and give it away for free? To draw people into the system and get them comfortable with staying there. Now that they have monopoly power, they can start tightening their grip with the good old embrace, extend, extinguish.
Google isn't unique in this. Similar arguments could be made about Facebook.
To be sure, Google and Facebook have produced tremendous advances in science and technology. But let's not forget what funded them, or why they were developed in the first place. Lest we forget why these things play out the way they do.
> Was it really just to become the world's most comprehensive search engine? Hell no, it was to get people into the ecosystem and to stay there, and be content with doing so.
>Chrome is the same thing. Why would a for-profit company develop a web browser more or less unprompted and give it away for free? To draw people into the system and get them comfortable with staying there.
I think you're giving them too much credit, bordering on a conspiracy theory. There's no way they knew all of this would happen and that they would now be in the position to dictate the plug-ins we use.
I'm not saying they didn't want that and they are not happy to be in this position.
What I'm saying is that their initial plan with Gmail/maps/chrome was to gather more data and give themselves the competitive advantage in the ad selling business. If you remember, when most of these products started they looked nothing like each other, quite telling that there was no central plan to create an ecosystem.
The ecosystem idea started much later, with the failed attempts to make social network and with the not-so-failed Google Now which is when they finally started bringing all the data they had together.
Surely they took advantage of their position eventually and managed both to create an ecosystem and to successfully lock people in, but that doesn't mean that they planned it since 2003.
> I think you're giving them too much credit, bordering on a conspiracy theory. There's no way they knew all of this would happen and that they would now be in the position to dictate the plug-ins we use.
Think about it. The time when Chrome came around, Firefox was poised to become the dominant player in the market, barring Safari, which was big on mobile but small on traditional desktops. To some extent, Mozilla depended on Google's money, but was still an independent body, and I suspect, Google came to the conclusion that Mozilla could thrive without their money.
I would not rule out the possibility that their top management decided that it'd be better to have another, Google controlled browser in the market, just in case.
So the Idea would have to be there, to have another browser in the market, just so to make life tough for Mozilla. And being nerds, their engineers put focus on speed, performance, etc.
And Google put a lot of money into the Chrome branding. I remember seeing ad's for Chrome, on huge banners, on prime real-estate in Indian tier-2 cities. Nobody does that sort of advertising, just to get people to use their browser.
Google's thought process absolutely involved the concept of getting people into their eco-system and keeping them there.
May not be a 'conspiracy', but definitely wasn't simply "Hey lets build a great browser, just because we love technology and we can do it"
The ad-Blocker market blew up in response to privacy concerns. And it directly threatens Google's revenue bread and butter, Ads and data collection. Google will fight till death to maintain the status quo, disable any meaningful ad-blocking software. Even if it means risking being a monopoly and paying fines. Even if it means it gets branded as evil.
Alternate motivation: People stay engaged on the web more when pages are faster. One thing that makes pages faster is a faster browser. When people stay engaged on the web more, Google makes more money. Chrome's initial big selling point was being faster.
The core motivation wasn't speed -- it was ads. Chrome was designed to send people to Google Search instead of websites. The address bar auto-completion in Firefox would always suggest actual URLs, but Chrome's would send people to Google Search to click on ads on the way to their desired destinations. The ads started to become camouflaged so that most users couldn't tell them from organic results, and Google continued to make the ads harder to distinguish over time. I'm sure that Google also had long-term worries about ad-blocking and wanted to control the browser.
"We just want to make the Web faster" is what they tell their employees so that otherwise ethical people will write code that does unethical things. (AMP/portals is another example.)
As pointed out by another reply to your comment, Chrome was about pulling the users into googles ecoststem. By combining search bar with address bar, it blurred the line between a google search and a web address.
Just like Facebook, with its free basics internet deals with mobile providers in emerging markets, sought to confuse people into thinking Facebook is the internet, Google also sought to confuse people to think Google is the internet.
The investment in speed and other user benefits was a loss-leader for the goal of increased control. The speed and other benefits were indeed a net positive for users (and motivated Mozilla to focus on performance in Firefox) but they were investment made for payoff.
You've said it yourself: "the big selling point was being faster."
And yes, there is obviously some marginal value for Google in simply making web experience as fast as possible because it keeps people using the web. But that extended web use is of no good to them if users are doing so at non-Google properties that don't see Google ads, so it seems implausible that control was not a key motivator.
> That extended web use is of no good to them if users are doing so at non-Google properties
Extended web use is often punctuated by google searches, even if a lot of the engagement happens on non-Google properties. If slowness causes people to get bored and go do something else, Google suffers.
I agree with everything but Safari was never a contender for the status of "standard" browser. It's a browser limited to a single vendor platform and a relatively small one, expecially on the desktop. Apple never had plans to move it anywhere else, with the exception of an aborted version for Windows XP more than 10 years ago.
Google was frustrated giving Mozilla tens of millions of dollars and all they saw was that Mozilla was dicking around with the UI. And every year Firefox got slower. By losing the browser wars, the Google business model could be put to death.
There were efforts to make chrome faster and more portable but that wasn't the main idea.
Google considers itself the ultimate place for programming on Earth. The open Java reimplementation, the state of the art just-in-time compiler (I know the designer), and Octane and extension architecture were imho the main reason to take over the browser platform. Microsoft dominated PC programming. Google would dominate Internet programming. This is why they started the chrome team and de-funded Firefox.
>Was it really just to become the world's most comprehensive search engine? Hell no
Many around these parts still believe this, and that Facebook is trying to connect everyone, and that Tesla wants to accelerate renewable energy. Such altruism!
Bingo. Deanonymize the majority of users to the point where only privacy activists, power users, and grey/blackhatters are running adblockers, compile a comprehensive list of such users, and then either find another way to track them or straight up ban adblockers because "only the bad guys use them, what do you have to hide?"
Their concern about losing the ability to "personalize" ads smacks of NSA/GCHQ concern that they can't reliably track X percent of the population (UBO/pihole/etc users). Sure, Google makes a ton of money from ads, but with their stranglehold on the search, mobile, navigation, and video markets, do they really need to make such a controversial change just to squeeze a few extra million a year out of Ads?
No. There's plenty of other ways to make money for their business, there aren't many other ways to remain in good standing with and funded by the surveillance state. There's something else at play here.
This is compelling, but possibly too charitable to the ad tech industry. Google isn't a hapless vassal of the surveillance state. They make a fuckton of money off of your attention.
It's more or less just capitalism in the tech era.
Google became 'evil' as soon as it realised that exploiting user data was the route to profit. We live in the information age - any for-profit business is going to pursue that which makes it the most profit.
Since most people are used to the idea of online services being 'free' - they got used to the relationship imbalance, and few companies ever really inform their users about exactly how much value their usage produces.
The surveillance aspect is definitely real - but I suspect it emerged over time as Google was forced to work with law enforcement, security services etc as its reach and power grew.
It's the never ending cycle, they will start to tighten the grip and other options will pop up with time, people need to learn that losing a bit of convenience is not the end of the world.
So you claim that there are many people who have been hurt by Google's data collection, but Google has censored their cases away?
Can't you find them with Bing?
Give an example!
It's a problem if Google bans your account, and you built everything on your services. But that is not the same thing, and any other company banning you from doing business with them would also hurt.
If it doesn't work in Firefox I don't use it. I've been using Firefox for as long as I can remember, probably shortly after it was made and Netscape died. And would never use Chrome. Chrome users are now going to be betrayed and just didn't see it coming. I look forward to Firefox' upcoming rebound which I have every faith this news will spur on.
I came back to Firefox a while back - maybe a year or two? - and have never regretted it. It's a fast, modern, respectful browser. Maintained by a non-profit. I do all my front-end dev work in Firefox, and also test in Chrome. Never had a problem. Come on back folks!
The only thing I miss is the ease which you can create and use different profiles in Chrome. Facilitates testing web apps. Firefox's support of multiple profiles is kludgy.
This is funny - because I actually resisted switching to Chrome initially because it didn’t support multiple profiles at all.
FWIW, I find Firefox’s support for multiple profiles fine - it just requires the -P argument, which you can wrap up in a shortcut or launcher script. (On Windows, add -no-remote to force a new instance to launch). -P takes an optional profile name, so you can add it to your scripts to auto-launch a Firefox with a particular profile preloaded.
Sure, it could use a proper UI, but if enough people clamour for it I’m sure that can get added. The fundamental support for multiple profiles is quite good.
No, this suggestion is wrong. I looked into using FF containers as a chrome profiles replacement and they’re not the same thing at all. I can dig up my notes if you like, I think they’re posted on HN in fact. FF has true profiles and those are what you’d use.
I agree with GP. Chrome profiles are what’s stopped me switching to FF.
It does depend on exactly why you want profiles -- for my use cases, Multi-Account Containers (and Temporary Containers) do a much better job of helping me achieve what I want to do than either browser's implementation of profiles.
I maintain that it is a widely-propagated misconception that Firefox containers can be used to replace Chrome profiles. Here are my notes from when I investigated:
I went deeply into trying to use containers as a profile replacement, replacing Chrome with the new Firefox beta for one month, and I can report that it is not the right direction to go in:
- New tabs do not inherit current container
- No way to make Ctrl-T do this by customization (I investigated extensions (can't remap Ctrl-T) and even system-wide Ctrl-T remapping with Karibiner; neither gives you what you want)
- History is shared across containers. So e.g. work URLs mixed up with personal. That's contra to one of the main purposes of Profiles.
- External applications do not open a tab in the current container. So e.g. clicking in a link in work slack will fail because it will not open in a tab which has work cookies / google account etc.
Evidently Containers are not designed as a Profile replacement. I'm not sure what they are for but I don't think it's a need that I have.
As I understand it using the long-standing Firefox profiles feature is the way to go, but personally I switched back to Chrome after a month of the new Firefox Beta because of the convenience of Chrome profiles. I should try Firefox profiles, but I exhausted my experimentation energy on Containers.
You're right about the limits of containers. When I'm browsing normally, the default is not to be logged into Google, Twitter, Facebook, etc. Things like my "Gmail this" bookmark fail because I set up Gmail in its own container.
However, I find it very useful to have specific containers for Facebook, banking, other social networks, most of my gmail accounts, etc. So I can be doing stuff for my son's soccer club's email without it affecting my own gmail.
I think of containers as a user tool. It seems like profiles are more of a developer tool.
> So I can be doing stuff for my son's soccer club's email without it affecting my own gmail.
In Chrome you could click on the profile icon in the top right and add a profile "Son's soccer club", and I believe that would also prevent it affecting your Gmail, etc.
> I think of containers as a user tool. It seems like profiles are more of a developer tool.
In Firefox, yes it seems like it. But not in Chrome. The point I'm trying to make (if by any chance there are Firefox people listening!) is that Firefox would benefit from making their hidden profile feature easily available to users, as Chrome does. But then they'd have the confusion of containers vs profiles, so it seems that they should just make containers behave like Chrome's profiles. But Firefox has Profiles! So why did they introduce Containers? IOW it honestly seems like they've made a mess there and the Firefox would be improved by fixing that mess.
I suppose a corollary of that is that Profiles are a poor substitute for Containers. And Containers largely match what I actually want to do, which is to have a shared history while being able to split out certain login (or cache) contexts, without needing a separate window.
The “workspace” and “blackbox script” features in Chrome devTools are real paradigm shifters and a huge productivity boost for me when developing/debugging in JS/TS or SCSS/CSS. With Workspaces, most changes to source files are reflected immediately without reloading (no LiveReload-type tooling is needed).
Firefox has yet to release either of those features, as far as I know. I’ll switch dev browsers as soon as they do.
Something I miss in Firefox, is that the webextensions `management` api does not allow to toggle off/on extensions other than themes. It's no secret that some of the most popular browser extensions require a wide range of permissions to work, while being useful only occasionally. This restriction prevents to have on Firefox a quick toggler without leaving the page like (author here) https://chrome.google.com/webstore/detail/extension-manager/...
I switched to Chrome the same day the beta was released mostly because of the speed. Nowadays I stay because I keep my passwords synced over my Google Account and I'm not sure if Firefox works with that.
Regarding speed: Firefox will soon switch to a new rendering engine, WebRender. It's at least as fast as Chrome's. Then Chrome's speed advantage should disappear.
Did that years ago when I switched back when Quantum came back; Firefox still has the good old venerable "import passwords from another browser" functionality. You just need to have your passwords sync'd to a local install of Chrome and Firefox can read them all and sync them with FF Sync (which does not force you to use Mozilla servers, btw; you may host your own server for max privacy).
It's not just their service. You can run self-hosted service for sync [0] and fully control it. From privacy point of view that's far more interesting than any other solutions I've seen.
I've been doing this for a while but the part I haven't managed to solve is that using your own sync server on Firefox for ios seems to require you to also run your own Firefox accounts server - which is much more painful to set up. Is there a similar easy guide for setting that up?
The way I have switched password managers is to run both for awhile. Overtime you move the ones you need. Eventually if you leave a couple behind you can always reset them in the new browser/manager. It's not as big of a pain as it looks going in.
Sure it does. I keep my passwords synchronized with a Firefox account through my Windows, Linux and Android devices. Works like a charm. I won't trust google to keep them.
That is the second most important thing why I'm using Firefox, I sync my passwords via my own instance of the sync server https://jeena.net/firefox-sync-15. I don't even trust Mozilla not to do something stupid with my passwords (by accident) and Firefox was always the only browser which allowed to use your own sync solution.
Same here, once I started using Firefox in the late 2000s I never needed anything else. Even when Chrome was the new hotness, I never considered their speed benefits beneficial enough to abandon the privacy-conscious, open-source player I grew to love. Admittedly, I even contributed to their Fire Stick project on Kickstarter and don't feel even a little disappointed that they couldn't deliver.
I never needed anything more than a kick-ass web browser.
Firefox plugins are signed and quite tightly controlled, very easily blocked or banned for business or political reasons.
While there are workarounds to install/run plugins locally, they have been made quite inconvenient and inaccessible to the average user, as seen a few weeks ago with the expired certificate incident
How so? Either they already have access to my computer, in which case browser addons are the least of my concerns, or they've compromised Mozilla in some way in which case who's to say they can't sign their fake addon too?
Firefox are bankrolled by Google, and seem to make changes according to that relationship.
FF seems to be "still give your data to Google" (or other partners we force install for you) but for people who have realised that Chrome is "give all your data to Google".
Note that charitable donations (of the 501(c)(3) variety) in the U.S. can only be employed for a very specific whitelist of purposes, and software development and marketing are not among those (the Mozilla Foundation uses donations to fund things like education programs instead). If you're a software developer who's interested in the health of Firefox specifically, consider instead donating your time by volunteering (not limited merely to Mozilla or Firefox; there are dozens of open-source projects whose development directly or indirectly ends up benefiting Firefox).
I'm told that this is incorrect and that the FSF has been doing it for 34 years, the EFF does it, the ACLU has done it and others have even paid for proprietary software development (their apps etc).
Yep, I'm about to ditch Chrome altogether now. I gritted my teeth and just handled a few of the more bone-headed changes the Chrome team made, but this latest change is a deal breaker.
The truth is, I actually don't need Chrome for anything. Firefox is faster now, gives me more freedom and has less corporate affiliations. I'll stick with Mozilla.
I tend to have to use it at work. The reason I haven't shifted at work is a ridiculous one really - it's the way you can tear off tabs and move them to another screen - it doesn't tear the same way as Chrome. Like I say, ridiculous.
Drag the tab away from the title bar and to a different screen, it should just birth a new window and live in that. You can also drag tabs from one window to another, or drag a window with only one tab to merge it with another window
oh, for sure. it's just that at this point in time, on my macos firefox is simply inferior to macos chrome.
fyi i'm a power user (tens/hundreds of tabs, lots of web development etc), so this might not apply to everybody.
For me, it was Chrome's peerless DevTools (although Firefox has been catching up recently) and the fact that entering full screen on Firefox freezes up the entire UI for me (it's probably some bad interaction with BSPWM).
On my low-end mac (Macbook) Chrome is still noticeable faster than Firefox. I ran Firefox the last year, but then I tried the new Opera and it was so much faster. Trying out others I realized it's the blink engine. So now I am using Vivaldi...
This is much less apparent on faster machines or on Linux/Windows though, but for me it is more efficient to use the same browser on all my machines.
No. This is a misconception that is being spread around. I don't know what FF containers are useful for, but they certainly do not function to replace Chrome's profiles:
I went deeply into trying to use containers as a profile replacement, replacing Chrome with the new Firefox beta for one month, and I can report that it is not the right direction to go in:
- New tabs do not inherit current container
- No way to make Ctrl-T do this by customization (I investigated extensions (can't remap Ctrl-T) and even system-wide Ctrl-T remapping with Karibiner; neither gives you what you want)
- History is shared across containers. So e.g. work URLs mixed up with personal. That's contra to one of the main purposes of Profiles.
- External applications do not open a tab in the current container. So e.g. clicking in a link in work slack will fail because it will not open in a tab which has work cookies / google account etc.
Evidently Containers are not designed as a Profile replacement. I'm not sure what they are for but I don't think it's a need that I have.
As I understand it using the long-standing Firefox profiles feature is the way to go, but personally I switched back to Chrome after a month of the new Firefox Beta because of the convenience of Chrome profiles. I should try Firefox profiles, but I exhausted my experimentation energy on Containers.
I don’t know about profiles, so I can’t speak to that. All I can do is say that I use containers to open multiple tabs of the same site with different users.
For example, I have multiple AWS accounts. To login to two different ones at the same time, I use different containers. Way better than opening one is FF, one in Chrome, etc. When I open a link from a container in a new tab it opens in the same container.
Containers are great to "containerize" cookies but that's about it.
If you have multiple "personas", like a personal account with bookmarks and a work account with another set of bookmarks and settings, you'll want a separate profile.
Yes, that use case sounds very much like what Profiles are for. You'll find that Profiles are an improvement over Containers for tasks like that (logging into the same URL as different identities) for all the reasons I list above.
> Way better than opening one is FF, one in Chrome, etc
> Yes, that use case sounds very much like what Profiles are for.
