Comment by rkharsan64

6 days ago

People here seem to be underestimating the advantages that Google gets just because of Chrome:

- When you sign in to Google, you sign in browser-wide. Google now gets all of your browsing data, perfect for advertising. (If you ever doubt it, go check out Google Takeout. You'll be shocked at the amount of data you see there.)

- They have special APIs and features that they get to use, and nobody else. Only because they own Chrome. [1]

- They get to move forward with enabling and pushing features that allow for more advertising: see Manifest v3, FLoC.

- Google specifically serves a worse version of Search on Firefox for Mobile. You have to get an extension to get the full experience.

This isn't an isolated attempt. You can see more of the same thing with Android.

- AOSP (the open source counterpart of Android) is now unusable. It doesn't ship with most essential apps, including a Phone app. In previous versions of Android, all of these were a part of AOSP.

- Most third party launchers/stores struggle to implement features because they are only available for Google themselves.

- The signing in with Google thing from above continues here too: you sign in to Google system-wide.

[1]: https://x.com/lcasdev/status/1810696257137959018

> This isn't an isolated attempt. You can see more of the same thing with Android.

Here are more:

- (jumping off of your second point...) Play Services does more than just handle stuff you sign into as a user -- it's also a dependency for everything from push notifications to screen casting. This actively poses issues building competing platforms, in that in order to give developers a path to shipping in your ecosystem you have to provide functioning alternatives to all of those ancillary features. The compatibility issue also impacts user adoption, and then the user adoption and the barren marketplace impact each other... Even the combined resources of Amazon and Microsoft weren't enough to overcome this. (Facebook did, but I'm also not sure forking the OS into a separate VR platform is necessarily the same thing.)

- It also comes with integrity checking, so even if you do find a good third party image, and sideload Google packages, numerous things won't work unless you take part in a dumb arms race that ironically requires you to also root your device. By which I mean a feature that was originally built for banking applications is now used everywhere from streaming services (as an additional layer of DRM) to gacha games (for anti-cheat). This is actually the entire reason I dropped Pokemon Go, personally.

Obligatory link to the excellent Ars piece on this topic: https://arstechnica.com/gadgets/2018/07/googles-iron-grip-on...

  • > Even the combined resources of Amazon and Microsoft weren't enough to overcome this.

    Pedantically, their resources were never combined. They independently tried to compete, and they independently failed.

    For what it's worth though, Amazon seems to be doing tidy business with entrypoint tablets and FireOS, which is a fork of Android, but still one they own.

    Microsoft's exit of mobile was a short-sighted decision IMO. They have the entire office suite. They have windows and Windows has essentially become an app store model too.

    I can easily imagine a future where Microsoft leaned in hard on Microsoft Mobile-exclusives for Word/Outlook/Excel/Teams/etc., bundled it with the rest of Office/Windows subscriptions, and had every office worker in the world carrying a windows phone for their work device.

    I know, I know - everyone wanted only an iPhone. But it feels to me like Microsoft didn't try very hard.

    • They can't properly port the office to android today, imagine years ago, also office in cellphone were never great

    • > Pedantically, their resources were never combined. They independently tried to compete, and they independently failed.

      I'm talking about Windows Subsystem for Android, which leveraged Amazon's app store.

      Though honestly, their application strategy since Windows 8 has been fascinating. Watching them:

      - ship Linux and Android app support

      - replace their first-party browser with a Chrome derivative (and inherit all the PWA support of its parent)

      - ship first-party support for .NET and PowerShell on other platforms

      - ship React Native builds for Windows and macOS

      ...all suggest they're really trying not to get into the same hole as what happened with Metro, by giving you a lot of different ways to build something that can then run on modern versions of Windows -- regardless of whether that's even necessarily your goal.

The problem is this:

- the browser is undeniably critical as everyone's window through which they view the online world;

- the user gains a huge amount of value by a browser being integrated into the OS, webviews in other applications, etc

- browsers aren't really a self funding product

- having a single for-profit US advertising company control everyone's view of the online world, however slightly (e.g. by obstructing adblockers), is Not Good

Splitting it off solves the latter problem but immediately raises the question of how to pay for it. A very artificial arrangement where Google pay "arms length browserco" to maintain Chrome?

