1. 83
  1. 30

    Chromes unfair advantage on Google pages: being recommended as the preferable browser in places so prominent on the website, places Google wouldn’t even sell to advertisers.

    1. 12

      does this feel exactly like the windows browser saga from about turn of the century to anyone else too ?

      1. 9

        Honestly? As someone who remembers the Halloween Documents clearly … not much.

        Microsoft’s power at the time is nothing compared to the power Google wields now.

        1. 18

          I believe the specific complaint @signal-11 is referring to is that IE 4 had a special mode that made connection establishment faster with IIS, but it only worked if the client is IE 4 and the server is IIS and this was seen as an example of using the browser to get an unfair advantage in the web server market.

          1. 3

            yes, that’s exactly what i was referring to.

      2. 9

        Google also has Firebase preferring code written in AOSP that I read when reversing some apps a while ago. A competitor would simply not be able to compete with certain Firebase features.

        1. 4

          I couldn’t parse that sentence.

          1. 6

            It’s missing a hyphen; “Firebase-preferring code”.

            Even then it’s awkward but it makes sense.

            1. 1

              Sorry, can’t edit anymore, but that is what I meant. Thanks for clarifying for me.

              1. 1


          2. 6

            What percentage of users use another default search engine? I would guess less than 0.1%. such things, while susceptible to valid criticism, barely matter at this point.

            The big advantage of Google, is being in a position of advantage and power and being able to essentially force website owners to follow their rules, under the threat of removing them from their index.

            Google started by crawling webpages. If someone did to Google what they did to the web. Not only would they be blocked immediately, they would potentially face legal consequences.

            In essence, building an index of the web was up for grabs and Google is allowed to legally prevent other player to ncrawl Google.

            1. 17

              It’s around 7%, but it’s also 100% of users of 100% of Google’s competitors.

              Google has a track record of upholding their dominance by “oops, but who cares” with their sites serving slower or inferior “fallback” code to other browser engines, buggy/discriminatory user-agent sniffing, browser “upgrade” prompts, AMP that happens to have better implementation for Google’s ads, etc.

              1. 1

                I wonder if this preferential preconnect behavior also benefits AMP. Don’t forget that the idea of “AMP’s performance comes from the restrictions it imposes” is mostly a lie: the real performance gain is from AMP pages being served from Google’s CDN and not the real origin.

            2. 5

              This definitely feels like a lose-lose situation for Google as already identified by the post. I’m not sure how the author would suggest Google experiment with this feature (it’s marked as an experimental feature) before going forward with the RFC or OSD option.

              1. 22

                I think the real question is whether Google would reject a patch from, say, DuckDuckGo that enabled the same feature. Remember that, although Chromium is open source, Google is the gatekeeper of everything that is accepted. If someone is looking at antitrust violations, I’d look at the fact that they’ve rejected patches to support *BSD, which significantly skews the market for desktop and mobile operating systems. They can’t really argue market share here, because they have accepted patches to support Fuchsia (which is a home-grown Google OS and had zero users when they started accepting patches into Chromium to support it).

                1. 10

                  You’re imposing unnecessary constraints, they could have and should have gone with the OSD extension option (or similar) from the start. It doesn’t need to be standardized, but it does need to be available to anyone.

                  1. 3

                    This is also how most good standards start off. Someone implements something, that implementation’s value becomes proven in production, and then it gets standardized. The other way around is a recipe for a “standard” that nobody uses. This is also why many (most?) new web technologies that end up in browsers are W3C Working Drafts, and why e.g. TypeScript or Babel often implement Stage 2 or Stage 3 TC39/ES6 proposals. It’s so people can experiment with the standards and see if they’re as useful or good as they were thought to be on paper.

                2. 7

                  The commit says it’s an experiment so I don’t see what the problem is. And I wouldn’t expect Google to create a new OSD spec and involve all search engine providers for something they’re only experimenting with.

                  1. 34

                    I don’t think “experiment” is a magic word that lets you blatantly create unfair advantages for your own services. Anti-trust law doesn’t (as far as I know) have an “experiment” exception.

                    1. 1

                      I think if Google was experimenting by suddenly getting all Android phones to automatically connect to search engines, they’d be blamed too.

                      1. 9

                        This could also have been an opt-in behavior on the search engines side, though. Perhaps a non-standard header or meta tag.

                        1. 2

                          but at least it wouldn’t be potentially illegal.

                          anyway they always have the option to not implement the optimization.

                          1. 1

                            I see your point but I honestly see that as an experiment by developers who want to see if their idea makes any sense. And the best way for them to test is with the Google search engine, since they can’t reasonably start spamming other search engines with pre-fetch requests.

                            1. 1

                              what’s the relevance of it being an experiment? does it exempt them from legal or social obligations?

                        2. 1

                          From another point of view, Google by creating a spec so that other search engines would benefit from the same optimization is actually sponsoring better usability for their competition. That gives the competition an unfair advantage, because Google has actually paid for developers time when developing this feature, and competitors didn’t.