I'm not quite sure how profiles work in Chrome, but from a Firefox point-of-view this is only true if you also want to keep your history/bookmarks/settings/add-ons completely separate, too.
Yes, history, bookmarks and extensions are separate. I have one profile for "me in my personal life" and "me at work", and a few for testing logins to different interfaces at work. So this degree of separation is mostly what I want, although it can be a little annoying to maintain the extensions you want under both profiles.
Keeping history separate in particular is very valuable since nowadays the fastest / least hassle way to bring up a site is to start typing the beginning of the URL in the navigation bar, so you wouldn't want work URLs mixed in with non-work URLs.
> “Only works in Chrome,” is poor engineering for anything more than a prototype or proof of concept.
Can't we, developers, do the opposite? Like introduce a small annoyance, like showing a pop-up, when visitors use Chrome? It could say something like "today is free web day, upgrade your browser to Firefox or any other libre browser"
What you, as Web developers, can do, is use Firefox while developing your site to ensure it really works well on non-Chromium browsers. And do your level best to encourage your developer friends to do the same. If they're on Mac and would rather use Safari, that'll do too.
Then, every time you see any site or any Web developer satisfied with "works in Chrome", do what you can to let them know that's not acceptable. In a polite, loving, and extremely firm manner of course.
A decent idea of course, but you are going to run in to the same issue as everybody else who tries that: unless Firefox has a significant market share, it is just not worth throwing resources after it.
What you can do is develop automatic filters to make chrome specific CSS prefixes general, etc. Those are probably worth using since a few hours/days of engineering time easily is worth the larger market share.
Firefox already implements a bunch of -webkit prefixed features for compatibility reasons. Are there specific Chrome-only features that aren't yet supported that you think should be? If so, file Firefox bugs!
I'm still working out the details, but when I get to the point where I'm selling software on my site, I plan to offer small discounts (maybe a dollar) to anyone I detect on a Firefox browser. I'm considering doing the same thing for Ad Blockers.
I know I can't convince normal people to care about the web ecosystem, and yelling at users about behavior is just another way to annoy them. I like the idea of having a small (likely secret) list of browser behaviors that reward users, rather than punishing them. It makes it feel more like a game or a cheat code or easter egg to me than a heavy handed lecture.
If someone is visiting on an unconfigured browser, or something I don't recognize, they won't get an error message or performance hit or any notification at all. But if someone visits and they're doing the right thing, maybe they get a "good job" and a discount or extra download. And then hopefully if they recommend my software to a friend they'll also let them know about the "secret."
> I like the idea of having a small (likely secret) list of browser behaviors that reward users, rather than punishing them. It makes it feel more like a game or a cheat code or easter egg to me than a heavy handed lecture.
While that is nice, you run the risk of nobody noticing.
For some strange reason, I wanted to comment to your post and you didn't have a reply button initially. I had to refresh and then is showed up. Huh, never seen that behavior on HN.
Agreed, but the point is specifically not to nag people -- I don't think that works for Open Source communities; people just get mad. So yeah, tradeoffs.
To put it specifically in advertising terms, I'm also optimizing for conversion rate, not impressions. I'll make a (light) prediction that the few people who know in advance about a system like this will be more likely to try out a browser to save a dollar than they will be to switch a browser to stop a negative experience they're already in.
And on a less practical note, I think I'm OK with fewer people discovering something like this if the ones who do notice end up feeling really good. I want someone's reaction to be, "You noticed! You're right, I am awesome for using an ad blocker!" I want the feeling to be, "sometimes people don't hate me for doing this, and sometimes doing the right thing has benefits."
When the reply button is hidden - to reduce arguments and "over-posting" - you can click the date to go to a page only showing that comment, that page always has a reply option.
It's largely proven by decades of experience at this point that you cannot make the average software user care about ideological concerns. Especially if you use terms like libre that don't makes any sense to people that haven't drunk the FOSS koolaid.
That's not the place to target such a change, if anyone's to do it.
If there's the support of a major maintainer of a popular library, merging changes which incrementally incur larger performance penalties in Blink may be more effective.
If there's no support from a major maintainer, then simply writing contributions which are largely tested in firefox for performance but are tested in Blink for mere functionality should succeed over time in inducing the same.
In the end, the libraries are lock-in for larger SaaS providers far more than they might be aware, and if such changes start making it into e.g. React, there's not all that much that many product teams can do to work around it other than replying via support channels that Firefox seems to take less of a performance penalty.
That's the black hat in me talking. Resuming white hat status now.
No need to blackhat; rather than intentionally degrading it for one browser, just optimize for the other. I'm reasonably confident that there are areas where FFx is more performant; use them.
> If there's the support of a major maintainer of a popular library, merging changes which incrementally incur larger performance penalties in Blink may be more effective.
That library will get forked by industry.
If its license does not permit forking, it's incredibly unlikely that it can get traction in the first place.
You could just add more intrusive ads to your site. The site owner will be thrilled, and the users not thrilled at all. It would work very well for all concerned!
Most sites and content are monetized by ads. Your suggestion is to slow down the browser that allows those ads to exist? What publisher would accept this?
Well, if we're fighting fire with fire, they know because you put in a little dialogue on the page that says, "switch to Firefox for better performance".
We have that already! Except it goes the other way; using non-Chrome on Google properties gives you a slightly worse experience. YouTube on Firefox desktop is still terrible.
> It could say something like "today is free web day, upgrade your browser to Firefox or any other libre browser"
Love it!
I'd suggest a small change: It could say something like "today and every other day is free web day, upgrade your browser to Firefox or any other libre browser"
Huh, developers using electron and/or whole NodeJS/ Javascript ecosystem for their products are already causing big annoyance for users.
If there were whole bunch of conscientious developers we would not be in place where browser engines embedded or otherwise would be the most dominant way to deliver services or products to users.
You could make use of features that work in Firefox but don't yet work in Chrome, such as SVG favicons and position: sticky (Chrome supports but only on th elements)
There's probably a way to compile a list of such features from caniuse data.
People depend on Chromium/Blink because there's no other options. You can't really embed Gecko, and WebKit (Safari) can sometimes lack desirable web platform features.
Yet still, we have Rust (a phenomenally innovative programming language) that was conceived, as far as I understand it, to build a new rendering engine. I've not checked in on Servo for some time but they were going for full on ACID2 compliance with CSS, meeting the specs for HTML...A lot of that work became Electrolysis and then Quantum. Mozilla has been a serious lab for innovation without the conflict of interest in ad revenue.
Even if that effort has changed focus, the community now has Rust and I don't think we've seen such a fresh language paradigm since we got Lisp, Haskell, Ocaml, F#, Scala...
In which case, the world has benefitted from browser competition as a total side-effect of competing with the incumbent browser; we got the various evolutions of C and C++. Exactly the same way we got V8 and then nodejs and the whole server-side rendering paradigm with React and JSX.
If we all fall back to Chromium for everything, then Google has achieved a Pyrrhic victory. They need a disruptor to up their game... and it isn't WASM either.
Servo is very promising, but not ready for commercial embedded applications yet (as of approx 3-4 months ago last time I checked it).
It will likely be a better engine than embedded Chromium when it's done (if for no other reason than that it seems to have fewer dependencies[0]), but I wouldn't start building an application on top of it today.
I am highly interested in finding a more performant alternative to Electron. I'm currently building a web-first game that will also be available as a fully offline native app. I'll probably use Electron unless the ecosystem changes drastically before I'm done. Some really interesting projects out there -- not just Servo, but also thin wrappers around OS web views, even a few re-implementations of CSS/HTML that just force you to cross-compile or port your Javascript to another language.
But I haven't found any that were mature enough that I felt comfortable using them. Servo was the most promising project I personally have seen so far, but it needs more time.
I've been really liking rust thus far, its pretty awesome and I like the community. I didn't know it was made for a rednering engine. It would be nice to have something built by someone other than Mozilla though, just purely for more competition. I could care less that IE sold out, they made their bed in the 2000's and there was no escaping it. As a corporate programmer I am glad they a
I've been really liking rust thus far, its pretty awesome and I like the community. I didn't know it was made for a rednering engine. It would be nice to have something built by someone other than Mozilla though, just purely for more competition. I could care less that IE sold out, they made their bed in the 2000's and there was no escaping it.
Disclaimer: Work for Mozilla but not on the browser core, thoughts are my own, grain of salt, etc.
My understanding is that on Desktop it's much more difficult than it ideally should be.
With that said we've got a pretty solid story being worked on for mobile via GeckoView [1] and Android Components [2], there's a post on the Mozilla Hacks blog about our use of them in Focus [3] and they're also what is being used for building the "next generation" version of Firefox for Android currently code-named Fenix [4][5].
I wouldn't be surprised if there was an effort to get some of the GeckoView work back onto our desktop platforms.
Speaking as someone who worked on Gecko at Mozilla for many years, that page is years (if not more than a decade) out of date and embedding Gecko is a rather miserable experience.
The problem is the only way to combat Google in this regards would be for smaller players to come together and form an open-source working group based on a chromium fork. (The train has already left the station for anything besides WebKit or Blink, since the lack of accessibility of Gecko means it will never be adopted by anyone else.)
But most of them are directly financed by Google and have almost no common ground (e.g. Opera, Firefox).
And unfortunately no single player involved can gain much by going against Google. What would Microsoft gain from forking? Nothing.
I think 10 years in the future we might see WebKit and Blink merge together into a single core engine.
Modern Capitalism almost dictates this development, as corporations strive to save money at all costs.
Crack the DRM and do it in a country that does not care about DMCA and has lax copyright laws. You could create an "illegal browser", and judging by how effective all the other antipiracy efforts have been to date (i.e. not at all), I doubt you'll face too much opposition...
Or go the "insert DeCSS here" route that some used back when that was a thing. "If you happen to have somehow acquired a WideVine lib, having hash [some string], stick it here if you want to watch Netflix."
Is that true? IE was incumbent and Firefox unseated it. They had the legacy of Netscape, sure, but they didn’t throw the towel and reskin IE the same way other browsers of the time did.
Besides that, at which point does the dominant search and advertising company of the internet get an antitrust case for this? People think google is trustworthy yet they do nothing to earn that trust.
This kinda erases history. IE was riddled with security flaws and stagnant beyond belief when Firefox usurped it. There was a compelling reason to use an IE alternative back then - they actually offered something IE couldn't give.
Do these changes make Google less stagnant, as they disempower the discerning user who wants at least some control over their browsing experience?
Google's new security flaw is that they are going to moderate ad-blocking, become responsible for it, and then depend on their inhuman machine to ignore all of your complaints. No different to how you can't find a human connect when you get screwed out of Gmail or Google Wallet because of their mathematics.
As a total thought experiment on where I'm coming from with this: in Star Trek TNG, Data (the Android) spends seven years with us exploring the human equation. Strip out the narrative imbalance and do you think the current automation of help and support is anywhere near aligned to the ideals of that 30 year old TV show that didn't know better? Or is the algorithm an overfitting to Google's commercial needs?
It's tangential to this thread but for some reason it felt worth writing out.
Mozilla can (and does) offer putting the privacy concerns and an open web first.
Google has worked very hard over the years to make itself into an organization with deeply baked-in incentives to do the opposite. It'd take a decade of sustained effort working counter to powerful incentives to fix that.
This is distinct from making a capable well-performing browser, which Google has done. So has Mozilla.
Standards often shift be a substantial reconsolidation, in which numerous side features are discarded.
HTML itself emerged from such a consolidation (from SGML) and went through this process a few times. There are numerous other examples of technical recapitulation.
HTML and the menagerie of related standards has never, as far as I'm aware, become simpler. And the old versions have never been removed from a browser.
XHTML is hard-structured, and among its negatives, requires being fully downloaded to be parsed and validated.
HTML, including H5, has soft-fail modes.
At least that's the justification I generally see. See the Criticism section of the Wikipedia article:
You're responding to a thread pointing out that browsers are too complicated to implement, which has forced everyone but Google and Mozilla to give up on providing browsers and web standards.
And Google is currently the funding source for Mozilla, giving them a more or less complete monopoly on the future of the web.
Chrome might not have that many security holes, but if you browse without adblock you will quickly see just how festered the web is with horrible, horrible ads.
Chrome without adblock is essentially ie6, with tabs.
I'm not sure this analogy works, because internet explorer was pure garbage. It was worse than garbage, it was like the nasty water left over at the top of the industrial drain, and by just looking at it you risk losing your eye sight. IE was more ripe for disruption like few products ever are, it was so ripe we got not one, but two competitors (chrome)!
Pound for pound, chrome is ridiculously far ahead of internet explorer mid-2000's. IDK what the hell someone is going to have to do to upend chrome and its rendering components but is going to have to be great.
It's not what someone had to do to upend IE or that they will have to do to upend Chrome, it's what Microsoft didn't do to keep IE in power: they didn't do anything.
I agree with that completely, but personally, I seriously doubt Alphabet is going to let its flagship product's flag ship delivery vehicle falter like ie did.
I mean, who knows. Never say never, but they are going to be tough competition. Not to mention that browsers are becoming so advanced it will be tough for a small team to match the engineering prowess of a company like google.
Microsoft basically shut down the IE team after they beat Netscape, which gave Firefox a chance to produce a better product. Web browsing is so closely tied to Google's core business that they're unlikely to mothball Chrome development.
Genuine question, will Apple leave all the tracking/advertising money on the table in order to best Google on privacy? Or, once they win enough market-share will they just sell out?
What do you mean "lack of accessibility of Gecko"? Did you mean "lack of embedability"?
Gecko's embedability sucks ... on desktop. But that's not inevitable, it's just a matter of priorities and resources. On Android it's getting a lot better and something like GeckoView on Android could be made to work on desktop too ... if there's demand and perceived benefit.
(Mozilla's in the middle of upgrading Gecko's multiprocess support for fine-grained "site isolation" so now might not be the ideal time to stabilize embedding APIs.)
That might be true. They're all incredibly complex beasts. I don't think it matters much for adoption in terms of embedding or forking to create a new browser; the engine internals can and should mostly be hidden behind an API.
> We still have WebKit and Firefox and (sooner or later), more of Servo at least.
Do remember though, Mozilla's income is heavily dependent on Google. If we want a browser engine that can remain completely free, we really need a engine that is community driven. Although that might not longer be possible given the complexity of the web nowadays.
> This is why it’s so unfortunate and frustrating that people are depending so much on Chromium/Blink
Chromium is generally fine, until Google packages it as Chrome. The issues are not inherent to Chromium, they're failures of principle at Google.
I know people like the idea, in concept, of engine diversity on the web, but the alternatives are terrible.
Anyone who says that Firefox runs anywhere near as well as Chromium on Linux is either incredibly lucky, extremely knowledgeable about custom building Firefox, or just lying. On Windows, the story is a bit better, but it's still just not comparable. On top of this, in my experience, I've found the Firefox UI extremely frustrating.
My take is that when Mozilla ousted Brendan Eich, something changed culturally at the place; it's no longer a culture of competence, but one of paranoia, reluctance, excuse-making, distraction, and (sometimes) bullying.
Regardless of the causes, we are in a situation now where the only competent browser which handles basic webpages the way normal people expect, without much fiddling with configuration, is Chromium. Everything else relies on excuses and wishful thinking.
Brave is looking good, it has all of the extremely popular and well-thought-out UI of Chrome, total compatibility, and a backbone. In the worst case, it can survive on its own.
Chromium is really great, and whoever packages a principled, non-user-betraying browser based on Chromium (and convinces people to use it), will be on the most pragmatic path to preserving the open web.
By default, google disables hardware acceleration for chrome on a large percentage of linux. As well, Ive never had it work well with open source graphics drivers or new products like wayland. And the desktop integration comes nowhere near firefox
My understanding is that this change in the extensions API will land in Chromium first, and so any browser depending on the extension mechanism of Chromium will be impacted, like Opera and Vivaldi. If I'm not guessing wrong, Opera and Vivaldi are compatible with Chrome extensions because they use the extension mechanism of Chromium, so they'll need additional maintenance burden to keep the `webRequest` API working.
Anyone who says that Firefox runs anywhere near as well as Chromium on Linux is either incredibly lucky, extremely knowledgeable about custom building Firefox, or just lying.
Firefox used to be much sluggish on linux, but I tried out quantum and it seemed just as snappy as chromium. This announcement is enough to make it my daily driver, I think (I'll have to figure out the dev tools, hopefully it's straightforward).
>almost the entire ad-blocking market is controlled by the company behind Adblock Plus (eyeo GmbH), who has contracts with Google. It appears they also own AdBlock, and uBlock (not confused with uBO), so during the last years they basically tried to capture the entire market.
This makes me wonder if there's any way we can give gorhill a bunch of money in donations. He's in a much more important position than I previously realized, and despite the statement in the uBlock readme ("Free. Open source. For users by users. No donations sought.") I think it would be an excellent idea to make sure he has permanent financial stability for as long as he continues to work on uBlock.
If he wants a multi-million dollar corporate buyer to come along, I'm sure he can get it. But as long as he's incentivized to continue building a truly pro-user ad blocker, we should at least make sure he's not in a situation where he could be forced to sell out.
That may be fair (though there's some ambiguity to me in the word "sought"), but I hope he knows that if he loses his day job or has health problems or something like that he can go to the community. uBlock origin is worth hundreds of dollars a year to me, and I'm by no means rich.
Well, I'm not seeking money when I walk down the street, but I wouldn't consider it rude if someone decided to throw some at me. (assuming cash, not coins thrown at my head though even then, if they're quarters, I'm picking those up)
Currently, we can go use Firefox or Joe's homemade browser to to back the features that Chrome removes, such as the ability for an ad-blocker to access and remove content. But we've already seen just last month [1] that Google actively prevents Firefox from using some features of Google products (need to find the HN link). Thus Google as a content provider can easily steep people into using the Google web browser.
This is really starting to smell similar to the Microsoft - IE debacle all over again.
This thing about Google preventing Firefox from using some features (Edge as well I believe) - I'm assuming this is all done using the user agent header? or are they detecting Chrome through another means. I guess I'm wondering how effective spoofing the user agent header would be in resolving that.