  • You raise some very important points.

    Specifically, this one:

        > browsers aren't really a self funding product
    

    I feel the same. I also feel the same about a modern C library and C compiler (and C++, if you like). They are essential to build any modern system and applications. Yet, those are also (mostly) no longer self-funding products.

    What do you think will happen if Google is forced to divest Chrome?

    • Netscape used to cost the equivalent of $100 inflation-adjusted dollars and was only forced to go free to compete with Internet Explorer. Now the genie can't be put back in the bottle, and anyone trying to sell you a browser would become irrelevant the same way Delphi's paid compiler lost out to free C compilers.

      Maybe you could carve out a niche that's willing to pay, the same way C# did before dotnet core. But for a mass product the best-case scenario would be something similar to today's Opera.

      However what it would do is open up the market to competition. Right now Google is spending a lot on Chrome development and Chrome advertisement. Opera and Edge both have given up on their own engines because they couldn't keep pace with Chrome development, and Firefox kept its engine but can't compete with Chrome's ad spend. If Chrome had to compete on a more even playing field there would be more room for diversity and competition. That could be a net positive, even if it makes Chrome worse.

      7 replies →

    • > What do you think will happen if Google is forced to divest Chrome?

      The new Chrome company will struggle for a year or two then Apple will try to buy it but lose out after Oracle submits a higher bid.

      14 replies →

    • Firefox has made Mozilla billions over its lifetime by selling the default search engine rights to Yahoo and Google. Chrome, having a much greater user base, would demand a correspondingly higher fee (probably around $10b a year). Now, the other problem is there is no other search engine to compete with Google at that level, but that might change with independence of Chrome.

      2 replies →

    • > They are essential to build any modern system and applications. Yet, those are also (mostly) no longer self-funding products.

      so, an utility.

      create a (partially?) state-owned steward with a legislated mandate to develop the browser, self-funded via extra tax on digital goods and services.

      14 replies →

  • Any potential buyer will have to be looking to use Chrome to accomplish the same kinds of synergies that Google is using it for, to get ahead in some adjacent market. Depending on the buyer that could be good for competition, at least in the short term, but it's not clear that it will be better for us as users.

    • Hypothetically, why would some buyer using Chrome's monopoly to establish a market advantage in an adjacent domain be different than Google using Chrome's monopoly to establish a market advantage in an adjacent domain?

      1 reply →

  • How does splitting off Chrome as a separate company solve anything? They would still rely on Google for funding (like Mozilla) and being close friends they would do whatever Alphabet tells them to do.

    A better solution is to implement a bill like DMA in the EU to enforce competition among web browser vendors and fight monopolies.

  • > A very artificial arrangement where Google pay "arms length browserco" to maintain [a browser]?

    Sounds almost like Firefox.

  • The issue is who controls Chromium. I would create a non profit and staff it with a handful of maintainers. Their primary job would be to ensure safety and squash exploits. Their other job is to curate and approve pull requests from volunteers for enhancements. They should make it open source with the caveat if it is used for commercial purposes, there will be a licensing fee to pay for security enhancements, bug bounties, and the like.

    • Whoever contributes to it controls Chromium.

      If 90% of the contributors were non-Google, then it would effectively be controlled by non-Google, because they could fork it and their fork would still get 90% of development.

      See Terraform for a live example.

      The only reason Google "controls" Chromium is because Google funds almost, but not quite, 100% of its development.

      On a similar note, there's nothing stopping Microsoft from investing equal or greater amounts and forking Chromium (well, arguably they might already have with Edge). Except that they're benefiting from all of Google's investment, for free. Why turn down a massive developer investment from your competitor with no strings attached?

