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.
>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.
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.
Disclaimer: ex-googler.
"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.)
Or also, faster ads. The explanation that faster web better compete with native apps and make for better revenue.
(I agree that it is likely not the only factor, but it a reasonable one)
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.
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.
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.
https://en.wikipedia.org/wiki/Safari_version_history#cite_re...
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.
disclaimer: ex-Googler 2013-2018
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!
This is not "surveillance capitalism". It is surveillance in guise of capitalism.
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.
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.
Nevertheless, they offer good products. I'm not sure the negativity is really warranted.
I've yet to hear about a case of anybody being hurt by Google collecting data to them and serving ads?
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.
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.
I agree with GP. Chrome profiles are what’s stopped me switching to FF.
https://news.ycombinator.com/item?id=15688651
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.
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.
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.
Firefox has yet to release either of those features, as far as I know. I’ll switch dev browsers as soon as they do.
It’s not integral to my work, but it does contribute significantly to my productivity.
[0]: https://mozilla-services.readthedocs.io/en/latest/howtos/run...
upd: typo
I never needed anything more than a kick-ass web browser.
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
In that regard, Firefox is not a worse option, just equally suboptimal.
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".
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 keep seeing people write stuff like this, and I just don’t get it.
What’s keeping you? Why haven’t you switched already? I mean... it takes minutes at most.
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.
Here are my notes from when I investigated:
https://news.ycombinator.com/item?id=15688651
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.
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.
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.
> Way better than opening one is FF, one in Chrome, etc
Agreed!
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.
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.
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"
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.
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.
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."
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.
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."
> you didn't have a reply button initially.
HN rate-limits replies to recent posts.
Good luck convincing your product owner of the necessity of that change. Most large enterprises won't do something like this.
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.
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.
Yes we can, go ahead and do it. This is a "Do not do X" nudge not a "Do Y" command.
It is proven by thousands of years of history that humans are best served by negative commandments.
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"
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.
Java didn’t cut it. Neither did Tcl/GTK/Wxwidgets.
Web apps won because of universality and zero touch deployment. Nothing to install, it just works (mostly).
Platform-native apps simply doesn’t achieve the major “it just works” goal, and always havw their own compromises.
There's probably a way to compile a list of such features from caniuse data.
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.
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.
[0]: https://gist.github.com/flibitijibibo/b67910842ab95bb3decdf8...
https://developer.mozilla.org/en-US/docs/Mozilla/Gecko/Embed...
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.
[1] https://github.com/mozilla/geckoview
[2] https://github.com/mozilla-mobile/android-components
[3] https://hacks.mozilla.org/2018/09/focus-with-geckoview/
The truth is the browsers are complicated beasts these days for better or worst and so goes the complexity of building and depending on them.
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.
Google can and does use DRM to block even this from happening: https://blog.samuelmaddock.com/posts/google-widevine-blocked...
Even Mozilla needed to license Widevine from Google to support modern video: https://support.mozilla.org/en-US/kb/enable-drm
Based on the policy evidenced in Maddock's post, it seems impossible to develop an open-source browser in 2019 that supports video streaming.
Of course, I'm one of those who doesn't care much about mass media in general. Addictions could make this shift impossible.
You can also enjoy lots of high-quality, DRM-free content on YouTube, Vimeo, peertube, fosdem and god knows where.
There’s more to internet video than Netflix, Hulu and Prime.
There's nothing particularly modern about DRM.
These responses... Dont support amazon or netflix, twitch? Even FOSS needs to be realistic and see what is a solution and what is not.
A FOSS product needs to support the things people want to do. Otherwise it will not succeed. We should have learned this in the last twenty years.
Without Firefox and its compromises, Google would have 99% market share and FOSS people would browse wuth Lynx or wget
If a website is using Proprietary DRM A, your browser needs to be licensed to use DRM A.
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.
Chrome doesn't have that problem.
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.
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.
It's not a product-making sale point for the vast majority of people.
It's not a completely ethereal concept like privacy, democratic freedom, etc.
That ship has sailed.
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.
At least that's the justification I generally see. See the Criticism section of the Wikipedia article:
HTML5 breaks soft. It's easier to write, which is what drives content. A parser for which there is no content becomes moot.
(I'd prefer far more rigorous document specification. That's not the Universe I inhabit.)
And Google is currently the funding source for Mozilla, giving them a more or less complete monopoly on the future of the web.
Chrome without adblock is essentially ie6, with tabs.
It managed to get to around 20% marketshare[1] before Chrome surpassed it[2] and then also surpassed IE.
[1] https://en.wikipedia.org/wiki/Usage_share_of_web_browsers#Th...
[2] https://en.wikipedia.org/wiki/Usage_share_of_web_browsers#W3...
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.
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.
They are competitors in the hardware space and Apple needs an 'enemy' to help target its pro-privacy agenda against.
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.)
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.
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.
I disagree with you after there.
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).
Chromium is where this change is originating, not chrome.
Amen! I mean, the layout look as if it were something made in the early 2000s
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.