This naming is a catastrophe. I just tried to find out which is which and it isn't easy, even if you know what you're looking for.
I'd think that uBlock Origin should consider changing the name. Yes, "it means giving up", but it's a battle they can't win anyway, and at least users will be able to tell which one is owned by eyeo.
EDIT: AdBlock and AdBlock Plus are similarly confusing.
> I recently had a look at another system of someone who uses ABP I noticed that the blocking really is only visual, theres still a profile that is being sold to data brokers
It's quite shocking, anti-intuitive, and really shows what the web advertising business is really all about.
Adblock plus have a committee that is supposed to be independent companies that they consult but they invested in some of them and made those people the head of each group so its not even independent
eyeo is a pure scam and they make 99% profit because they have no costs
> Someone at Google in the higher ups probably realized at one point that giving the user so much freedom and control could theoretically backfire enourmously.
If that is the case, it seems very short sighted.
While this change is obviously not in the interest of any user, there are multiple ways around it.. the alternative solutions are not so easy right now, but they will become more user friendly as they rise to the surface as the new way to block ads. The affect on Google's revenue will be short term (if any), but I suspect the affect on Google's public image will be significant and lasting. It doesn't seem worth the risk to Google for such a short term gain.
> Someone at Google in the higher ups probably realized at one point that giving the user so much freedom and control could theoretically backfire enourmously.
That's ... well, so much for the internet ... :-(
It's really against what I imagine to be some of the core principles of the internet.
Welcome to the point when even free market people will no longer defend your monopoly, Google.
The number of egregious examples of their centralization backfiring against the freedoms within the day-to-day life of internet users keeps growing bigger at a seemingly exponential rate.
The barrier for competition may be high but history is littered with examples of giants withering under their own decision making. Nor will they forever be immune from antitrust laws.
I don't consider Google a monopoly. I figure many "free market people" define a monopoly pretty strictly. They're not forcing me to use their browser, search engine, maps, email, or many other services.
However, they do have a disturbingly high level of market dominance. So seeing as I am a part of the free market (after all, the "economy is us"), I have opted to minimize my use of Google products and services and encourage others to do the same. This wasn't the tipping point for me; I've been advocating alternatives for quite some time.
I don't want Google punished with anti-trust laws if that can be avoided. I'd much rather see users apply corrective pressure. Of course, that assumes users appreciate the importance of their privacy, which makes this a lengthy game of messaging and persuasion.
What we call "monopoly", is "being big enough to put a thumb on market scales" in legal theory these days (note that this doesn't even require a company to have even 50% of market share...)
Even if you decide to believe that Google is doing this for the "right reasons" (improving performance for instance) it should still be a huge red flag. Due their conflicting business model they have no incentive to improve the adblocking interface later on. You can therefore expect the adblocking capabilities to degrade every time they can come up with a reasonable reason to do so while they'll never take the time to actually come up with a better solution.
Switch to Firefox people, it's not perfect but at least it's not Chrome.
Yes we can ditch chrome. But what about general population? Now that websites may choose to give incompatiblity prompts to Firefox since it impacts there revenue through ads. They will ask users to switch to chrome. This might affect Firefox in long run. It's market share might drop.
I'll bring out the "computer has viruses" argument. As in:
"Remember how you asked me to clean your computer from these toolbars and make it faster? This problem may happen again if you browse the web without uBlock Origin installed. Chrome doesn't support it now, so you have to use Firefox. Ignore websites pleading you to switch back to Chrome. Some of them will try to sneak in this toolbar garbage to you. Actually, stop visiting such websites because they're staffed with assholes who want to abuse you. But if you must, do it in Firefox. Yes, preferably in Private Browsing."
As others said, regular people follow the tech crowd. My family uses Chrome because that's what I've been telling them to install and/or installing for them for the past few years. But the very first thing I also install for them is uBO, and if that stops working, Chrome goes out of the window.
How about Vivaldi? Or Opera with built-in ad-blocker (which I'm currently using and it works great, at least with uBlock Origin installed too)? Unfortunately I can't use Firefox on my system ( https://www.reddit.com/r/firefox/comments/a42xei/state_of_fi... nothing changed for years with these problems and I believe it will unfortunately stay that way)
Aren't Opera and Vivaldi just a re-skinning of Chromium?
If yes, then it probably won’t save you.
And here you have the actual problem with Chrome — unless you fork it and have resources to maintain that fork, its open source nature is absolutely irrelevant. And even if you manage to fork it, you’re not operating at Google’s scale so your fork will be irrelevant.
PS: I use Firefox on top of MacOS and works just fine for me ;-)
I don't know how much different they need to be from Chrome/Chromium to not be affected (and if some built-in blocker like in Opera still will be fine), that's why I'm asking :) There is also Safari, but I would rather cook my CPU with FF than use Safari.
Basically everything except the fact that it doesn't eat huge amount of RAM. I can't stand iOS either, so not being able to sync tabs/bookmarks between my devices is probably the biggest drawback.
"The deprecation of the blocking ability of the WebRequest API is to gain back this control, and to further now instrument and report how web pages are filtered since now the exact filters which are applied to web page is information which will be collectable by Google Chrome."
In the same way a company trying to determine what a user looks at on their computer screen is a privacy issue, this is too. Those filters should be private.
I say this all the time when this kind of thing comes up.
Please use Firefox!
Even if it's worse. Even if its slower[1]. Even if it doesn't have that one feature or bug fix that you personally consider really important. Just use Firefox anyway. Find a workaround. Suffer whatever it is you dislike about Firefox because in the end if we don't act as individuals against the chrome monopoly then google are going to own the web and we'll suffer a far worse period of monoculture than the IE6 ever was.
If you can't go all the way, going part of the way is still valuable. I personally have chrome installed still because there are a couple if internal sites at my work that have problems on Firefox, so I use Chrome for those but Firefox for everything else.
Firefox for Android is also solid browser, and as a bonus you don't see any AMP stuff.
If you're a website/app maintainer, check for compatibility in Firefox.
It's worth supporting Firefox to keep the web the way it should be. I know they make mistakes sometimes, but we need a viable alternative or it will be too late.
([1] I don't think it is, it's made soild improvements in recent years, but lots of people seem to have their own specific issue they hold dear against it)
I prefer Chrome for sure. It full screens video properly (which I need to automatically trigger turning off f.lux), it seemed faster, it does spell checking of form fields.
I'm sticking with Firefox though as these features aren't much of a hindrance compared to the increasingly shitty behaviour of Google, and one benefit is that my Macbook Pro (2014, max spec), doesn't run out of resources and stop responding anywhere near as often as it used to do.
Firefox dsoe nto spel chcek fomr feilds. This comment didn't get any spell checking whatsoever. Last time I investigated this the suggestion was to install a plugin, a hoop I think is unreasonable to make people jump through. I'm using a vanilla Firefox install on a Mac. I've not fiddled with it in any way.
I switched from Vivaldi back to Firefox within the last year for the same reason. I'm growing tired of Google and have been moving away from them piece by piece for the past year.
I prefer the customizability of Vivaldi but Firefox has been fine. I can't think of anything that is a thorn in my side at all. In fact, 99% of the time I can't tell a difference functionally, but I'm not a typical web browser (I disable JS except on trusted sites).
I think for the general population they wouldn't notice much difference between chrome and firefox.
Firefox just runs so slowly compared to Chrome on my 2015 MBP. I have zero extensions and it's a default install. Such a bummer too. ( I use Firefox on my desktop PC though. )
what do you think about the new Edge browser from Microsoft, its based on Chromium.
So you would get most benefits of chrome but its from Microsoft, which personally, I trust more than I trust Google.
I installed the beta for macOS and I'm not happy about it.
It installed a daemon that updates Edge even when the browser is closed. Even if you delete the .app the daemon keeps running and when there is a new update it downloads it and installs it.
There are no settings to change this intrusive behavior.
I sent feedback and complained on Twitter. Apparently the team is looking on it.
Do we know how much the new Edge will phone home to Google? What about raw chromium? Alternatively, are there settings you can change in Chrome to prohibit it talking back to G?
I think, this situation is the same as when you tell people to "vote with your wallet" as to not spend money on the inferior product when the issue is clearly far beyond where people can freely decide this.
Edit to add: Not defending Chrome. I've used IE and Edge since IE9 came out because I saw this thing coming. And I promoted other browsers all the way. Didn't help. I am not moving to Chrome ever and will try and avoid the Chromium Edge as long as I can as I'm not a stranger to sticking with a strange browser choice :)
I'm just saying that maybe something else should be done about Google's browser market share.
> Except there's absolutely nothing inferior about Firefox nowadays.
I mean, maybe for you. But for me, with the add-on that I use, and the computer that I am on, FF is not the most optimal, Chrome is. FF is slower for me and my set-up.
EDIT: I'm not joshing here or trying to trash on others. FF may be faster for you, that's fine. But in my own personal little corner of the world, FF is slower than Chrome. I've no idea why, maybe I could spend a few days trying to figure this out and then optimize FF to make it faster. But, for me, out of the box, FF really is slower.
Google is essentially saying that Chrome will still have the
capability to block unwanted content, but this will be
restricted to only paid, enterprise users of Chrome.
Never heard of a paid version of Chrome before! Can anyone elaborate on this?
I gotta say I'm kind of glad Google is doing this. It will force me to finally abandon Chrome, something I should have done awhile ago.
Googler here, but I do not work on Chrome and I’m speaking on a personal capacity.
I believe this is a misnomer. The enterprise deployment stuff I’m aware of in Chrome/Chromium is accessible by any user, as far as I know, and in the past I’ve used it to force private browsing on always for my own personal usage.
That said, I switched to Firefox when Quantum came out and haven’t looked back. Mozilla has done some annoying stuff over time, but imo the browser itself is really solid, and with some tweaks is very good.
I don't believe it's accurate to say it's "accessible by any user", the net effect of the policy will be (over a time span of months or years) to limit ad blocking only to the most technically capable users who understand how to do what you describe.
Moreover, extensions that continue to use this behavior are required to adhere to manifest v2, and support for that manifest will likely eventually be deprecated. Chrome blocked the store from serving and updating manifest v1 extensions after a deprecation period.
Google rapidly deprecated version 1, from Chrome 18's release adding version 2 in March 2012 to blocking updating in March 2013, to removal of all extensions that had not updated from the Chrome Web Store in September 2013.
If Manifest v3 follows a similar trajectory, Chrome is on a path to remove uBlock Origin from its web store in 2 years.
Just to clarify, what I am referring to is the claim that it’s limited to paying enterprise customers. I could be wrong but I believe that the enterprise policy bits are what matter, and as far as I know all of those are available in Chrome/Chromium today, via group policy systems. When I say “accessible” I am specifically referring to that bit, definitely not suggesting ordinary users should do that.
One of my favorite things with Firefox is the ability to “edit and resend” requests in the browser devtools. I’m sure there are plugins that do something similar on chrome/chromium. But I use it all the time and my colleagues are always blown away when I show them this.
The one thing I always liked about Chrome dev tools was the ability to introspect WebSocket frames. Last I checked this couldn’t be done in Firefox :(
Otherwise, I have found Firefox dev tools to be fairly competitive. I liked the CSS Grid stuff they had around the launch of Quantum; it made it easier for me to jump into Grid.
I use Firefox personally, but develop with Chrome. I have a hard time getting stack traces from Firefox reliably. Oftentimes one of the following occurs on Firefox:
There is no error, but the page won't load.
There is an error, but the error message is generic.
There is an error message, but no stack trace.
When I switch to chrome, I always get the error and stack trace without any further code modifications.
I make sure to clean up for Firefox in releases. But it's just impossible to develop without error messages and traces. I'm on Ubuntu using the latest stable releases.
Ugh. Is this really still broken? I'm not a web dev but I remember having to switch to Chromium 6-8 months ago when I needed to debug a websockets issue.
Ah yeah, on occasion I will boot Firefox at work for this capability.
Truth be told I know of nothing similar in Chrome/Chromium, but you can do something similar using the healthy number of “Copy Request As” context options. I like to copy as fetch and manipulate it in the console. It’s not perfect but it’s useful.
ADM/ADMX templates with 300+ user and device policies
ADM/ADMXs are a list of settable policies (rules) that Chrome will read from the Windows Registry on startup[1]. I'm going to make a huge assumption that the Chrome.exe for Enterprise users is no different to the Chrome.exe that everyone else uses, and it switches into 'Enterprise' mode if it sees certain key policy entries in the Registry[2].
These polices are typically set on a Windows Domain Controller and are pushed to Domain member machines (and users) on the network on a regular basis.
However end-users can simulate the policies by setting these Registry entries manually (There are about 4 places in the Registry where they live). Depending on where Chrome looks, the user may need local admin privileges to do this.
So, what we really need is a 'Chrome Enterprise Enabler' tool, that does this automatically for non-technical users. They would run this tool, the required Registry keys would be enabled, and WebRequest API based Ad-blocking would continue to work.
Unfortunately, none of this helps Linux users as I don't see a Linux Enterprise Chrome package.
---
[1] I've got 25+ years of Windows domain administration under my belt, this stuff used to be my bread-and-butter.
[2] Even if not, extracting just the Chrome MSI installer from the Enterprise bundle is trivial.
Chrome MSI and chrome EXE installers work a little differently. The MSI installer by default installs to the program files directory. I've seen the EXE installer often installed to the local user profile. to convert a regular install to Enterprise you would probably have to uninstall the local version and then install the MSI.
ADM templates are something Firefox is sorely lacking at the moment. I remember reading that it is on the road map but that does not help out much. Nevertheless, I installed both chrome and Firefox and my organization. I like Firefox and use it myself, but sadly it's just not as easy to administer as chrome.
So two things. The MSI installer will convert a local (EXE) installation to a global install (updating shortcuts and removing the local profile install whilst preserving the chrome profile).
Second, Firefox does have ADMX policies (ADMX being used primarily after Vista): https://github.com/mozilla/policy-templates. I believe it took them so long because they were initially reluctant, user freedom etc. But I guess Chrome's success in enterprise made them reconsider. Honestly I never considered Firefox for our Windows Domain because of the lack of Group Policy support. Now I deploy both and users have options :)
FWIW it Is supported under Linux I believe, but unfortunately the documentation is lacking. If I recall correctly, you have to drop JSON files in the right places.
> These are just features and settings, there's no requirement to use them.
I think you missed my point. Google say that the Enterprise version will continue to support WebRequest API. If the Enterprise version is the same codebase as the normal version, then it's a good educated guess that the Policy keys are whats controlling that.
Looks like some useful stuff for deploying to Enterprise, but actually as far as I know the enterprise functionality exists in normal installations. Like for example, this stuff:
What platform are you on? I tried Firefox quantum yesterday and could not stand the text rendering. Gmail looks awful. But I don't have a 4k monitor yet...
People were predicting this as the Chrome endgame years ago. It's playing out exactly as everyone cynically expected.
It's time to eliminate Google from your life as much as possible if you haven't already. Too many wake up calls. They are not a tech company, they are data monopolists. Stop giving them your data.
I see hardly anything even about advertising in there? Let alone anything about ad blockers. It's just generic statements/quotes about monopoly, privacy, browser wars, etc... and a few notes about how promoting the web helps promote Google's products/services indirectly too, which helps their bottom line... which Google itself has been saying openly since forever. There's nothing there particularly predictive of this particular chain of events.
Is there such a thing? The "paid" part comes from the article's gloss, not from Google. I could see it instead being "this is a switch you can turn on in Group Policy."
They mean Chrome extensions installed via gsuite - Google's enterprise management system. Ie - you can have managed Chromebooks and force install extensions. A lot of school's use it for web filtering.
Well, without live heuristics. Chrome ad blocking is basically now limited to a dumb, limited list of URI patterns. Better than nothing, but much less capable than before.
And without privacy invasion implications. Apple introduced content blocking 4 versions ago implemented nearly the same way. No one would accuse Apple of trying to help Google make more money.
Every application has privacy implications, and it isn't difficult to manage this one. Having a giant warning around installing extensions with this level of access would be a good way to protect people while also allowing them to make their own decisions.
Because warning people has historically worked well with PCs not to mention Android.
See Also Vista UAC.
It’s kind of amazing for people to say that they care about their privacy and then they install extensions that can record their entire browsing history and install VPN software on their phones that intercept all of their internet communications.
There's a mountain of difference between "can" and "will", and it's called "trust".
For decades, people have been using software on their PCs which can essentially delete all their data or worse, yet what are the chances of that actually happening? Of course malware would exist, but at the same time, the unrestricted nature lead to a seamless exchange of data and creativity unhindered by any bureaucratic permission-maze.
A world of "perfect" security in which malware and cybercrime cannot exist is basically a dystopian police state.
And we have found better ways on a newer platforms not to implicitly trust apps with unfettered access to user data and surprisingly enough, no one has started rounding up people en masse for looking at cat videos....
... That's how computers work, yes. If you have a way to make a VPN that doesn't see its own traffic, I'm all ears. In the meantime, I trust a VPN more than my ISP.
In my personal case, yes I have because I'm running it myself on a VPS. In general, though, the commercial providers that I would use have stood up to court orders / warrants, so I reckon they're fine.
Unfortunately, I don't have a fast enough uplink to comment on speed concerns.
You don't see any hazard to showing a giant warning to people when they install something that almost everyone should use? If people see a big warning when they install a legitimate ad blocker, people will start ignoring the warnings.
The 30,000 rule limit is a pretty big deal breaker for me, but what sort of live heuristics are you referring to here? Aren't e.g. uBlock Origin's filters entirely rule based (not unlike most other ad blocker extensions)?
"We are planning to raise these values but we won't have updated numbers until we can run performance tests to find a good upper bound that will work across all supported devices."
Ungoogled Chromium is Chromium with all google domains and google-specific code removed. It keeps the nice things about Chrome/Chromium like the dev tools and is compatible with Chrome extentions, though you do have to download and install them manually. Unlike Chrome, it is also 100% open source.