      1 reply →

    • Do you think Google will continue to invest money and resources into the development of Chromium if they were forced to sell Chrome? I don't. The first thing I would do is close source the Chromium project and work on a new closed source browser to compete with Chrome. I also don't see Chrome surviving when all of the Google/Chromium developers have left.

      1 reply →

  • They could sell a lot of the data that Google now gets for free and uses for its ranking algorithms, like Clickstream sells data to SEO tools like AHrefs and SemRush.

    • Google doesn't use Chrome data for Ads or Search. They're not allowed to based on the TOS, and also they have government regulators watching carefully to make sure they don't make a mistake like that.

      4 replies →

  • Sort of sounds like you are one step short of suggesting a browser is critical infrastructure.

  • I’m consistently fascinated to look at Chrome/Google and think of all the things we lost when we broke IE/Microsoft.

    To what extent and I holding a stupid belief, and why? I think I might like to be talked out of this, if reasonable. Want to try?

  • > browsers aren't really a self funding product

    They are, see how both Safari and Firefox, the 2nd and 3rd most popular browsers, have brought in tens of billions of revenue per year. Safari is immensely profitable, Firefox too would be if Mozilla wouldn't be run in an absurdly poor manner.

    > the user gains a huge amount of value by a browser being integrated into the OS, webviews in other applications, etc

    What is the huge value gain that e.g. Safari being integrated into MacOS is bringing me? Why couldn't webviews be backed by a browser of my choice?

    • Mozilla literally gets paid by Google, and not sure how you can quantify Safari as being profitable on its own when it's the default for all Apple products and realistically, the only browser on iOS.

      2 replies →

    • In what way is Safari profitable? How is that even measured? Has any consumer in the last 10 years ever specifically paid for safari? Or do you mean the payments by google to be the default search engine?

      6 replies →

  • > browsers aren't really a self funding product

    Yeah... Because massive companies use them anti-competitively as a moat against other companies, and as a loss leader to enable massive data collection and vendor lock in.

    "browsers aren't really a self funding product" is a symptom of dysfunction, not the inevitable conclusion of a fair market.

    • I'm not sure that software entirely obeys normal market theory. So much of it is zero-cost free.

      The synergistic effects are so strong that most users would prefer there to be The System, in which everything works together and there's no risk of incompatible choices. They don't necessarily care which system.

      The market in things like, say, file explorers is tiny. There's a few shell replacements (free), Midnight Commander and clones, and maybe over in the corner someone making a few thousand dollars a year from an Explorer replacement.

  • Browsers should be developed by an intercountry nonprofit. Funded by all the countries' governments.

  • >browsers aren't really a self funding product

    You can use Firefox 3, programs don't rot.

    • Not really true...

      In this specific case I'd be willing to bet that Firefox 3 probably doesn't handle current HTTPS/TLS standards and might not be able to browser the modern Internet at all (let alone display modern webpages, HTML5 video players, single-page web apps, WebRTC live calls, etc.)

      3 replies →

Google realized if they don't control the search distribution they gonna lose out sooner or later; which is kinda contradictory for them if they claim Google is the best search out there and that they are constantly improving it and that's why(they say), people choose it over other alternatives. But tbh distribution of your product/s is crucial.

Just look at Microsoft and their internet strategy, they chose the other route; push their internet browser(IE) down their massive distribution pipe called Windows and then introducing their search engine to this massive userbase. Fortunately this didn't work out for them but unfortunately that worked out for Google. And now Google essentially controls the Web in the more than half of the world.

  • > Fortunately this didn't work out for them but unfortunately that worked out for Google.

    No, Google was better, then they used Chrome as an extremely powerful moat to protect their situation. Google at first was like magic compared to the Altavista of the time.

    • It wasn't just the search that was amazing. You would go to google.com and be presented with only a Google logo, 1 line input box, and 2 buttons.

      No portal, no news, no header, no login, no advertise with us, no punching monkeys.

      It was a refreshingly different take on the web at the time.

    • That’s how I feel about Yandex lately. When it’s a struggle to find something on the common search engines, Yandex works like magic.