                          That maybe sounds like a stretch, but I fail to see why it’s an “unfair” advantage, since Google is paying for the development of Chrome. It’s like saying that Apple plays unfair because it wants iPhone to be integrated with macOS, but leaves Android behind.

                          Since Google is the most popular search engine, and Chrome is the most popular browser, then optimizing their interaction seems like a sensible thing to do (everything is cheaper and faster). And if competition doesn’t want to stay behind, they should negotiate with Google to be included in the idea.

                          1. 10

                            That’s exactly not how it should be. First of all, Google paid for additional developer time to add to the general “connect to the default search engine” feature a “but only if it’s our search engine”. So the point is not valid here. Also Google and Chrome are not part of a closed eco system. So the Apple iOS example doesn’t work either. And the whole “they paid for it, so why would they allow competitors to benefit?” is a very short sighted idea. Imagine Google and big companies would generally be allowed to use their their power and money to make everything a closed system that only works with their own services. We would be back to the dark ages of the web, aka the nineties, where Microsofts IE was the driving force.

                            And in case you need another point of few: Google, as well as most if not all IT companies benefited a lot from Open Source software and contributions. The kind of software eco system Microsoft wanted to crush back in the days. Without those, Google, their competitors and most of todays IT companies would not be feasible. If course they don’t care. They take what they can take and make the best possible profit out of it. This is why we need laws that make competition fair.

                            1. 1

                              I don’t really like entering such response-to-an-argument debates, because they tend to never end, but here goes.

                              “but only if it’s our search engine”. So the point is not valid here.

                              It’s cheaper for them to only care about their search engine; they can actually measure how this feature affects their expectations and then pull out the functionality without anyone complaining about it. Also, if a bug pops out because this function doesn’t work with DuckDuckGo, but it still works with Google, then who should fix it? Should it support Yandex, Baidu and Ecosia? Should it support your own search engine? Should Google hire additional support people for this function alone? Who should create documentation for it? What QA team should deploy testing for it? Suddenly this function starts to be bigger (and more expensive) than initially thought.

                              Also Google and Chrome are not part of a closed eco system. So the Apple iOS example doesn’t work either.

                              Well, it’s really popular, and there’s no alternative with a similar quality for it, but I don’t think it makes it “public”. Can you extend google.com or use it somehow differently than it’s sold by Google? You can take Chrome and extend it, but that just means that it’s possible to implement this functionality on one’s own (in a fork).

                              Imagine Google and big companies would generally be allowed to use their their power and money to make everything a closed system that only works with their own services. […] This is why we need laws that make competition fair.

                              They do it all the time. If not in the products themselves, then with lobbying and with “buying laws” that make their life easier and pose problems to the rest. And then people can’t have nice things, because legally they’re not allowed to, only big corporations can.

                              And in case you need another point of few: Google, as well as most if not all IT companies benefited a lot from Open Source software and contributions.

                              I would argue that if an open source project declares its license as BSD, but has a different idea how it should be used, then I think it simply has applied the wrong license. You can’t say “you can do everything with the code”, but at the same time have some expectations that result in a situation that some people are using the code “in the wrong way”. State it in the license and the problem will go away.

                            2. 7

                              The key for me is that much like Internet Explorer before it, Chrome is essentially a pit into which Google throws huge amounts of engineering resources in order to ensure it controls the direction of the entire browser market. There’s little to argue that Chrome brings in any revenue at all for Google in of itself as a product - it exists to dictate the direction of web standards and to provide a vehicle which guarantees the continued dominance and tight integration of Google’s Search.

                              It really can’t be understated just how much engineering is continuously going into Chrome. Microsoft couldn’t keep up with Chrome’s release cycles and feature additions. Firefox is continuing to haemmorhage market share due to its increasing inability to keep up with Chrome. Safari is always behind in standards support compared to Chrome and that’s considering they often decline to support more complex features at all (though, like many others, I suspect they could if they weren’t more interested in preserving their App Store hegemony). Google won’t stop until it burns out or crushes what’s left of Firefox, and I imagine it will be deeply interested in lobbying efforts to “crack” iOS too.

                              1. 5

                                Unfairness is ill-defined, but here it seems to be a proxy for what should be forbidden by anti-trust law. Those laws are specifically designed to curtail actions that would be the “sensible thing to do” for a company with monopolistic powers. Do you reject anti-trust altogether?

                                1. 1

                                  The thing is that I trust governments even less than I trust corporations. And with corporations at least I know they are driven by profit, which makes them more predictable. With governments, the politics are just hidden deals between powerful people. So if a government wants to control the market, then the first question I think of is who will control the government.

                                  1. 3

                                    It probably would save time to come out and say that you don’t believe in anti-trust, so that people can recognize that there is a more fundamental disagreement rather than some other confusion.

                                    1. 2

                                      I have never been allowed to vote for the chairman of a corporation.