I was under the impression that Chromium had some Google integrations but was pretty harmless from a privacy standpoint unless you sign in to Google. Is that wrong? Do I need to use Ungoogled Chromium instead?
Are brave in control of the web request API within their build? I know it's a fork of Chromium, just not sure how much is available for them to change.
I have been using brave for months and intend to continue if uBo keeps working. Braves internal ad blocking doesn't stop everything.
This kind of crap is why we need to be cautious in allowing Google too much control over web standards, including AMP and their not-iframe element (portals [1]). Whilst the engineers mean well when creating them, Google's main objective is to make money, not to make a better web.
At the moment you have awesome projects like Project Zero [2], but how long till they start strategically handling exploits for monetary gain? Contrast Project Zero to Project Dragonfly [3]. Nobody should be relying on them being good actors.
Also agreed. I've already long since pulled almost everything I use away from Google, with the exception of my cell phone (I'm a Fi subscriber).
Now I'm beginning to look into alternative cell phone plans, simply because Google's insistence on advertising doesn't sit well with me.
I currently work at an advertising agency (doing IT stuff), and I'm in the process of job hunting for this exact reason. I can't justify spending my time helping people who are insistent on finding new, increasingly intrusive ways to advertise.
If google moves forward with this we should all start randomly clicking on ads until advertisers start to realize that they're basically throwing their money away on advertising that doesn't work :) Let's hit 'em where it hurts.
Yes, let's beat Google by giving them more money by increasing the overall advertising hit rate in an organic, nearly untraceable fashion. Wait, what was the logic of that again?
From my experience, many advertisers do so little QA, they'll hardly notice that their conversion rate dips a little. And if they do, they'll look for problems elsewhere: Google is never wrong.
There are people who click on random ads professionally, and Google has been fighting them for a long time. Whatever equilibrium has been reached is probably stable. The best thing to do is block ads and to do our best to stay off the grid - the only thing that no statistician can work with is the absence of data.
They can absolutely work with the absence of data, because you are still doing other things around the web that generates data. They will see you doing other interactions but not ad requests, and they can figure things out from that.
Disabling JS in your browser is like cutting off your leg because one of your toes hurts. It might solve the toe problem, but it's not an ideal solution.
It does not allow blocking Google\any tracking and Google ads, essentially, and gives the Ad industry (and malware) a way to circumvent the blocker with modern tech, because it is a non smart ad blocker but a dumb filter list from 2002.
This is fully intended to give Google the ability to use sophisticated scripts to serve ads and track you despite blocking. Everything else Google says about this is bull!
They probably already have the tech set up. And that real blocking still works for Enterprisr customers is the nail in thr coffin.
If you care about a usable Web, STOP USING CHROME.
Google is has become the #1 anti user and anti freedom company. They want to make money from Ads and tracking, so dont think they will allow you to block ads.
> It does not allow blocking Google\any tracking and Google ads
As per the article, you can still use a rule set to block. Whilst undoubtably less powerful, I would expect that could still be used to block Google stuff.
Why would you expect that? It seems to me that Google does this precisely because they have tech to circumvrnt dumb- style list blocking, but not more advanced heuristics.
Except that list really is just bloated with rules that don't do anything anymore.
> We measure how EasyList affects web browsing by applying EasyList to a sample of 10,000 websites. We find that 90.16% of the resource blocking rules in EasyList provide no benefit to users in common browsing scenarios.
Part of the point of an adblocker is to protect you in uncommon browsing scenarios. This is almost like saying that a malware list is bloated because some of the hashes its storing are uncommonly downloaded. Having as close to full coverage as possible is important.
It also opens up an attack vector for advertisers that they absolutely will exploit. With unlimited rules, there's no reason for ad networks not to use a few domains and serve their ads from a few sources. With a hard limit, there's a strong incentive for networks to collectively try and flood the lists with tons of different domains until we run out of room to include all of them.
> the possibility of tweaking the 30k limit.
This has been the number one complaint about the proposal from day one. If at this point the Chrome team still hasn't decided to tweak the limit, I just don't see how there's not gonna be any new argument past this point that anyone can make to convince them.
Comparing it to a malware list is a bit extreme, isn't it? In one case, you actually get infected, in the other, you see one ad. Your next point about advertisers abusing it is a good one though.
I believe in the response, they said they are running benchmarks to see the performance hit, so they are definitely still looking at tweaking the limit.
I don't think the consequences are the same, but I do think the underlying idea is applicable.
Two things to keep in mind:
A) uBlock Origin doesn't just protect me from ads, it also protects me from many trackers. I'm in the (maybe minority, I don't know) camp that says that excessive tracking and de-anonymization attempts cause people tangible harm. They're not as drastic or as harmful as installing malware on my computer, but I put these practices in the same category as a malicious attack.
B) Under the most recent stats I've checked, Malvertising has almost surpassed general unsafe sites like porn/torrents as a source of consumer malware. Even Google Ads aren't immune from some of these attacks[0]. So for less computer-literate friends that I have, I consider a gimped adblocker to be a malware risk source.
But to your point, maybe a less emotionally charged comparison could be Chrome's automatic whitelist for autoplaying videos. Nobody is going to die if a video autoplays on their tablet, but at a fundamental level the point of blocking autoplay is to actually block it, everywhere. Not to block some of it. People didn't want Google trying to guess which video sites were the most annoying, they just wanted Google to turn off autoplay.
In the same sense, if someone installs an extension that says it's going to block ads and trackers, I don't think it's unreasonable for a consumer to want it to block all of the ads (or at least as many as is possible to detect), not just some of them on the X most popular websites.
It has the advantage of being faster, with the disadvantage of being less dynamic. Existing rule lists will need to be slimmed down. That's not necessarily a bad thing, because all they ever do is grow without being pruned of old rules.
It seems to me people are upset because they read the title and misinterpreted it to me that adblocking as a whole is going away. But to be fair, even the article itself does a poor job of clarifying that.
Google has not shared any actual evidence that there's a performance gain to be had. Non-google benchmarks of uBO show it not being a performance problem in the first place, so there simply isn't much room to be faster than it.
uBO is not the only adblocker around. It's likely not even the most popular. These changes will apply to all of them, bringing the baseline of performance up.
The article says that current ad blocking extensions stops the request from even being made to the ad server, while the new version of Chrome will remove the ability for a plug in to stop a web request.
No, I don't think that is correct. You can still stop a web request, but you can only stop it by providing chrome with a list of url patterns to block. The current method passes the url being requested to the extension, which can then decide to block it or not.
The new way has the extension give a list of patterns to block to chrome, and from there it will block any request that matches.
The stated reason is that this is for privacy, to prevent extensions from being able to gather data about every request a browser makes. Currently, a malicious extension could send every url a user visits to some central collection server. The new way prevents that.
Is the trade off worth it? I personally think no, but I do think many people don't realize that as blockers have the capability to collect every url you visit if they wanted.
It's not for privacy, the people who were saying that are either mistaken or intentionally lying. It's clear to see because the observational capabilities of the API aren't being removed -- only the blocking capabilities. So the privacy issue is still there. Supposedly, this change is just for performance.
What I mean is the fact that they are deprecating the blocking API. That wasn't an action taken for privacy's sake. Otherwise, why wouldn't they deprecate the observational API too?
I'm glad I already made the switch to FireFox. I only use Chrome to access Google apps, because for some strange reason they work a lot better in Chrome...
The only thing I miss is a Session Buddy equivalent. When my computer crashes, it's nice to be able to restore all my tabs and windows, and also it's nice to be able to close a bunch of windows when I travel and then go back to my tab state from three weeks ago.
Maybe wrap Firefox startup in a shell script that backs up your profile data before starting the browser, and then when the problem occurs send a bug report and work with the developers to give them the information they might need.
These bugs suck for everyone and I'm sure Mozilla developers would appreciate any help they can get to track it down.
Have you tried hitting Ctrl+Shift+N to restore the previously closed windows?
These will behave differently: (1) "Quitting" Firefox will multiple windows open. (2) Using your WM to close both Firefox windows. The latter will cause only the last-closed window to be restored, but you can pop the others off the stack too with Ctrl+Shift+N.
Just so you don't feel crazy, I've had this happen multiple times as well (on both Linux and Windows).
In fact, it's one of the larger reasons (tabs constantly crashing is the largest one atm) I leave Firefox for side jobs instead of being relied upon as my main browser.
Hm, I had session restore fail maybe every 3 years or so.
For more features, I'd recommend looking at either the Container feature, that might be able to help, or otherwise third-party extensions - maybe another commenter has a recommendation, since there's a giant number of tab management extensions, and it's hard to tell which ones do what perfectly.
Firefox has this if you are using a Firefox account at least.
If you search around a bit in the sidebar menu you’ll find a list of all the open and previously open tabs on any of your PC’s. Including ability to restore all of them at once.
Yes! Thank you! I don't know why my searching failed to find this, but it is in fact exactly what I was looking for. If this were Reddit I'd give you Gold. You're my hero.
Former Session Buddy user. I'm using Session Sync and it's pretty neat. I think Session Buddy UI is still better and Session Sync needs some improvements (such as selecting multiple tabs in a session simultaneously). However, once I got used to it, I didn't miss it at all.
In addition, I like the integration with Firefox Bookmarks. It creates a folder "Session Sync" and stores sessions as sub-folders. So, if anything happens, I still have them as simple, curated bookmarks.
This impacts you if you're on FF too... if this works and Google is able to monetize Chrome more directly than FF then FF is going to have less development resources.
It's not exactly sessions, but you can bookmark all open tabs (perhaps into a folder called sessions/worktrip_boston), and then later "open all as tabs" using that folder.
I find this useful when researching a particular thing, and wanting to save the resources for later. Sorry I can't remember the exact menu items, as I'm on Firefox mobile and it doesn't do it.
I don't get it. I've used Firefox and Chrome for years. Firefox is easily as good and capable as Chrome. The barrier to switch from Chrome to Firefox is almost non-existent. Is Google counting on the majority of people not being knowledgeable enough or motivated enough to switch browsers?
It's not "as good." I've tried every release of Firefox for the past decade. On Linux and Mac the video performance is noticeably worse than Chrome with significant frame drops, lag, and tearing. Yes, I've tried enabling/disabling hardware acceleration, flash, and all sorts of other flags and options, but the two are still incomparable. Furthermore, even with the quantum update Firefox is still slower for general web browsing and usage than Chrome. I've measured both dropped frames and load times to confirm this.
Yup, let's not sugar coat it. Most HN says "switched to Firefox and haven't looked back" but Chrome does a lot of magic that makes everything faster (or at least seem faster - doesn't matter to me as a user). Firefox also still occasionally freezes even for a few seconds with some busy javascript, haven't had that experience on Chrome in a long time. I'm using FF for ideological reasons but currently Chrome is simply better in terms of user experience.
And sadly web became so dominated by Chrome that some pages simply don't work correctly under Firefox. I'm not going to switch bank or my mobile operator just because they suck at properly designing websites.
Seriously, when is the tipping point? Mine was a couple years ago. Recently with Chrome, there is forced login to the browser when you sign into your google account. Also, when you clear all cookies- it keeps the google cookies. Now it’s crippling ad blockers. Chrome is no longer a user agent, it’s spyware. But I guess as long as that crappy page overloaded with ads and trackers doesn’t freeze for 2 seconds, then it’s still the best.
I only use chrome on my work computer now, and even there I use firefox for logging in to my personal gmail.
My son just got his first computer -- a windows pc. He has edge and firefox installed, but no chrome. I told him he cannot trust google :( and should avoid their software as much a possible.
Uh, that's not my experience at all...I use FF normally, and Chrome not uncommonly (sometimes it's handy to have an entirely separate browser, plus compatibility testing), and I find no noticeable difference.
I agree in windows, but in Linux and android FF has some issues. I still use it but sometimes if a page is being super slow I'll open it in chrome instead.
Hmmmm...not denying your experience, but I haven't used Windows routinely in years; my last several laptops have been Linux. Maybe I just use different websites?
I've been trying to force Firefox (latest Firefox Developer Edition) upon myself yet again for the past week, and aside from performance problems which still seem to exist on the Mac (initial rendering of pages is slow), some things just annoy the hell out of me:
- Search results are not highlighted on the scroll bar, which means no information on grouping at a glance, and can't just scroll to the relevant part. I have to go through matches one by one.
- Font rendering is subtly broken as always. A specific example is font-variant-numeric: tabular-nums, which seems completely broken at least for -apple-system (.SF NS Text); as a result a data aggregation site of mine which I read all the time looks awful, larger numbers may be narrower and vice versa. (Oddly, font-variant-numeric: tabular-nums seems to work for "Fira Sans" on MDN; I haven't found another font it works on.)
- With developer tools open, pages are really slow.
I'm probably forgetting something else.
(I should add that I also forced Safari upon myself, which is a shitshow, problems too numerous to list. Until Google completely removes ad-blocking capabilities I need, Chrome is simply the best there is.)
I've been using Firefox for the last two weeks or so. It's fine, but noticeably slower (on a 2018 MacBook Pro) when opening new tabs and especially dragging tabs into new windows. It also seems to get stuck resolving DNS sometimes, which is probably an artifact of the corporate network, but Chrome doesn't have that problem.
I'm having the exact opposite experience with Chromium: Sometimes, two to three tabs (regular websites, no videos, no fancy stuff) already make it freeze. My Firefox, in contrast, runs smoothly even with dozens of tabs open.
The DNS thing is probably caused by Firefox's OS-agnostic DNS. You see, in Windows Chrome will use IE's networking stack (so changing your IE proxy settings changes them in Chrome) if you have the "enterprise version" installed. Also, if your company has installed "security software" that screws with web requests and/or DNS resolution in order to "keep you safe" it'll treat Firefox's outbound HTTP requests like they're coming from a script and treat them with heavy suspicion (which slows everything down and can even result in timeouts).
It's super annoying. This is the state of things at my workplace and it's ridiculous.
That isn't an effective solution on a corporate network that depends on internal name resolution.
And FWIW I run into the same DNS-like issues with Firefox on my home network, which uses Cloudflare's 1.1.1.1 DNS. Perhaps it is something to do with the way my company provisions my machine, I don't know. All I know is that Chrome has no issues. I've made efforts to track down the issue but it's ultimately not worth my time.
Me too actually... I haven't spent any time trying to actually track it down. Unsure if it's the internal DNS servers, the Squid proxies, or what exactly. It doesn't seem to happen at home, but I don't have much runtime on FF at home yet.
I know this is contentious, but I think the problem might be that you're using a razor-thin device with extreme thermal throttling problems. I experience no such slowdowns on a 2011 Acer econobook with an i5 and 4gb ram.
How about a sane Profile manager? This is something I use heavily on Chrome. The "about:profiles" UI doesn't cut it.
Also when I use Firefox on my non-work Fedora machines (and this is going from Fedora 21 to 29) it seriously chews up memory and CPU. Facebook (yeah, I know) is rendered unusable after about 30 mins. And as other post(s) have pointed out, video playback can be a bit janky. These machines aren't slouches either, i5 ~2.5Ghz 4th or 5th gen, 8GB RAM, SSD, and dedicated graphics cards.
I'd love to switch, but these are things that are deal breakers.
> How about a sane Profile manager? This is something I use heavily on Chrome. The "about:profiles" UI doesn't cut it.
What's your thoughts on Multi-Account Containers? I think it's better to use if your use case for multiple profiles is to keep sessions separate (i.e., keeping work and personal logins to clearly identifiable tabs / windows)
It's not as thorough compared to switching entire profiles but sometimes the above is sufficient. (ex: your history, add-ons and bookmarks are still under the same profile)
I've always used the -ProfileManager command line option in the past and shortcuts to "-P <profilename>" when wanting to launch/send actions to sessions. (didn't know of about:profiles UI, neat)
I haven't played with MAC (Multi-account Containers) yet but I understand they're a lightweight version of Profiles that can co-exist in a single browser session and have a bunch of neat features (like auto-binding certain sites to certain containers when visited). Would recommend trying out if you've been dissatisfied with the previous approach.
My money would be on you having an ATI graphics card and that being a good bit of the graphics jank.
2013 intel chip+integrated graphics runs like a champ and with lower memory usage than chrome (though I am blocking all the ads, which does significantly change performance everywhere).
They're nVidia 750 Ti's (Asus). Sure, not exactly cutting edge, but work just fine on my Windows 10 machine with a similar spec (which I also game on and get decent frame rates even at 1080p resolution). Chrome is perfectly fine, it's always been Firefox that's the let down, even with the new Quantum releases. Trust me I'd love to abandon Chrome, but it just isn't realistic from a productivity angle.
Yes, I'm pretty sure this is what they count on.
Everyone who is knowledgeable or motivated enough, did already change. The rest obviously didn't so even if a part of those will change, they can keep on pushing it through bundling or advertising to regain a part of them. The bulk will just stay. This is the "internet program" for them.
Google Chrome (desktop) does not contain ads/"suggested content" on the New Tab page. For that reason alone I'll continue using Chromium on desktop until they actually pull the trigger and remove/neuter uBO.
From my experience, it seems that those who used the Internet in it's earlier years before and around Firefox 2 dominating the scene, have a much, much easier time going back to using Firefox again. Whereas those that first jumped online with Chrome was dominated, simply will never switch.
What extensions do you have loaded? Done anything else to customize it?
I've been running Firefox for over a year now on my Kubuntu desktop and haven't had that issue. Not with pre-Quantum or post-Quantum Firefox.
Also, some tabs will continually use more and more memory due to poorly-written JavaScript. That's not really Firefox's fault but Chrome seems to be much more aggressive in shutting those sorts of things down (which can be annoying because if you go over that threshold in your own JavaScript it'll stop working and you'll get ambiguous resource errors in the console).
My tabs might be using a lot of memory though (React + Redux gets reloaded an awful lot).
Regardless, I’d expect my browser to be smart enough to know that allocating even more memory when the system is out and it already uses 12GB is probably not desireable.
Yep, to be honest this is probably me. I use UBo, but I'm willing to make do with less effective rule-based blockers as long as the most obnoxious and intrusive ads are blocked.
it's not true even if some wish it were. Firefox isn't as performant, has fewer extensions, and is less polished. not to mention Dev tools are inferior (but typical users don't care about that).