      It’s not that Yandex is particularly better in any way, it just chooses to not filter the content one is looking for.

      6 replies →

    • Yes but without Chrome, Google wouldn't have 90% search market share that they have today. They are completely dominating the WWW industry.

      7 replies →

  • TBH, when Google did that, Apple was already threatening making it impossible for IOS users to use Google's services.

I don't disagree, but I'm also not sure what the alternative is.

Who's going to buy Chrome that also doesn't suffer from the same anti-trust problems? Who would want to buy Chrome? Who would want to fund Chrome?

What browser would Android ship? In one view I kind of like the idea that Google would have to shop around and 'buy' a browser for its OS (competition good!), but also that seems ridiculous and easy to fall right back into the same trap.

  • > Who's going to buy Chrome that also doesn't suffer from the same anti-trust problems? Who would want to buy Chrome? Who would want to fund Chrome?

    Hmm. It's a good question, and I don't know the answer. I think there's a compelling argument that the problem is the scale of the harm. That is, even if the new owner has the same problems, the new owner won't also be the largest web company. So the problem still exists, yes, but becomes smaller. In particular having the #1 web browser strongly tied to the #1 web company has a lot of problematic dynamics that the #1 web browser being owned by the #25 web company doesn't. Maybe that company would be more open to forming beneficial relationships with the #2 and #3 web companies, for example.

    • Google already funds Firefox and makes Chromium[0] as well, which seems like quite a lot of effort to go to as a single company in funding/enabling competition. Microsoft had to do far less to resolve their EU dispute: just give users other options for browsers on install of their OS.

      [0] Unless if today you take Chromium and make your own browser, and it still has all the stuff in about logins and tracking.

      2 replies →

  • > Who would want to buy Chrome? Who would want to fund Chrome?

    This is interesting question especially when companies are usually just use Chromium instead of creating new browser (not even making hard fork of Chromium).

  • Most Android devices ship Samsung Internet.

    Chrome is only the default on Pixel devices...

    • >Chrome is only the default on Pixel devices...

      WRONG

      Before a couple of weeks ago where a Google Play Services update changed the first set-up process, almost all global (non-CN market) devices forced Chrome as default: Xiaomi, OnePlus, Realme, Motorola, Oukitel and whatever other weird brands there are left in the Android world. AKA anything other than Samsung

  • How about this: sell the browser to the entreprise [1] and use the profit to offer the browser to the public for free, which in turns helps you secure a user base.

    [1] https://www.island.io/

> If you ever doubt it, go check out Google Takeout. You'll be shocked at the amount of data you see there.

I sign in browser-wide and I do takeouts regularly. I don't see my browsing data.

> It doesn't ship with most essential apps, including a Phone app. In previous versions of Android, all of these were a part of AOSP.

And back when they were part of AOSP I never saw these example apps in the wild. Every vendor ships their own phone app. Every single one.

There's some "hey we compile a extremely old and vulnerable version of AOSP"-style Android distributions, mainly advertised for builtin su/Magisk or "degoogle", which did use these example apps, though.

I agree with other critics, they are toxic.

  • You've disabled the option to see it. Your data is still being collected.

    • That's unfalsifiable conjecture. I could just as easily assert that dang is building secret dossiers on all of us from our IP-request logs - we just can't see it

    • Yeah, I really wish there is a working option to make me see it, then.

      I always enable every data collection option in my privacy page, because even if I don't they get the data anyway, so why bother?

      I do see my entire Google Search history since 2006.

> - When you sign in to Google, you sign in browser-wide. Google now gets all of your browsing data, perfect for advertising. (If you ever doubt it, go check out Google Takeout. You'll be shocked at the amount of data you see there.)

I have yet to see evidence that Google uses browser sync data for advertising.

Go do something in chrome (look for cruises maybe), then delete the activity from myactivity.google.com, then wipe and reinstall chrome. You will see that you aren’t advertised based on that activity yet it’s still in your chrome history.

  • >I have yet to see evidence that Google uses browser sync data for advertising.

    Then you probably don't get out much:

    https://www.theverge.com/2024/5/31/24167119/google-search-al...

    Another major point highlighted by Fishkin and King relates to how Google may use Chrome data in its search rankings. Google Search representatives have said that they don’t use anything from Chrome for ranking, but the leaked documents suggest that may not be true. One section, for example, lists “chrome_trans_clicks” as informing which links from a domain appear below the main webpage in search results. Fishkin interprets it as meaning Google “uses the number of clicks on pages in Chrome browsers and uses that to determine the most popular/important URLs on a site, which go into the calculation of which to include in the sitelinks feature.”

    • It sounds like that might be talking about the hyperlink `ping` param: https://developer.mozilla.org/en-US/docs/Web/API/HTMLAnchorE...

      It could be marked as chrome because Firefox doesn’t support it (without a default-off config value) and Safari technically implemented it at a later date than Chrome, with non-chromium Edge supporting it years after chrome.

      This would indicate Google is using chrome to tell it when (via its support of the ping property) but wouldn’t indicate they are decrypting and analyzing chrome sync data.

I know this is anecdotal and that I should look further into this but I recently had to switch to chromium because google suite products were slightly unusable in firefox: youtube used way more CPU than it should’ve (maybe this is due to codecs but I was not able to solve it), google sheets crashed constantly, google meet slowed my laptop to a halt and I couldn’t even share my screen, and for some reason google calendar would suddenly start to hog CPU and RAM randomly. Since I switched to chromium everything is smoother and I just can’t believe chromium per se is just this better than Firefox.

  • I'll ask what extensions you use on your Firefox, because I regularly use Google Meet on an 2014 MacBook Pro with latest Firefox, and it doesn't even make the fans spin, plus all the goodies like environmental noise cancellation is there. Meet team also recently ported some of the Chrome only features back to Firefox, due to some fear, I guess...

    The YouTube videos in higher bitrates (like 4K) is generally due to Firefox's ability to hardware accelerate things, and there's a bit of difference there, yes. But on Linux and macOS (moreso in sequoia), I see no extreme CPU use. Just testing it on Firefox 132.02 on Debian Testing with Radeon 550 with open drivers, While I see a spike in CPU load, there's definitely some GPU load is also being produced, pointing to at least some GPU acceleration.

    On the other hand, Intel N100 with on board graphics can visibly struggle at 4K as far as I can tell. That one runs Firefox ESR though, I need to retest.

    I don't use WASM based Google Workspace tools (docs, sheets, etc.) heavily, but they don't crash when we use it on other pepople's documents that we collaborate on.

  • That's strange; I use all of those Google products (except Meet, I can't speak to that one) on a daily basis on Firefox (on Linux) and they all work just fine. YouTube in particular works very well, perhaps because I have the "Enhancer for YouTube" extension installed, plus uBlock Origin.

  • You can use chrome only for those google things where Firefox is crippled on purpose, and use Firefox for everything else.

  • Teams was acting up on Firefox the other day, too. It wouldn't have Q&A and whatever button is to the left of it enabled. "Not supported in your browser", or something.

  • I keep Chromium only for Google Meet, Datadog and Google Cloud Console. Opening these apps in Firefox makes fans on my laptop spin like crazy.

> - AOSP (the open source counterpart of Android) is now unusable. It doesn't ship with most essential apps, including a Phone app. In previous versions of Android, all of these were a part of AOSP.

This particular example is a bit misleading as those apps are still available; they're just unbundled from the system image: https://source.android.com/docs/automotive/unbundled_apps/re...

  • Your comment unfortunately implies that Google still maintains them, which is farthest from the truth (as this document shows, it is only maintained for automotive use - it is not usable on a regular phone).

    • Why is Google obligated to maintain them? It's open source so other people can and should do so.

> Google specifically serves a worse version of Search on Firefox for Mobile.

I don't think Google owning Chrome is really a factor here, but just a raw traffic question where FF Mobile has basically zero uptake. The experience they serve on FF Mobile is just the "we arent subscribed to validating that all of our shiniest JS works with this version of this browser".