You must not have used any recent version of Firefox. It has vastly better and more useful Dev Tools compared to Chrome, especially in regards to the frontend. Firefox Quantum is also significantly more performant.
I am not surprised. If you look at the direction in which browsers have been "evolving" (or perhaps devolving...) especially over the last decade, especially after Google first introduced Chrome, the message has been pretty clear: gradually hide and remove functionality that helps users take control of how they consume content, and silence opposition by explaining that it's "for your security".
Chrome isn't the only guilty one here; it just happens to be the most user-hostile, maybe because it started the trend (good example being https://news.ycombinator.com/item?id=7329855), but all the other ones have made similar decisions. Firefox made extension signing mandatory (many people think Mozilla is benevolent, but that doesn't mean their views will continue to align with yours), and more recently IE, which could be said to have been the last reasonably popular browser with a per-zone configuration and site whitelisting/blacklisting feature by default, was deprecated for the far more dumbed-down (and now becoming even more Chrome-like) Edge.
But as long as you can still install a custom CA and set a proxy server, you're still in complete control over the content your machine receives; there have been many changes to frustrate that (first HTTPS, now DoH --- to protect, not just from attackers, but you), but it is still possible to MITM and control your experience. There's been a strong opposition to them ostensibly for "security" reasons, however, the way things are going, you will give up your freedom and security.
(I'm a long-time Proxomitron user. It's far more fine-grained than DNS-level blocking, although I also use a HOSTS file, and I can do more than just block. The best part is, it works for all browsers, even the ones embedded in other apps.)
As always, I think it's appropriate mention Brave here. They are very much going in the opposite way of other browsers. They've now for a long time had features such as being able to change scripts/third party cookies/tracking/etc on a per site basis with a single click. As an example, you can get rid of all bait and switch paywalls (such as NYT) by simply disabling scripts. In Brave that's a two click change - Lion Icon -> Toggle Enable Scripts button.
They've also added features like built-in TOR support. Right click a link and one of the context options is 'Open link in private window with Tor'. You can then change 'identities' (Tor exit node -> IP/country) with a single click as well. Absolutely awesome for geoblocked content or even just seeing what users from other countries would see. Kind of fun to play 'Google Roulette' and perform the same search from various places around the world.
And obviously there's built in ad blocking and all that good stuff. It's like a Chrome that runs absurdly fast (in part because all the rubbish is removed) and has put privacy as a #1 consideration. Of course the downside is that, as mentioned, it uses Chromium as the renderer. If Google starts bringing this stuff over to Chromium, which is probably a non-zero possibility, that's going to pose major issues to browsers like Brave.
Incidentally I was also a long time Proxomitron user!
Maybe it's unfair, but I can't help but think of Brave as just a scheme to push their cryptocurrency. They're removing the website's ability to monetize and replacing it with their own system. That seems gross.
I use Brave and have never bought any cryptocurrency, nor do I intend to.
I think it will be a net positive if every targeted ad company stops using that model, so Brave shutting this down is a huge plus for me.
Note, I used a similar setup in chrome and Firefox that some might find gross as well as it removed the ability for websites to monetize me using ads or data from surveillance.
No, certainly not. I want an Internet with lots of business models including paywalls and large players, but also OSS, amateur bloggers, publicly funded sites, etc.
I remember an Internet back when there were only university sites and personal sites. It had a lot of flaws, but was better, I think than our situation where our data are sold and resold with no way of customer control.
You're not wrong about what it boils down to, but, I'm all for it. I didn't mind ads, generally, until I did. Until they started hijacking your browser, redirecting you, doing popups. Ad people have shown they cannot be trusted on the whole. What Brave seems to be doing is doing their own 'safe ads' system, and sharing the money. It will likely be less for the content creators, but as a 'at-my-wits-end-with-ads' Brave user, I just don't care.
If you look at the benchmark results, aggressively stripping ads and tracking can result in pages loading twice as fast and hours of extra battery life. I'm happy to pay money to view your website, but modern web ads are a ridiculous and punitive payment scheme.
What you're doing is the equivalent of walking into a store, picking up $100 of merchandise, slapping $10 on the counter and walking out because "I'm happy to buy your products, but your prices are ridiculous". It's not a very moral position to take.
gorhill makes a moral argument on the other side of yours [1]:
That said, it's important to note that using a blocker is NOT theft. Don't fall for this creepy idea. The ultimate logical consequence of blocking = theft is the criminalisation of the inalienable right to privacy.
Your analogy is flawed because it assumes the customer walking into the store retains their privacy and is not charged any hidden costs.
A more accurate example would be one in which the customer walks into a store that:
* data mines as much about that person as possible to profile them
* sells that data without consent to unvetted global buyers
* increases the risk that customer's devices are compromised (drive-by malware installs through ad placements on mainstream sites)
* charges them a hidden surprise fee (in the form of increased data usage, battery usage etc from poorly-designed ad systems).
Presenting naive analogies is harmful to our ability as a society to reason about the costs of systems like this.
I'd urge anyone to have a read and/or follow Ad Fraud Historian to understand how bad / harmful the existing ad-tech ecosystem is. [2]
It's truly awful, and anything that moves the internet away from this dystopian, incompetent and fraudulent monetisation approach to content is doing all of humanity a huge favor.
I'm much more sympathetic to the privacy point of view than the "pages loading twice as fast and hours of extra battery life" is my right so I'm going to remove your ability to make money so I can have your content faster.
The people most impacted aren't the entitled well-off first world citizens you allude to, but second and third world citizens and youth.
The rational choices of market participants do not remove the ability of sellers to sell, but they certainly do create more economic pressure for sellers to find and support better systems.
Also consider that the problem is not audiences vs publishers. It's corrupt middlemen harming audiences (privacy / malware / hidden costs) and defrauding publishers (fake traffic, bots, perverse incentives).
At this stage Brave's system of on-device ML in-browser for relevance and privacy, and opt-in ads seems a much better model, and I'm sure others will pop-up too.
We seem to be on the cusp of much better approaches to funding content, so it's really up to us to start advocating and supporting these new approaches.
Society has created legally enforceable ways to be compensated for copyrighted content. If you choose not to take advantage of those mechanisms, whose fault is that?
What you're describing is clear theft. Blocking ads is much grayer, since there's no real cost being charged by the site. I'd say it's closer as an analogy to say going to a store and using the bathroom but not buying anything.
I mean we would be paid for viewing ads this way instead of being paid nothing as it is now, and losing all our privacy instead.
Seems like a win/win to me. Advertisers would actually pay the eyeballs that should be paid for seeing ads instead of the middleman in the attention economy. The technology in the Basic Attention Token performs all the functions instead.
Maybe you will pay, but the market as a whole has been very clear here. Free wins. Personally I dont mind either. I'd prefer to see ads relevant to me rather within some limitations of privacy.
I feel government needs to set the limits here rather than expect business to self regulate. Nor will the market self regulate as this is more technical than most people are interested to understand, and back to free wins.
I seem to recall that the Safari content blocking API is like the proposed Chromium API, but that doesn't alarm people because
a) Apple is less of an ad company than Google, so less of a perceived conflict of interest
b) Google is perceived to break stuff that is already working
c) Safari/Webkit aren't dominant outside of iOS where they're mandatory
UI in javascript = still to this day it can take 2-3 seconds when opening new tabs or going fullscreen. Plus they released a statement about "observing" Google adblock decisions instead of going hard in with "we will not deprecate webrequest usability in our fork".
Does the webRequest API deprecation also impact Chromium, and therefore Brave as well? Brave seems like it's willing to backfill in changes to Chromium, but I'm wondering how this impacts the browser overall.
I use Brave as my daily driver (with Shields Up and uBlock Origin), but I'm not sure what the actual impact will be when it comes.
The only thing I'm missing on Firefox is when I am on my Macbook, is that fling when using my touchpad. It's essential to any MacOS app and it feels weird not to have it. I'm used to it on Linux and Windows though.
That's in macOS System Preferences. I use it all the time. Go to Trackpad / More Gestures and check "Swipe between pages". You can also control how many fingers you want to use. I set it to three fingers so that it doesn't conflict with horizontal scrolling.
I mean, mobile Safari doesn't allow extensions, and only recently allowed content blockers. Safari also has essentially the same model that Chrome wants to adopt:
> Apps tell Safari in advance what kinds of content to block. Because Safari doesn't have to consult with the app during loading, and because Xcode compiles Content Blockers into bytecode, this model runs efficiently. Additionally, Content Blockers have no knowledge of users' history or the websites they visit.
I see a lot of comments saying to switch to Firefox, or people saying they're
already using Firefox. This is fine, but I don't think people should be totally
comfortable with this. Switching to the one other choice still leaves us in
danger of the whole web being controlled by a few programs. We should encourage
other browser projects as well. I've seen some cool browser projects, but mostly
they use webkit still. I'm not sure what you call that part. The core maybe.
We need more browser cores. I think this[1] is one, I encourage people to share
more that they know of.
Also, Mozilla is largely dependent on Google via their default-search arrangement, so even if it's not Webkit/Blink based, Google still has significant influence.
Another big issue is that if chrome makes it difficult to disable ads such that 99% of chrome users aren't able to do it, websites may simply choose to block Firefox as it would be easy to do so without losing a large part of user base while making sure no users are blocking ads. Right now there are far too many users using ad blockers.
That is rich!!! There are far too many companies selling my private information without my consent. As long as companies can sell user information, it is only fair for users to block access to such information.
Blocking a user agent? Oh no, whatever shall we do? Thankfully, we can still just change it, though I suppose a really insidious web site could use feature checks.
I mean, those annoying “join our newsletter”, “quit using adblockers”, and “allow us access to cookies and all your data or else” popovers went from nonexistent to everywhere in the span of 2 weeks.
This should be good news for Firefox and Safari. I'd be interested to hear from Gorhill on whether uBlock Origin works with the Content Blocker API for Safari and if not what needs to change to make it possible.
Safari has essentially the same content blocking model that Chrome wants to adopt.
> Apps tell Safari in advance what kinds of content to block. Because Safari doesn't have to consult with the app during loading, and because Xcode compiles Content Blockers into bytecode, this model runs efficiently. Additionally, Content Blockers have no knowledge of users' history or the websites they visit.
Development has stopped on the Safari extension due to the kneecapping of its capabilities in Mojave. Maybe the upcoming Chrome change will spark renewed interest in getting it to work to the extent possible.
Yes but as far as I understand it doesn't actually use the native API. The Content Blocker API has benefits from a performance and privacy perspective.
Yes, content blocker extensions are many times more lightweight than old-style blocker extensions (virtually zero negative impact on resource usage and site load speed), can never be hijacked to do anything malicious, and know nothing about you or your web traffic.
I use them on my iOS/macOS devices and the their effectiveness is quite apparent, especially on the iPhone where CPU and bandwidth aren’t as abundant.
I just switched back to Firefox after using Chrome since pretty the day it came out. I switched after updating to 74 and noticing how many features I depended on were removed. I originally intended to just revert some of the changes (using Chromium on Linux) but I decided that amount of work was silly. Switched back to Firefox. After this news, glad I did.
The only particular thing I can find is that hyperlink auditing can no longer be disabled. I doubt this is what you meant when you said features you "depend on".
Google never wanted to have these adblock extensions on their store in the first place, it just turns out that when chrome was released and had zero market share they had to make this huge compromise to gain territory in the browser arena and eventually overthrow Firefox and the competition. And when (not if, when – it will eventually happen) they do that I will jump off from the Chrome bandwagon.
"When" they do that, it will be because they have judged that they have, by then, thoroughly extinguished all competition. If you (the royal "you") want there to exist an alternative browser to which to jump to when the time comes, then consider making the jump today, when your influence might still make some small contribution to the competitiveness of the browser market.
Chrome serves me just fine nowadays. Unlike you, I'm not some kind of Gandhi of the tech scene or anything like that. If it serves me fine I won't stop using it. If Chrome ceases to support adblock extensions then I will go look for some other browser, I would probably change to Vivaldi or anything like that, or even Firefox, they sure as hell will be there.
For as long as you, I and everyone else continues to use Chrome, we effectively support Google's ability to monopolise the browser space.
Google's continuing monopolisation of the browser space will (and is?) leading to a dearth of alternatives.
So, by waiting for the time when Google finally puts that final straw on the haystack which makes this whole thing too onerous to bear - and then you say "right, now I've finally had enough - I'm switching ...."
... you may find that there's nothing left to switch to.
For what it's worth, I've had a great experience with firefox for the past year (since quantum convinced me to give it another chance). All though, it recently basically factory-reset itself. Signed out of sync, extensions gone, custom settings gone. Any one else have this happen?
It's also got a weird memory leak, which I think is related to the pdf viewer. Never really checked in detail or tried to measure.
Yes!!! I've had this happen several times, and it's incredibly annoying! I was just about to comment on this issue before I saw your post (as a "this is a good opportunity for Firefox but they still have some seriously irritating bugs").
It seems like my entire profile (about:profiles) got switched out several times with a brand new one. Exact same symptoms as you - bookmarks, custom UI tweaks, extensions, etc. all gone (basically consistent with creating a new profile yourself). For a regular user, I think that only needs to happen once for them to uninstall and never come back.
Most of the time I've been able to restore it to my old profile, but one time a file was corrupted and I had to start from scratch again. It happened before the extension signing fiasco, and those two things combined led me to explore new options (Vivaldi, Chromium, etc.) although those didn't pan out for me so here I am, still on Firefox and crossing my fingers hoping it won't happen again.
Perhaps relevant: I do have Nightly and Developer Edition installed alongside the regular version. I used to use Nightly as my daily driver, but switched to the regular version after the first time my profile got messed up.
Hmm. My situation might have a weird circumstance as well: I installed a new version of manjaro, which ships a profile by default with manjaro homepage etc. bookmarked. I signed into sync; normal one pulled down. The manjaro one is what got mysteriously applied. Old profile gone. I can only imagine what would be causing it; don't remember changing anything whatsoever. Did you by any chance sign in on a new computer before it happened? My windows was the one that went wonky; linux was the newly signed in. Glad to hear it's not just me, though. I was to the point of thinking a rabid script went crazy and deleted my profile.
The new version that just came out added a new feature where different channels (Release, Nightly, etc.) have different profiles, and for some people it seems to have not picked the right profile for the channel they use. You can try looking at about:profiles. Your old profile might be in there, and then I think you can switch back to it.
Firefox has been doing a few rather questionable things. They had the Mr Roboto add-on which was pushed out to advertise a TV show, they stuck ads in the "Recommendations" [1], they proactively destroyed peoples' bookmarks when they dropped support for RSS, they dropped support for RSS (because, apparently, a decentralized way to track website updates is anti-Google), and they destroyed a lot of useful add-ons.
Waterfox is good if you miss the old add-ons from pre-quantum Firefox. But it is the old model, and one lagging page will still hose the whole browser.
I tolerate it because I need my Downloadthemall extension, amongst others.
Devtools are also in standard firefox. The developer edition just includes new features that haven't yet made it to the current release. The chrome analog is Chrome Dev (https://www.google.com/chrome/dev/)
Do you have some reasons you prefer chrome dev tools to FF? I consider FF to have the more featureful devtools, for example the sheer amount of stuff under the styles tab, you can tweak fonts in browser.
I do remember back when I used chrome there would be new `console.x` features that were chrome only from time to time, but after I switched to FF I never really looked back
Chrome's DevTools handles developing large unbundled webapps better. When your loader is reaching to the fs for thousands of source files for a single page load, Chrome performs significantly better than Firefox. Same goes for Ctrl+P jumping to files and debugging performance.
It's gotten a lot better over the last two years, especially when FF DevTools got support for source maps and TypeScript debugging, but it still lags enough to keep me using Chrome for my job, Firefox at home.
I've been a Firefox user throughout - never made the switch to Chrome.
But here's what worries me, what I'm wondering now: As far as I'm aware, Mozilla/Firefox tried to follow Google for extensions, deprecated their own API for Google's/Chrome's instead.
How likely is it that Mozilla will further "follow the spec" so to speak, doing a change like this for compatibility or whatever?
Seems unlikely. It's in opposition to Mozilla's clearly stated stance toward privacy.
In addition, their reasons for deprecating their old extension API was purely technical. Even then, they were extremely conservative in their deprecation timetable, waiting long past when the old API was massively harming performance. But once they made the decision to replace it, it made sense that they would choose an existing API as a base, not make a separate incompatible API.
But do note that, while Firefox and Chrome share the same base extension API, Firefox has APIs Chrome doesn't. It's reasonable to believe that this new deprecation will just become another thing that Firefox has and Chrome doesn't.
Yes, that sounds exactly right. But I would add that there is a technical reason for the more limited blocklist setup, in that it's much more reasonable to do a simple list check asynchronously than to run JS code.
But to me, that just suggests that firefox might want to consider an additional async API. I can think of many options that are more powerful than what Chrome is moving to. eg use the exact logic uBO needs. Or register a WASM function with no access to anything that would problematic to run async.
Chromium / Blink is open source, we should be doing everything we can to get as many users using non-Google Chromium browsers. We really shouldn't be having to worry so much about Google's conflicts of interests. Brave and Blink-based Edge are looking to be better and better options.
Chromium is still controlled by Google. Exhibit A: this ad-blocking API issue, where Google decisions will affect all Chromium browsers. (Sure, they can maintain patches against upstream, but over time that becomes non-viable.)
If Google's behaviour concerns you, you need to use Firefox or Safari.
This is a strange comment. On Linux at least, Firefox is far easier and quicker to get building from scratch than Chromium. I've done both relatively recently.
I’ve been hesitant to agree with calls in the past to break Google up, but now it’s time. This is Google choosing what’s best for their investors over what’s right for the community. This is Google doing evil.
You're right, and I stand corrected. I mistakenly thought their quote was "Don't do evil" when it was "Don't be evil"[1]. Still... I'll take the downvote because I was wrong, but your clarifying comment did more to educate me on my error than a downvote could.