The extension spoofs the user agent and arbitrary obscure features that only trigger on specific queries may be broken.

Google does do the effort of validating on other browsers where the traffic threshold is higher, including Firefox on Desktop. If they didn't own Chrome nor Firefox they still wouldn't really have incentive to spend more time supporting the tiny fraction of users.

A browser with a huge userbase would be extremely lucrative for an unscrupulous owner. A new owner could sell full and identifiable clickstream data for all browser activity. A new owner could siphon information from the non-public web for AI training, corporate espionage, or any other purpose.

> Google specifically serves a worse version of Search on Firefox for Mobile. You have to get an extension to get the full experience.

Huh, I wonder if this is why I have perceived a drop in quality from Google Search. What a stupid move from them -- not only have I stopped using Google Search and now pay Kagi (yes I know money still flows from Kagi to Google but even still) and have been evangelizing Kagi as well as taking every opportunity to shit on Google Search.

Great job G, you made the product worse and made me a customer of someone else

> Google specifically serves a worse version of Search on Firefox for Mobile.

Can't say I've noticed that, but YouTube definitively feels like it's getting an especially slow-loading version on Firefox.

Everything here you accuse Google of doing, Apple is running circles on. Ultimately, if this case goes through Google are right about one thing. The UX on Chrome is going to take a steep nose dive.

The APIs thing: it's not just Chrome, but Chromium too. I first noticed this when trying to replicate some of the screen sharing UI (buttons to share different things) from Google Meet, only to find that no non-Google domains have access to those APIs in the Chromium source code. Made a huge deal about it but nobody seemed to care.

>AOSP is now unusable. It doesn't ship with most essential apps, including a Phone app.

That's OK: these days, the phone (voice calling) app is only useful for receiving calls from scammers and telemarketers. I wouldn't miss it.

Yes, Google gains advantages through the Chrome browser so do we as consumers. The competition for Chrome should come from a better Browser not by being forced to sell it. DOJ is wasting time and taxes on this pointless verdict. Even more economic wastes are coming when the selling happens.

  • Thinking things just "should" happen is how we get to this issue in the first place.

To add: Google uses Chrome's established user base to bundle other products in the same way Microsoft uses Office subscriptions to push Teams.

The most recent example being Gemini now deeply integrated into Chrome. Had Gemini been a stand-alone product, it would have to fight for every user. Now billions of users have it at their fingertips.

I sincerely doubt that forcing Google to spin off Chrome is the only or best solution for this.

  • What other solution do you have in mind? Legislation about architecture decisions taken in software products seems preferable?

    In principle there is nothing wrong for example with a shared account for multiple products from the same company, many even prefer it. The problem only appears when this gets concentrated into too much power and can be leveraged in ways that distort the market and hurt consumers.

    • > Legislation about architecture decisions taken in software products seems preferable?

      To me this option seems more practical. And we already have some precedence for this kind of solution.

      For aviation we have entities like EASA issuing standards like ED-109 and for healthcare we have the HL7 organization issuing the HL7 standard. Another example in the healthcare industry is the DICOM standard created by the NEMA organization. This is not a new idea.

      I'm not arguing this approach is without problems. But we are already doing this for some pretty important topics, and I don't see why we couldn't use the same strategy for an "open web standard" that all browsers have to implement.

      5 replies →

I'm happy to give all this to a cohesive experience google provides. There are competitors in this market too, but google was the first and best not sure why they deserve this blatant overstep in abuse of power

> Google specifically serves a worse version of Search on Firefox for Mobile. You have to get an extension to get the full experience.

What's the difference, and where can I find this extension?

  • On Firefox, the design of the search results page looks about a decade older. Some people would argue that's a good thing :)

Uploads to google drive from Firefox are throttled to 10% of the bandwidth I get using Chrome

You missed one:

- They get to move forward with enabling and pushing features that allow for more hardware and software lockin to their platform: see unexportable passkeys