Apropos of my original comment; Apparently their new motto is "Do the right thing." which I question how they can justify this move to prohibit ad blockers as doing the right thing. So now (to your point) they're being evil and doing the wrong thing.
I think "do the right thing" has just the ambiguity they want. If you're trying to win at a video game, you do the right thing. It doesn't matter if you're playing a good character or an evil character. Same if you're trying to learn to golf. Google is a faceless and unprincipled company that has the profit motive and nothing more. They're shaping up to the perfect example of https://thecorporation.com/ only they're playing a very long game compared to the Exxons and Wal-Marts of the world, and part of it is tricking people into thinking they care about not being evil. I think they might harm the environment every bit as much as Exxon but perhaps it will be in the form of paperclip maximizers. https://wiki.lesswrong.com/wiki/Paperclip_maximizer
DNS / domain based blockers are entirely too coarse in my experience. There are plenty of cases where an ad or a tracking script comes from the exact same domain as real content. A real ad blocking extension can match only parts of the URL, or block certain file types, etc. Ad blocking extensions can also do CSS to clean up the page after blocking the ad, or "display: none" ads that for some reason can't be blocked any other way.
My fear is something like pihole may end up being a temporary stopgap; it can only help with DNS queries that go through it, and so it too may end up in the same position as chrome + ublock origin. For instance, if chrome decides to use a trr or dns-over-https and removes options to change it.
Depending on how powerful your laptop is, you could always install a lightweight version of Linux using Hyper V (assuming you're running Windows), and install a pihole server in that. It might be a bit of a pain changing the dns settings for every network, but maybe someone else knows a way to make that easier.
> Firefox is available on all platforms (including Chrome OS via the Android or Linux app)
Maybe this is a viable solution on brand new chromebooks, but on my Acer R11 both Android apps and Linux containers run pretty poorly. I use an Android app as my password tool just fine, but trying Firefox was a rather poor experience. And the linux container just dogs trying to do anything.
I have Ubuntu loaded on via crouton (to use tools like GIMP or actual VLC), but if chrome gets rid of ad blocking that removes much of the point of ChromeOS for me entirely.
I'm starting to question whether it was such a good idea to make web browsers virtual machines (for Javascript or WebAssembly apps).
It makes web browsers big and complex. I know both Chrome and Firefox are open source, but does that really matter so much when the codebase is so big and complex that only large and well-funded organizations are able to develop it? I'm not sure if it's realistic anymore for a few guys to get together and create a web browser (including rendering engine) in their spare time.
This centralization can lead to censorship, as we're starting to see here.
We also end up with a mono-culture (well, not quite there yet, since there are still at least 2 web rendering engines), which is horrible for security. The Irish potato famine is good example of what can happen when you have a mono culture. These days, you often have to let strangers execute code on your computer, if you just want to read an article or look at some pictures.
And this pops up on the same evening that Google chooses to roll out unblockable advertising in Android System App "Duo" in the form of an unsolicited video message from Virat Kohli, apparently the lead player on an Indian cricket team?
Way to work that Lily Tomlin "Ernestine" vibe Google.
Edit: also, "Oh, we're so sorry that our 'security' update is going to gut something that users love but our accountants hate (for anyone else). This is completely an unintended consequence."
No, $24k gross per year does not even begin to cover the amount of work put in to create and maintain software at the level of Pi-Hole. Not even remotely close. Not even 1/10th.
Pi-Hole should be making $1m ARR at least. 53 releases (going back to 2015), 2,765 issues on Github, multiple devs, supporting multiple platforms, extensive documentation, and on and on.
The top contributer has 500 commits since 2015. Its clearly a side project so $24,000 per year for a side project is very nice especially if you live somewhere with a good cost of living.
I have used a pi-hole based blocker for a while, and I have noticed increasingly ads showing up because they are using same domains, both in-app(iOS) and normal desktop. I've settled for a mix of network wide ad-block via pi-hole and adguard(uses mitm for https blocking) to remove almost all ads. In my experience uBlock Origin has made my aging laptop battery drain a lot quicker.
PS: I'm not sure if I would recommend Adguard, but works fine, but somehow the root certificate keeps me left a little bit dreary about possible future Adguard intentions.
Yes, it's an issue and there's nothing a pi-hole can do to solve the issue. YouTube is the most notable server of ads from content servers, but one can imagine more content providers following suit eventually.
However, you can have your pi-hole up and running in a few minutes and enjoy excellent network wide ad blocking for virtually every service that isn't YouTube.
Also related, I found by Bravia running Android tv would come to crawling halt if I used a pi-hole and loaded YouTube. I did the right thing by buying YouTube premium but still Youtube was doing aggressive pings back to their ad servers even with telemetry disabled. ¯\_(ツ)_/¯
Pi-hole will also catch all kinds of tracking and analytics requests that are sent via your mobile devices, non-browser apps, etc. There's a whole lot more of your data leaking out into the world than just the stuff that comes and goes via ad networks.
You only need to set it up once per network, instead of once per browser, and it is capable of filtering (most) ads even on closed platforms (e.g. apps on non-rooted phones).
The downside is that it is much less granular and harder to set up exceptions.
I'm not the maker of the app not can vouch for its integrity but you get a sense of how much network bandwidth ad and tracking takes up on your network.
This doesn't really explain how a PiHole compares to an ad blocking extension. Both block ads at the network level so your browser never has to download them.
Now they are at the position where they can make the users dance to their tune. Im quite not sure if any decision makers at Google have a glance at HN on whats the reflection on their decisions. They just know that users are hooked to their platform all around (Search, Android, Chrome, Maps, GPay, GMail, etc.,) and they can decide where to be a dictator and where to be a humble servant. In the end, all they care is how much data can they harvest & how to use that data to be more relevant for their service offerings (advertisement). Monopoly at the best level! Also, when Google gives a statement/blog post on their decision, it becomes a news for media & HN. But, the reverse doesn't happen; whats the reaction across different blogs/media/HN doesn't get much visibility.
You have just described business. A for-profit corporation is explicitly required, by its fiduciary duty to shareholders, to do what you are describing.
This is untrue. It's a particularly rapacious Wall Street meme, but a company can do whatever its board dictates, including not being a monopolizing monster.
I think it's time for Microsoft to rethink their decision to base Edge on Chromium.
It's a painful decision, to be sure, but the Microsoft team has to make a decision here: fork or adhere. The immediate benefit of forking and not following Chromium on this change is obvious. However, the cost of the fork will grow over time and it will be in Google's interest to deprecate and replace all of the infrastructure that supports the webRequest API to make it maximally painful to maintain the functionality in a fork.
On the other hand, had Microsoft decided to base Edge on Gecko, they wouldn't be working off a forked codebase whose owner has a perverse incentive to make the fork as painful and expensive as possible.
How many engineer hours, months, years are going to be wasted just shoring up features like this?
Why wouldn't they just patch that part of the code? I'm seriously surprised though Microsoft didn't build an adblock in IE much earlier, that would have stopped google in the tracks.
That part of the code doesn't exist in a vacuum, and it's baked in at a rather low level in the browser. That's how it can block requests! In JavaScript!
Google will begin by deprecating the webRequest block API (that's in progress). Then they will begin deprecating the code that supports it.
Then they will begin removing and deleting the "unnecessary" code. Then, at some point in the future, they may make other breaking changes to the webRequest flow that assume that requests cannot be blocked. Each time Google writes new code that assumes webRequests cannot be blocked, Microsoft must patch that code too.
With each successive change, Microsoft will be forced to make a choice:
1. Support both Google's webRequest API and the Microsoft's, at increasing cost with every merge and rebase where Google has cut off or altered another piece of functionality
2. Adopt Google's API and break ad blockers
Google has a profit motive in shrinking the ad blocking market, and therefore they have a perverse incentive to make their browser worse for users, and more expensive for any forks that want to maintain ad blocking functionality.
Google Chrome Enterprise didn't exist when Manifest version 2 was introduced (and version 1 deprecated), so it's difficult to guess what will happen this time around. But if the policy now is the same as then, here's what will occur:
Chrome will deprecate in a release, approximately 1 year later Chrome will block updating of any extensions using the old manifest, and after approximately 18 months they will remove any extensions from the Web Store that have not updated, and approximately 2 years after the deprecation later the code to support the manifest will be removed.
Well, now is as good a time as any to switch back to Firefox. They've managed to come out of a long stretch of mismanagement, technical debt and security issues to a much better place.
Running Chrome without adblock is a security risk. Google has turned a blind eye to malicious ad behavior for their own profit, and this is going to lead to infected corporate networks.
Microsoft might just have the horsepower to keep up with any particularly shitty changes that Google might potentially try to push through, better than some of the other Chromium skins. At least if they want to.
Unless they backfill it... I can see Microsoft going all out with Edge Chromium and Brave also trying their best. Both have manpower to try overcome Google's decisions on Chromium with their forks.
For those of you not familiar with what is being proposed ...
Chrome engineers want to replace the ability to block any request with a standardized ad-block functionality based on a list of provided rules and inspired by Safari. In other words the ad-blocking functionality becomes built-in.
However this facility excludes dynamic capabilities that plugins like uBlock Origin or Privacy Badger need.
Also I’m speaking with first hand experience in this field: the ad-blocking functionality provided but Safari sucks and can be easily fooled. That there are publishers that don’t implement anti-ad-blocking measures, that’s only because they either don’t have the resources or because they don’t want to piss off users.
And at this point uBlock Origin is by far the most aggressive extension out of the popular ones and the nightmare of advertisers, not only because of its capabilities, but also because it doesn’t have a commercial entity behind it.
It is no accident that Google is hitting these extensions.
Switch to Firefox folks. The grass over here really is greener ;-)
The Looking Glass extension, the whole deal with pocket, and Studies being on by default and allowing them to push changes remotely with no user action.
I don't use Firefox currently, so these are just what I recall from HN posts.
I love Pocket, I love seeing them experiment with such services and I don’t get what the big deal is.
Sure they could have pushed it as an extension instead, but note that nowadays Mozilla owns it so the issue is irrelevant.
The source code for the Looking Glass extension was open source and was a mistake, they are only human.
Nowadays every platform maker has the capability to push changes via updates. This increases security for regular users and it’s what the default should be. I agree that there needs to be a way to turn off these updates for power users.
No, those issues haven’t violated my trust in them and when you compare it with what companies like Google are doing, the double standard should be clear as day.
I hope that this makes browsing with Chrome so bloated with ads and tracking that it is faster and efficient to browse with browsers that allow users to get rid of unwanted cruft. That may be the selling point to switch from Chrome in the future.
I've been going back to Firefox for 6 month now, both on desktop and mobile. I'm a developer so I still need to use Chrome for testing, but I'm never going back to it as my main browser, for obvious reasons. There is an obvious conflict of interest going on here, and I'm not going to give Google any more money.
Firefox works fine now, didn't get any issue whatsoever. It used to be slower than Chrome, yes, but it's not true anymore and it uses way less memory than Google's browser.
That's why I don't trust Google with anything, Golang, Goggle PAAS, whatever, they'll go full Oracle in a few years, mark my words.
For work stuff I rely on multiple profiles for different client work (I use about ten of them...so far) and so I stick with Chrome.
If Mozilla didn't have such a demented mechanism to operate with multiple profiles then I'd happily switch from Chrome. This is a feature that should be upfront in the UI, not hidden behind about:profiles and it's generally janky UI and behaviour. This needs to be a first class citizen, not what feels like some half baked afterthought.
Sorry but multi account containers aren’t even close to the same thing as having separate profiles. This is one of those things for me that keeps me coming back to chrome.
I switched to the Brave Browser a few months back (learned about it via Coinbase Earn). Its a great alternative to chrome with solid ad-blocking & I got $10 out of it.
This makes sense from the viewpoint that google is entirely beholden to their shareholders and has to maintain their insane growth. That doesn't make it ok.
It's a shame all these companies follow such a predictable pattern. Any resources to help actively ungoogle? I've switched search, email, and browser so far, and will be pushing everyone I know to avoid. Long time coming.
It's mostly a thing for companies that deploy Chrome in their images and want to retain control over it (group policies). It's not for your common end user.
I found this post highly editorialized. Here's the actual response if you want to understand the context, instead of simply seeing bits and pieces with a journalist's spin.
Also, the headline seems mostly incorrect. The gist of it, as I understand, is that Chrome is enforcing a migration from the webRequest API to a new declarativeNetRequest API. The latter API doesn't currently have all the capabilities of the former, which is important for context blocking extensions. However, features are still being added and the team states that they are interested in more feedback from extension developers.
And if those features happen to never be implemented?
I really think it is a bad practice to force adoption of a new API that can't fulfill the same functions as the previous one.
Nobody would be here talking about this if they just made the new API without breaking add blocking (useful add blocking), and then forced people to move.
Even more so given that it looks really bad when your main goal is to serve people ads.
> I really think it is a bad practice to force adoption of a new API that can't fulfill the same functions as the previous one.
I agree, which incidentally is the reason I switched from Firefox to Chromium a couple years back as Mozilla dropped XUL extensions for their own development convenience, toward an ideal of cross-browser extension compatibility, and in hoping to win over Chrome users by making Firefox more alike in both looks and performance.
How many of those motivations have changed, and do we know that Mozilla aren't actually going to follow suit with a similar restructuring of APIs some time after such changes have settled into and been normalized in Chrome?
I'm tired of major corps doing this with their APIs, either the new API is a drop in replacement for the old API or you're doing shady shit in the guise of "it's coming! soonish~"
This sounds like a wonderful gift from Google, to Mozilla.
As you're deciding what browser to move to, consider using two browsers: Firefox for any sites you need to log into or otherwise identify yourself, and Tor Browser for everything else. I've been trying this a while, and it works pretty well.
I'm not a fan of antitrust as a tool to punish tech companies, but I think there would be great benefit in cleaving Chrome off of Google. The tension between Google's business model and what is good for users is just too great.
I use Firefox with NoScript and found that leaving most domains disabled by default eliminates nearly all ads. I have to fiddle with enabling a few domains on a new site, but if the site is one of those where it uses a million 3rd-party "X as a service" domains, I just leave.
This is the reason I quit using DigitalOcean: every page on their web site required me to enable 5 new domains in NoScript. Other VPS providers just required 1 or 2 enabled.
Or, if the site is coded so that it only works after enabling ad-looking domains, I leave. Vote with your feet folks.
Yes and no. With uBlock you can allow ads at the site level, the problem is that half the reason to block them is the tracking they do, even if I like I site I don't like it enough to unblock google's spyware.
Chrome is faster!
Yeah but FF blocks ads, trackers, analytics and social media buttons. Basically half of the shitty internet. That saves some time I would think.
Will these changes affect the pure Chromium? I really don't want to move to Firefox, all my settings, auto-fills and stuff is in Chrome, and Chromium seems like a better alternative if it still supports ad-blocking. I know that Firefox has improved a lot over the years, but I still have a few too many issues whenever I try using it, whereas Chrome and Chromium "just works".
Is this a chrome API or chromium API? For the later case we have Chromium Edge(https://www.microsoftedgeinsider.com), which is just as good!
If it's chromium API then Microsoft may actually end up taking a fork off chromium codebase right away and decide to keep that alive.
This is a good move by Google to promote Firefox. We should use this opportunity to inform laypersons about Firefox and the freedom it provides with ad blocking (which includes protecting oneself from malware too).
Mozilla should use this for a PR move (even though the bulk of its revenues come from having Google as the default search engine in many geographies).
I've been a Firefox/Safari user for a long time now because of concerns about Google, but I don't really get what this outrage is all about. As I understand it, Chrome is moving to the Safari rules-based model? Can someone explain why this is such a bad thing? It's not like Chrome is stopping all ad blocking.
Hoping Microsoft keep the APIs enabled on the new Edge-ium. Solid browser. If they port it over to Linux (they're already doing OSX) and keep the APIS required for ad-blocking enabled I see it picking up speed very quickly.
Opera has it built-in to the browser so I think they'll be safe too.
Wonder if this will apply to Chromium or just Google Chrome?
Might be a silly question, but is the block list per extension or is it a global block list? Ie, if each extension can have 30k rules, and the current block list requires 70k rules, could there be 3 extensions, 2 with 30k rules and 1 with 10k rules? This would surely be an inconvenience, just wondering
This just means that ad blockers will need to work with a natively-installed proxy.
Remember, with https you can install a certificate and man-in-the-middle yourself. This is how tools like Fiddler and Charles allow inspecting and debugging HTTPS.
In this case, a natively-installed proxy can deny requests to blacklisted domains.
At the end of the day, it's their browser, their rules.
The only way this could be changed is through (EU?) law, but i'm not sure what would be the basis of such law. It would be like forcing every car manufacturers to implement free HEPA air filters in their cars.
This will probably get lost, but Raymond Hill confirmed that he will keep developing uBO and uMatrix for Brave (and any other Chromium-based browser that exposes access to webRequest) after Chrome deprecates webRequest for non-Enterprise users.
more reason to use the Brave browser instead of Chrome. Essentially the Chrome browser without the Google bindings. Use the crypto features or not, it has been working for me and my mac runs way more cooler when i had been using Chrome.
Well, I find this to be great news! Now I have a great way to convince my non tech friends to switch to Firefox. "It lets you install a thing so you don't see ads anymore, unlike Chrome where you have to pay for it"
Apparently, some misguided folks seem to think there is room for more than one browser in the world, and a non-Google one to boot. If you smell the scent of roasting flesh, those heretics are being burned at the stake.
Will this affect just Chrome or will browsers built on top of chromium also get manifest V3? If so, I assume they can implement the old web request API themselves to keep as blockers working?
You could fork Chromium, but then Google will prevent you from watching videos like netflix and amazon online, which means no one is gonna use this browser
I think this is a positive thing. Somebody has to accelerate FF popularity, and this is a start of a big push. Combination of this and FF efficiency equals more users on FF.
That's ok Google. You may not force me to look at your stupid advertising. Plenty of decent browsers out there that take my privacy seriously. You just shot your selves in foot
The Android Firefox browser is great. Not only does it do everything you expect a modern mobile browser to do, but also supports extensions, including uBlock Origin.
Can't use ff on linux unfortunately. It has this weird bug where it clicks on the first item automatically if I right click on anything. It's a super annoying bug.
google slowly showing it's grip with it's android 'open' model falling over and it's chrome now going corporate... :')... soon you will need to pay-per-search not to get only advertisements and nonsense results. enjoy your internet ppl..
Slightly off topic, I'm finding it very disingenuous that people in this thread(and any other threads that come up about Firefox) have actually had significant performance issues with Firefox. No bugs that are noteworthy, sans that fiasco with the expired certificate that disabled addons briefly.
I've been using Firefox since "quantum" on both my MacBooks(one of which is old AF) and on Linux. I've yet to have problems playing video, streaming, or anything of the sort. I keep tons of tabs open. I just can't really say anything wrong about Firefox at this point.
It was once the case that Firefox had significant disadvantages in contrast to Chrome, but now the only reason I have to still keep Chrome installed is when work forces me to use some Google-proprietary page that doesn't work in other browsers.
If you had problems with Firefox 2 years ago, try it again before bringing up performance when people are considering it as an alternative to deleting Chrome. The more people who uninstall Chrome, the better.
"Disingenuous" has become a way of insinuating astroturfing, which the guidelines ask HN users to refrain from. Please don't do that here. If you have better information than other users, that's great, but please simply provide it, so we all can learn. Attacking the other users' integrity poisons the well, and is mostly a cheap side-effect of the irritation we all feel when encountering views we disagree with.
Real astroturfing exists, of course, but is far rarer than insinuation. If you or anyone are genuinely concerned you might be seeing it, email hn@ycombinator.com so we can look into it.
These are just a few issues that I'm directly involved in and suffer from every day, and into which I've sunk over 100 hours profiling and reading code of, simply because they make it an absolute pain to use.
Provided this is anecdotal evidence, in my experience it's the opposite. I gained an enormous amount of performance-related comfort after switching to Linux Firefox from Windows Firefox/Chrome (where a JS-heavy web site that I use frequently would cause memory overflow and would stall the laptop). I've not experienced any problems with CPU usage since. I've not experienced the other problems either, though I don't use WebGL. You make it sound like Firefox is heavily broken for everyone. That's not the case.
I can't really speak to the Windows (FF/Chrome) -> Linux (FF) switch, but I have seen so many Windows setups that 100%-CPU-loop for unknown (likely browser unrelated) reasons that I would not be surprised if people considered switching to what I experience now a major improvement :)
> You make it sound like Firefox is heavily broken for everyone. That's not the case.
I didn't say it's unusable. But these things matter. If you get constantly annoyed by such things and you know that with Chromium this annoyance would stop, it is no surprise people use Chromium a lot and claim it's "faster". If we want them to be happy Firefox users, this has to be fixed.
Because it's still happening, also on Quantum. I still see the 5-15% CPU usage on idle. If you strace Firefox, you'll see the syscall spam, even when the browser is apparently fully idle. This consumes a lot of battery power because it prevents the CPU going to sleep, even when you're looking at a still screen in Firefox (compare to gnome-terminal as mentioned in the bug report, which stops spamming as soon as nothing moves).
My laptop lasts only half as long on battery when Firefox is on and showing a still screen.
I've certainly had something equivalent to the first one in Chrome frequently, seemingly when it's waiting for an unresponsive child process, often when closing a tab. It won't lock up as the UI seems to be separated from the other processes well enough to allow it to keep rendering loading animations etc. However every part of the browser will stop functioning including showing any tab content or opening any menu, for 10s of seconds at a time.
I've also had Chrome destroy my browser profile completely many times. It will start up and say something like 'your profile was unrecoverably corrupted and has been deleted'
Both are very complex pieces of software with lots of bugs.
Often, closing a tab in Chromium simply does nothing.
Chromium has lost my tabs multiple times (not by profile corruption, but not restoring them on startup). Firefox hasn't lost my tabs for over 10 years -- one big reason I use it (it did lose them ~5 times but I could always restore them from the backup file in the profile directory).
I wonder if Chrome is using Sqlite under the hood, I know Electron apps will lose their data entirely if any bit of the database is corrupt (Sqlite doesn't handle inconsistent states well :c).
On a slightly different tangent, you likely have a failing disk or RAM.
Can confirm. For me also sometimes input "buffers up", not showing up for over a second, and then appearing at once. I attribute most of that to the first bug in my list for now, as it freezes everything.
I’m guessing most people don’t suffer from the first issue because they don’t have 800 tabs open? The bug reports were very nicely done, however.
I’d recommend using bookmarks instead of keeping tabs open and unloaded, but I do think that insane numbers of tabs should be something that Firefox can handle.
I've had a few instances of firefox on Mac and Linux starting to spin up the fans, and show high CPU usage, and I've had trouble pinning down which tab was causing it.
Is about:performance the best place to do this or is there something like Chrome's task manager that shows CPU and memory usage per page, and per extension?
I just switched to Firefox today after a few years hiatus, and I agree that it is hugely improved, and rendering speed is at least on part with Chrome if not faster. They've done a great job with it, and the dev tools seem greatly improved in the developer edition as well.
The issue I have with firefox on both Linux and Windows is not speed, but stability. I think it's the process model. Windows freeze if I open too many tabs, and is an issue I've not really had much with Chrome (though it had its own issues). I've noticed it a few times today as well. I wonder if that's the speed issues people are describing?
With that said, removal of ad blocking is a deal breaker for me and Chrome will no longer be my default browser.
I'm wondering if Google degrades performance on its properties, like youtube and gmail? I use Chrome for GSuite and that's all. I'm fine with FF's performance on YT, but I'm not very picky about such things.
How many tabs do you consider to be too many tabs? Most of the time I open up to 10 tabs, and Firefox is performing fine. But I have also seen people opening 40 tabs.
At home I have like 60+ tabs open. And at work I have > 100. Firefox is pretty rock solid now with many tabs on windows. Chrome can’t cope with that. Or it couldn’t a year ago when I switched to Firefox. The tabs are mostly blogs and things I’m slowly reading or such. I just open more than I close :(
I sometimes have hundreds of tabs open (Firefox on Windows), and a feature that I'll leverage if things start chugging is that, upon starting, Firefox will only load the contents of a tab if you put focus on the tab. So if I open a ludicrous number of tabs, I can quickly restart Firefox (just timed it; took six seconds) and all the tabs but the one that last had focus will be unloaded. I believe they're working on making this more automatic by actively unloading tabs if certain criteria have been met, but no convenient heuristic here is ever going to be as outright effective as this. As a tab addict, I can live with it. :P
I'm currently over 1000 tabs in Firefox. Most aren't loaded in memory. I know this is terrible, one of these days I need to KonMari my tabs and start using bookmarks again. Thankfully I'm not a hoarder of non-digital goods.
Either way, I only get bad performance from a few tabs at a time, usually a Fandom wikia or the pinned Gmail tabs. The first can be solved with a few seconds on about:performance, and the second well... after reading this article and Gorhill's reply, I opened a Fastmail account. We'll see how it goes. But if they're really planning to put all users at risk with an automatic update, Google has finally crossed the evil line.
I sometimes creep above the 1000 mark as well. At 250 right now. I recommend the tabstats extension; it'll especially help when you discover that you've opened the same url in 12 tabs, which is easy to do when the tab count gets that high.
I also recommend periodically declaring tab bankruptcy. It feels so light and clean! For a little while, at least.
You, sir, are a gentleman and a scholar. I'm under 1000 now! (ok... 999) Deduplication of tabs is pretty useful for the Jira issues (I usually follow links from Slack). If only Jira pages would load faster.
Tab bankruptcy scares me. It's hard to know how much of it is an action I still need to take. Some of it is research I don't want have to find again, but I need to take the time to bookmark.
My firefox session is persistent (tab are restored when I open firefox again). Apparently, I'm around 500 right now, which is a bit above average for me (I usually stay around 300).
Firefox's lazy loading is very good, which means I only have 10 to 100 tabs that are actually loaded. Firefox is still fast in these conditions. Since I use tree-style-tabs, most of the tabs are neatly sorted in trees by topic.
The fact that people do this should not be viewed as a personal failing on their part. This should be seen as an indictment against bookmark UIs. If bookmarks were friendly to use, people would use them instead of tabs.
Ive never gotten anywhere near 500, but if a legitimate number of users are needing 500 tabs open that's a perfectly reasonable use case for a browser that fails to support it.
It's kind of like saying "oh no, people are trying to do x with our platform but it's not built for that..." Well, now enough people are doing this, so it's time to change up your UI / performance.
And performance (iMavc v.17 Retina) is usually accetable, at least initially. Somewhere between a few hours to days in, ahard kill becomes necessary.
But; On Android or Linux, both older machines, even with just a tab or two open, within recent weeks, performance has been abysmal. Several other options (console browsers, Dillo, Konqueror, Surf) are at least responsive for very light browsing.
I've wiped Chrome and Chromium from Mac and Linux. That's not possible on Android, and though I'd much prefer avoiding it, I'm typing this on Chrome/Android.
(The question of why I've got so many tabs open, or how to manage/adjudicate them, is ... a longer post. TL;DR: tab management user state management, workflow, and ergonomics all suck.
And Chrome is far worse about this than Firefox.)
It is absurd to suggest that anyone needs more than a few dozen tabs. Once you get to the point where you're leaving tabs open for more than a couple of days, you should recognise that you're misusing tabs as a brittle form of bookmarking.
(In my opinion, Firefox and others could solve this by blurring the lines between bookmarks and tabs with some clever UI.)
More like 40. I usually try to run < 10, but sometimes it gets out of hand...I tend to have multiple "tasks" of research going on at the same time. Like if I start researching at home at night I'll sometimes leave it open through a whole workday until I can return to it at night. Then multiple things at work, etc. So yeah...probably my browsing habits that may need to change.
You should try the Tree Style Tabs add-on for Firefox. You may need to tweak some of the settings a bit, but once you experience the power of this extension I guarantee you really won’t ever go back to Chrome.
Another gripe I have with Firefox (which I only use on the Lubuntu laptops I have) is that when it wants to update it insists that you restart it and won't let you open another page until you do.
Huh? This never happens to me and I usually don't restart when it nags me to for weeks, because it has a tendency to forget the previous session on updates.
It definitely does it. I'll be looking in to some issue (because that's 90% of using Linux on a device) and suddenly I won't be able to open any new tabs until I restart the browser.
Firefox has several experience-breaking bugs for me.
One where on websites that don't explicitly set their form background to white, which is the default on most browsers, Firefox paints them the color of the user's desktop theme, even when the theme is dark. Black text on black background. This bug had been known about for almost 20 years and is marked "won't fix" because the devs still assert that universal user theming of websites is just around the corner. The bug manifests on mozilla.com
Another is that hard refresh sometimes flat out doesn't work. I'm a developer who occasionally makes changes to frontend code, and sometimes Firefox just decided that it's just going to cache everything and ignore what I tell it to do. Ctrl-refresh, shift-refresh, clear cache, all do nothing, I have to killall firefox and restart the browser to get my updated web page to show.
It's the default in Firefox for Linux. It's a browser setting, not an OS setting. And in order to turn it off you have to do really obscure stuff in the Firefox launch options, it's funky and makes no sense.
I am unable to reproduce either of your bugs on mac. I've never had firefox ignore a "force" reload in over 10 years of using the browser, and despite having the "dark" theme running, going to mozilla.com shows a white background just fine.
Yeah, because it only looks for GTK themes, not whatever theming engine windows and mac desktops have, which is further evidence of the absurdity of their wontfix explanation.
I've had significant performance problems with Firefox recently. It's always felt less snappy than Chrome, but it got much worse these past few weeks -- every time I try to close a window, it would freeze with a busy spinner for 3+ seconds, often 5-10 seconds.
I did a profile refresh last night and I think it's resolved the primary issues (despite retaining all my extensions). If it hadn't, I would have seriously considered switching to Brave or Chrome.
Ideologically, I believe it's very, very important to support Firefox, and I also think the work they're doing with the Stylo, Quantum, etc. is very exciting, but performance issues with the UI itself got me off of Firefox for years once before, and almost did it again recently. (The first time it happened it was shortly after they started using sqlite, and I suspect that both of these instances were sqlite/profile data issues.)
From my own personal experience, UI speed (partly influenced by "bad" profile data) is the area where Firefox is failing, and the only thing that keeps me from unabashedly telling everyone I know that Firefox is better (user-experience-wise).
I did say I have performance problems (in addition to other rendering/UX problems) in this thread and I did say I've been forcing Firefox upon myself just this past week, in addition to numerous attempts in the past.
That "you've yet to have problems" means I'm "very disingenuous" for pointing out that I did is simply ridiculous. This is borderline ad hominem and I call guidelines violation.
That a sibling implying that we're somehow paid by Google to clean up their PR mess is plain insulting.
I use firefox as my daily for a few years now, I switched for panorama view, and almost switched back to chrome when the extension was broken, but stuck it out until it was fixed.
Quantum improved it, but I still have lots of issues with Firefox on Linux. To name a few: (For reference I'm running Firefox 67 on NixOS, xmonad, no DE, 7700K, 64GB RAM, RTX2070, only extensions I use are uBlock Origin, Panorama View, but I've run into all the problems below in safe mode also)
* Still slower than Chrome if you use lots of tabs and windows (generally 100+ on 3 windows depending on current jobs). It's much faster than it used to be but I can't deny when I have to switch back to Chrome for some things, I'm surprised at the (slight) speed difference.
* Visual glitches with lots of tabs the tab bar will occasionally spasm, sometimes all the text will disappear etc.
* Sometimes a window will just stop working. All the tabs are still there, but they all show blank white pages, new tabs are also blank. Different windows will work and a new window will work. Restarting firefox, the same window will be all blank pages. The only solution I've found is open a new window and copy the tabs you want to save one by one to the new working window.
* Bunch of websites just don't work in Firefox. In particular several of my governments payment portals don't work in Firefox on Linux but work fine in Chrome.
* Sometimes I get crackling audio from Firefox that can only be fixed by restarting pulse, never had it happen in Chrome.
* Some streaming websites choke firefox out, for example ufc.tv PPVs are sometimes entirely unwatchable in FF, the audio and video fall out of sync after a few minutes, sometimes it can makes other tabs sluggish. Never have a problem with Chrome, will usually just open a chrome window to watch live streams.
I chalk most of these up to Linux weirdness, but I do quite often find myself annoyed at FF considering its running on a moderately well specced machine, and its literally the only app I run regularly that gives me heartache.
That being said I've stuck with FF over Chrome and still prefer it for various reasons and recommend people try it. I just disagree with the idea I see floating around that there's nothing wrong with FF at all and its just as fast and stable as Chrome. It's not been my experience.
Another data point: I haven't had any of these in Firefox on Linux (Kubuntu) except for occasional crackling audio. For me, that happens (I think) after a non-playing video tab sits there for a few weeks, and I can fix it by restarting Firefox. I have ~800 tabs open.
That's the cause of your visual glitches (and the occasional "window stops working"). Switch to Nvidia's 430 driver and that should improve things considerably. It did for me anyway.
Never have these issues with Firefox on another PC I regularly use with an AMD GPU. Ditto for my Intel-only GPU laptop (though I don't use that much).
I've only switched to the 2070 in the last month, previously I was running a 1060 with all the same problems (The "window stops working" bug has haunted me on and off for a year or so, but you may be right that the tab bar glitches could have gotten worse since switching to the 2070)
I am on 418.56, I will look into upgrading. Really, I've been suspicious that Firefox might just not like running with no desktop environment, but I feel like it shouldn't make a difference, are you running a DE?
I use both Chrome and Firefox daily on Mojave, and see significant performance differences between the two. My understanding, based on reading[0] and experimentation, is that this may be due to use of display scaling on macOS. Turning this off does help, but makes using external 4K monitors difficult (for example).
People who have different experiences or preferences than yours are not 'disingenuous'. You're hardly going to convince anyone to try FF by calling them liars.
I've been using Firefox since Quantum, too. The one place where I consistently run into problems is the debugger - it's much slower than Chrome's, and sometimes it gets stuck with a "Loading..." in the source view that never loads. I use it... But sometimes I get frustrated enough to fire up Chrome to debug something.
I have never had any performance or memory related issues with firefox, on a modest xps 13 9343 with 4gb ram. In fact, with regular usage, firefox rarely exceeds 2gb usage, and cpu usage is never an issue. Though I have tweaked my firefox quite a bit, I don't think the core issue is with the browser technology. The technology is fantastic and computers are fast. Rather, I see it is a combination of two things: poor web browsing habits (tabs open, websites visited) and the ever-increasing badness of "modern" websites.
2Gb seems to totally reasonable for my habits, but I can't keep to it, there are just always leaks. Eg; Slack or Gmail just need reloading every few hours to reclaim RAM.
Also have a bunch of umatrix rules, but not inclined on sharing those. Regarding gmail, keep a pinned tab open on about:memory and periodically click on GC and minimize memory, that might be a better solution than reloading.
I have been using Firefox, and only Firefox, on the desktop since before Firefox 1.5 (released in 2005) and Firefox has had plenty of performance issues over the years and still have some. The performance issues still reaming after the Quantum release are mostly caused by shitty websites who only optimize their crappy code just enough for it to work in Chrome. A couple of culprits which slow down my browser significantly are Trello and Twitch. Another issue is that 3D acceleration does not work well on Linux so it is disabled by default and often cannot be enabled without compromising the stability.
Before Quantum on the other hand I had to regularly restart Firefox since it seemed to slow down over time and hog all of my CPU no matter if I had many or few tabs and no matter if I closed tabs so I am not surprised many got burned by Firefox and how it could not handle today's bloated websites until Quantum. The UI also got jankier and jankier the more CPU Firefox started using.
Chrome probably has had its own performance issues but I cannot comment on them since I have not used it much other than on mobile.
I use Firefox as my primary browser. I still have performance problems. Today this happened. After a long session using 5-30 tabs at a time, and devtools variously open, Firefox sometimes completely locks up for tens of seconds. This is on Windows 10. It happened maybe 3 times today. I've never seen it happen on chrome ever. This is a recent high-powered machine with fast (for a laptop) everything.
I've been using Firefox every day on the same mbp 2015 with the same addons for almost 4y now. It might be uBO, but either way: FF drains battery. The Energy tab in Activity Monitor is very clear about it. When I switch to Safari (also uBO, fwiiw), fans stop spinning, the browsing experience feels like I upgraded to broadband and I get 2h more out of a charge.
Before quantum, during quantum rollout, since quantum, doesn't matter.
I've just accepted that slightly increased privacy comes at the cost of being plugged in to the mains. That's fine, I accept it, but I hope it's hardly disingenuous.
I tried Firefox again right when quantum landed, but at the time there was a significant delay in opening/rendering pages compared to Chrome - as in 4-5 seconds to open Reddit compared to 1-2 seconds on Chrome (this was in macOS on a 2013 MBP with both browsers running uBlock Origin) and an hour of troubleshooting and following tips on the web yielded no improvements - I don't know what's disingenuous about that...
With this latest change to Chrome it's probably time for me to give it another go. I'd love to switch, but the experience needs to be roughly on par.
There are still issues with Firefox but you are correct about people maybe not knowing. I tried it a few years ago before the speed update and it felt unusable. Nowadays it's a little more sluggish than chrome on certain things (JIRA for one) which perhaps is just a result of people optimizing SPA for chrome.
It is VERY important for people to give things a try with and without a proper ad blocker. Any browser gets WAY slower and buggier when it has to deal with the nutty page takeovers with autoplay videos and javascript running.
Yeah, I have been using Firefox since 2007. I don't experience any crashes or bugs regarding performance or usability. But I have been facing issues with websites which don't work at all or developed with only Chrome in mind. And I have been facing this issue enough times that whenever I see a problem, I have to reopen the site in Chrome to confirm that Firefox is not the problem. I think this is a major problem, and as developers ignore Firefox or any other browser, web will be unusable without Chrome.
My issue with Firefox is essentially one bug where they refuse to fix. In private mode, previously closed tabs are, for some reasons, restorable. This doesn't happen for Safari or Chrome.
That's a feature. Chrome will refuse to remember history so that you can reopen the tab, however it will remember all other data from the closed tab such as cookies. If you log into a website in an incognito window, close all tabs of that website and then open a new tab and navigate to the site again, you'll still be logged in.
It tricks you into thinking that session state is kept only per-tab by disabling a useful feature (history), however the incognito session's state is actually scoped per-window and not properly cleared/forgotten until you close the window.
Firefox knows how to have temporary window-local cookies/site data _and_ history. Chrome doesn't know how to keep the history. It's a missing feature IMHO.
I thought there was some kind of time limit after which closed private tabs are no longer restorable? I'd prefer that to immediately losing the closed tab, if the time limit is reasonable (say, 30 seconds).
I believe this is within the same Private Browsing session, so if you're closing the window (default behavior with the last tab I think) you may not be seeing it.
I actually use this feature regularly... I consider a private browsing session gone when its window is gone, typically. Either way, nothing is saved to disk
Wouldn't getting able to restore a recently closed tab even in private mode be somewhat similar to having your browsing history being kept, even if for that particular browsing session?
I prefer this behaviour of limiting the browsing history per tab per session instead of the way Chrome implements it
Just started Firefox and opened a popular local website.
- the browser address autocomplete felt like I am waiting, while in chrome when I type few letters it shows suggestions like literally instantly
- when I hit enter the website is loaded fast, but still slower than chrome. Like 2,3 times slower. In chrome it is loaded for like < 1 sec definitel, for firefox it is like 2 sec.
Closed Firefox and tried this again couple of times. Both issues are still present.
My one issue with Firefox is the disk usage. Running on an internal SSD, the entire browser still slows to a crawl for 30+ seconds upon opening, regardless of the previous tabs open (although if a video is there, it takes even longer).
I've done all the tweaks and configurations found online, and nothing has made a marked improvement whatsoever. Not going to switch, but it's incredibly frustrating.
Are you experiencing this pause on Windows, Mac OS, or Linux? I ask because my employer has some crap McAfee security theatre software installed (Windows) that causes this same issue for me for any application that loads any large-ish file (or large number of files) on startup. Since Firefox is going to load a bazillion things from cache on startup it (or similar software) could be causing the problem.
I had support temporarily disable McAfee's "on-access scanner" thing so I could test it and the problem went away immediately. Everything was nice and snappy again!
Of course, they turned it back on after the test...
I'm experiencing it exclusively on Windows, as I don't use Linux for web browsing. I don't have any anti virus other than MalwareBytes and Windows Defender, and I've completely turned off all forms of caching for Firefox in a vain attempt to fix this.
I guess maybe it could be Windows Defender? Doesn't quite explain why the Firefox process is the one that is sucking literally my entire r/w bandwidth, slowing my computer to a crawl and inhibiting all OS functions. Defender does this every once in a while, randomly, but it doesn't show up during Firefox's start-up.
It works fine after like a minute of warm-up, but I still think that's unacceptable on an SSD. Still gonna use Firefox of course...
Speaking of extensions, I really miss FireBug in FF. The new "inspect element" tool lacks some very important features like easy XPATH finding/defining which are critical for web scraping.
I only keep Firefox around as a back up for when there's a one-off website that doesn't work properly with Safari. Otherwise Safari is my default. It just performs better than anything else on OSX/iOS.
Is there a way to fix the scrolling? That rubber-band effect is horrible. Usually people get mad when websites screw up scrolling behavior but here you have an entire browser doing it and nobody seems to mind...
The first Firefox Quantum version broke on my work laptop and I never managed to work out what it was. It probably interacted badly with some corporateware, I don't know, but it would just crash on start.
Linux is a third-class citizen for Firefox and its very obvious. Not only is the rendering performance and responsiveness subpar compared to Chromium, but even very basic things like XDG spec support have still not been implemented after sitting on bugzilla for more than a decade.
To me, its incredibly ironic. On one hand, it makes sense for Firefox to focus on Windows because that's where most of its users are. On the other hand, what's the point of using Firefox in such a privacy-hostile OS filled with ads, telemetry, and tracking?
I did give Firefox a _serious_ exclusive try for the past two months. It's terrible for heavy websites. Clickup, PivotalTracker, JIRA, Zeplin - the browser just chokes on these sites and get janky. On Chrome they are butter smooth.
I don't know what to do. I want to use anything other than Chrome, but I also want a sane workday and not have the browser jank my day up constantly.
I would use Brave but they don't have a proper sync system yet. I switch between my imac, macbook pro and macbook air throughout the day, I need sync to be there.
I'd suggest you try creating a new profile (after saving your bookmarks elsewhere) and start from there, installing your favorite extensions one by one. An uninstall and a fresh install could also be combined with this. Check the Firefox knowledge base for instructions on creating a new profile.
If you have Firefox sync on, you can also do a profile refresh from about:support and it will re-add your extensions via the sync. I did this last night and I think the only things I lost were userContent.css (which I'd backed up) and custom search engines.
Yesterday, whenever I closed a window, it would freeze for 3+ seconds. After refreshing my profile, it's back in full working order again.
EDIT: Oh yeah, I also lost all cookies (which is probably a good thing), notification settings, and my custom containers. tbh, the only thing that really matters is that I still have all my bookmark keywords. Everything else could have been wiped and I'd be fine.
This is a legitimate reasoning to keep using chrome. If for a person the benefit of better performance outweighs worse experience with ads, it's perfectly reasonable to use the thing that creates more value to you. I guess it will take more than hard limit on filters to force majority of people away from chrome. And Google is betting on that, for good reason too.
This breaks the site guidelines. They require you not to call names and not to make insinuations about astroturfing when commenting here. Would you mind reviewing and following them?
nah i've had it too, granted it's on a 10 year old Xeon box I got for 40 bucks 4 years ago on an equipment auction at my work, but Asana and clickup are unusably slow, so when I last tried to switch for Firefox I'd just use them in a chrome instance ( where it was actually usable) ,the trouble was I would keep accidentally opening links etc in chrome and before I knew it I was using just the chrome browser.
It might really depend on hardware and OS... we would have to know a lot more about both of your setups to determine why you guys are getting such different experiences.
I have an iMac 2015 and a MacBook (the one with the slow dual core ultrabook CPU). I use the normal edition of Firefox most of the time. I've never seen Clickup or JIRA run faster in any other browser.
It may be related with your screen resolution, I agree that Firefox has serious performance issues on Macs due this bug: https://bugzilla.mozilla.org/show_bug.cgi?id=1404042 [Poor battery life on OSX with scaled resolution]
I currently use Firefox, and recently (likely owing to multi-processing) Firefox eats up 2G of RAM for me[0]. I have ~700 tabs open so that may be a factor as well. Other than this the browser works great for me, and I have always used Firefox as my main browser.
Yeah it was a higher total than I expected myself, but I do use a fair portion of the tabs fairly frequently, as I can witch to the tab I want by typing into the address bar.
Neither does chromium unfortunately, although Fedora now incorporates the out of tree va-api patches for chromium by default. This is one of the biggest things keeping me on chromium, video acceleration in the browser significantly improves battery life.
There is a package for Arch as well, and it was previously in the default package, but they removed it again temporarily. This is keeping me on Chromium as well, but if it were added to Firefox then I'd reconsider.
This is a rather ambiguous and incorrect statement. Hardware acceleration of some things is disabled by default because of a security issue related to the way X11 works (it's complicated but in short, one window can "see through" to another window--even if it's an entirely different app). You can re-enable it via the prefs (forget the name but it's under "Performance") or just toggle `layers.acceleration.force-enabled` in `about:config`.
Furthermore, acceleration is enabled by default for loads of things. It's just some certain WebGL features that aren't hardware accelerated. For example, when you play back a video it's going to use hardware acceleration by default (if the underlying library supports it).
There's also non-performance issues to be wary of Firefox. The Mr Robot surprise extension installation is something Mozilla said will never happen again, but lost them some trust. Not that Chrome is much better in that regard
The HN guidelines ask you not to post insinuations about astroturfing to this site. If you're genuinely concerned about abuse, please email hn@ycombinator.com so we can look into it (which we will). But it's not ok to shoot it as cheap ammunition into an argument—it poisons the well, and is a thousand times more likely to be an imaginary projection than anything else. I've written hundreds of comments about this over many years, so there's plenty of explanation for anyone who wants more: https://hn.algolia.com/?sort=byDate&dateRange=all&type=comme....
There’s a lot of bias in this article title. I’m kind of disappointed in them for it. “Complex ad blockers” would have been just as informative but without the blocker-favoring tilt.
Bonus irony for realizing that a properly configured blocker as they’re favoring with this title would reduce their site’s revenue, being both JavaScript and ad-heavy.
Vivaldi builds on top of Chromium. All Chromium browsers will eventually receive these extensions changes as well unless they fork their codebase. Most of them support extensions via the Chrome Web Store though which means they'll need to adopt Manifest V3.
It's such a shame Microsoft thought it will benefit more from piggybacking on the Chromium monopoly over building its next browser on top of Firefox.
If Microsoft would have donated only 50% of the Edge browser development budget to Firefox, Firefox would be in a much better position in the future to compete against Chrome.
Now, I fully expect Google to end its search contract with Firefox, or pay them much less, too, since it won't feel like it needs to fund its competitor anymore.
Yeah, it's time to unscrew the write protect screw.
I'm wondering how Firefox for Android manages on ChromeOS. For now it's too slow on Crostini. But my cheap Chromebook is not performing well with both Crostini and Android containers. So straight up Linux should be hopefully better.
It's certainly harder for my old Tegra K1 Chromebook.
> I'm wondering how Firefox for Android manages on ChromeOS
I've done this before, it works pretty well, but Firefox for Android doesn't have a great extension experience as far as I've seen, though it could have improved since last I tried.
That's a serious violation of the Hacker News guidelines, which ask you to assume good faith and not to post insinuations about astroturfing or other manipulations.
People need to be free to come here and express their views regardless of who they work for, and not have their integrity called into question for doing so. HN will be much poorer if some users are allowed to hound out other users, however unintentionally. Everything I've observed on HN has shown me that the temptation to assume that the other party in a disagreement is being disingenuous is a million times greater than it should be. We all have to work against that consciously in ourselves.
Actually no! In fact, responses like this make me afraid to post at all.
Because, either I make honest disclosures that I work for Google, when talking about Google software or other things, or I try to be pseudonymous and not disclose. If I do the latter, well, that could have a lot of repercussions, it would be a violation of the social media policy as well, so it's not really worth considering, plus it even just feels dishonest.
But if I do the former, I'm stuck with people reading it like I'm some corporate marketing drone, despite the fact that before I joined Google, I was already prominently on internet communities like Hacker News. I'm trying to follow both internet etiquette and company policy by disclosing that I work for Google, but invariably whenever I speak on something there's this weird conspiracy that I simply must be on the corporate payroll astroturfing; despite that I'm, practically speaking, just a rank and file software engineer.
Also, I am not trying to mislead people. Average consumers are obviously not going to use the enterprise policy features. They should, however, be available on any Chrome installation, if the article is referring to what I think it is. The article claims it is a 'paid' feature, and I don't think there is a such thing.
I don't even understand this accusation. I straight up have been personally using and recommending Firefox since Quantum came out.
As for the frequency of discussing Google stuff... I mean come on. The frontpage of Hacker News is sometimes covered in discussions about Google stuff. I can't escape it even if I want to. It is actually a big downside of working for a large employer; it becomes fairly difficult to disconnect from work when on vacation (as I currently am, actually.)
As I tried to explain upthread, posts like this are poison to HN. We need you, and everyone else, to resist taking arguments in personally accusatory directions.
No, it's not a fair question to ask if someone is dishonestly manipulating conversation because of who they work for. By doing that you turn a conversation into a cross-examination, turn feelings sour, and give the other person a reason to leave. Most people who post here don't want to deal with that kind of aggressive or passive-aggressive confrontation. It's not that they couldn't—it's that it's a gigantic bummer. In that person's position, most of us would just say "why bother" and leave. Then HN would be much poorer for the loss of a good-faith user. I'm sure your intention was to support the integrity of HN discussion by doing this, but what you're actually doing is making this place—which is always very fragile—more hostile. And not just a little more hostile; a lot more.
If you're concerned about HN being abused, you should email us at hn@ycombinator.com so we can look into it—which I promise we will.
> Google made an intentional and specific decision
If by "Google" you actually meant "any of a zillion possible entities within Google, including any number of intrusion- and bot-detection AI's which can be incredibly trigger-prone and - like any sort of self-adaptive AI - go haywire in all sorts of entirely unsuspected and creative ways", then you would have some actual basis for that statement.
If he is not a part of the browser team, do you think it is possible for him to be honestly mistaken?
I am a software developer for a company with less than 70 employees, I lead most infrastructure initiatives and I don’t know everything that goes on in my company and could easily misspeak about our software.
What's wrong with a google employee sharing their opinion on HN? If there are factual inaccuracies, you can down-vote or, more constructively, respond & refute. The OP disclosing a potential conflict of interest seems, to me, as being assiduously transparent.
Ugh. Now I'm going to have to switch my passel of family members, neighbors, and friends who rely on me for tech support over to Firefox.
Not that Firefox isn't going the same way eventually. Firefox crippled the ability for real customization with Firefox 57 and it's only gotten worse since then.
"
It's really unfortunate how many people seem to be commenting here without understanding what the proposed changes even are.
To be clear:
1. This change _IS NOT INTENDED TO GIMP AD BLOCKERS_. Rather, it is designed to make them faster and more secure. (Yes, even despite the limitations that might impact uBlock.)
2. The new proposed content blocking API _is not final_ and can/will be changed.
3. Threatening to switching to Firefox is not helpful and _WILL NOT CHANGE GOOLE'S MIND ABOUT THIS_.
If you don't understand this, please refrain from commenting as you'll only be decreasing the signal-to-noise ratio in this thread and thus making it more likely that the entire thread (including those who are expressing actual legitimate concerns about the limitations of the new API) gets ignored.
If you want to help:
1. Explain a specific use case that the new API can't handle (including a technical explanation of _why_ it can't handle that use-case)
2. Suggest constructive changes to manifest V3's API that will improve its capabilities _while still adhering to the stated goals of manifest V3_
3. If you can't do any of the above, please refrain from commenting so you don't just make things worse
Sorry if I sound aggressive. It just really frustrates me when constructive technical discussions get hijacked by large volumes of unconstructive, uniformed comments. It makes it way harder to get real work done.
"
I know the anti-google bandwagonning is strong, but seriously, I doubt most people actually tried to understand what is being changed here, and it's sad.
This reads just like propaganda, frankly. It doesn't say anything firm, denies reality ("won't interfere, except it will impact uBlock"), etc. Also, according to the article, Google has since confirmed that it isn't going to budge, at all, on any of the issues that the comment you quote (which was from January) is addressed to.
It's not something I would put any credence in if my app or org depended on it.
What we see are the public statements, for public consumption, they are designed to "sell" the changes to the wider public. What we do not see is what is being said in private meetings by officers who get to decide how to optimize the business. So we have to judge not by what is said for public consumption purpose, but by what in effect is being done, or what they plan to do.
This is how personally I see the deprecation of the blocking ability of the webRequest API in manifest v3:
In order for Google Chrome to reach its current user base, it had to support content blockers -- these are the top most popular extensions for any browser. Google strategy has been to find the optimal point between the two goals of growing the user base of Google Chrome and preventing content blockers from harming its business.
The blocking ability of the webRequest API caused Google to yield control of content blocking to content blockers. Now that Google Chrome is the dominant browser, it is in a better position to shift the optimal point between the two goals which benefits Google's primary business.
The deprecation of the blocking ability of the webRequest API is to gain back this control, and to further now instrument and report how web pages are filtered since now the exact filters which are applied to web page is information which will be collectable by Google Chrome.
https://github.com/uBlockOrigin/uBlock-issues/issues/338#iss...