Google Antitrust Remedies: Promoting Competition Without Punishing Users
from the some-ideas-are-better-than-others dept
Considering how to increase competition in the search space without damaging end users is a trickier question than it seems at first. Many of the suggestions that people have tossed out have tended to focus on ideas that are purely punitive to Google, but which would also have negative impacts on users (and even some competitors). As we reach the stage of the antitrust battle where remedies are actually being considered, it’s crucial that we focus on solutions that will truly promote competition and benefit users, not just score political points against Google.
Earlier this year, I was left troubled by the end result of the ruling against Google in the first (of a few) antitrust cases against it. I think the (currently ongoing) case about the company’s practices regarding advertising is a lot stronger. The case that was ruled on this summer, though, was about Google’s massive payments to Apple and Mozilla to have those companies have Google search as the default on Apple devices/Safari and on Firefox.
At the time, we pointed out that it was difficult to think of any remedies that actually helped solve the situation. Both Apple and Mozilla more or less admitted during the trial that users effectively demanded Google search be the default, and any attempt to use other search engines resulted in angry users. If the court demanded Google stop paying the billions of dollars to Apple or the hundreds of millions of dollars to Mozilla, it wouldn’t hurt Google. Indeed, it would seem to help them.
Since both companies admitted that users were demanding Google as the default, little would change there other than Google getting to keep even more money. And Apple and (especially) Mozilla losing a ton of revenue. That didn’t seem very helpful at all.
In the intervening months, I’ve had a few conversations with folks about possible remedies that make sense. The most reasonable suggestion seemed to be DuckDuckGo’s main suggestion: allow other search engines to build off of Google’s search corpus by enabling API access under Fair, Reasonable and Non-Discriminatory (FRAND) grounds.
The best and fastest way to level this playing field is for Google to provide access to its search results via real-time APIs (Application Programming Interfaces) on fair, reasonable, and non-discriminatory (FRAND) terms. That means for any query that could go in a search engine, a competitor would have access to the same search results: everything that Google would serve on their own search results page in response to that query. If Google is forced to license its search results in this manner, this would allow existing search engines and potential market entrants to build on top of Google’s various modules and indexes and offer consumers more competitive and innovative alternatives.
Today, we believe that we already offer a compelling search alternative with more privacy and fewer ads, relative to Google. We’ve also been working for fifteen years to make our search results on par in terms of feature set and quality by combining our own search indexes with those of partners like Apple, Microsoft, TripAdvisor, Wikipedia, and Yelp. However, we know that many consumers still prefer Google’s results due to the benefits of scale discussed above, and this intervention would erase that advantage, instantly making us and others much more competitive.
This remedy would certainly allow for more competition to arise, which has proven difficult today. No one (not even Microsoft’s Bing) really has the reach and comprehensiveness of Google’s index. DuckDuckGo is mostly built on Bing (I know it insists it’s more than that, but in practice, it appears to be mostly Bing — as we discovered when Bing banned BestNetTech, and we also disappeared from DDG).
Every attempt to build competing search engines seems to run into the scale problem eventually without access to Google results. Even Kagi, which was briefly a darling among folks looking for a search alternative, apparently makes use of Google’s search tech on the backend. It seems like a pretty reasonable idea to make it so that others can license access to the API and build Google results into alternative search products, as this gets at the actual issues underlying this case.
A few weeks ago, the Justice Department filed its preliminary thoughts on remedies, and there are a wide mix of ideas in there, some crazier than others. A lot of the headlines that filing generated were around big “break up” ideas: spinning off Chrome or Android. These seem preposterous and unlikely. Under antitrust law while breakups (“structural remedies”) are certainly one tool in the toolbox, they are supposed to be related to the violation at hand.
Given that the antitrust problem in this case was about the search payments, and not anything specific to Chrome or Android, it’s difficult to see how such remedies would even be allowed under the law, let alone make sense. Indeed, without Chrome and Android being attached to Google, those products would likely suffer, as both are subsidized by Google, and that would do a lot to harm users. That doesn’t seem like a good result either.
So the proposals from the DOJ that match DDG’s suggestion of API access are much more interesting (and probably better) overall.
Plaintiffs are considering remedies that will offset this advantage and strengthen competition by requiring, among other things, Google to make available, in whole or through an API, (1) the indexes, data, feeds, and models used for Google search, including those used in AI-assisted search features, and (2) Google search results, features, and ads, including the underlying ranking signals, especially on mobile
Again, this seems to actually target the issue. It creates a scenario for increased competition without a corresponding harm to users or to other competitors. Many of the other sections do not.
Also, arguably, the DOJ could have gone even further, conveying on users more ability to designate access to information and data as a way to escape the silo of Google. This is a bigger issue and one that doesn’t get as much attention, but the ability of large companies to lock in users has diminished the ability of competitors to grow and challenge the network effects of existing businesses.
For some users of Google, the fact that it tracks your history is not seen as creepy or privacy invading, but rather a benefit for that user (and yes, this is not true for everyone!). But if the user could retain control over their own search histories and preferences, and allow third party search engines to access it with the user’s permission it would also help users get out of an existing silo.
Just as one example, Google knows a fair bit about what I normally search on and click on. But if I could make use of that history and give DuckDuckGo or Kagi or someone else access to it for the sake of improving their own search results to my queries, that would be potentially useful for competition. And all it’s really doing is saying that the user who generated that history and metadata should have some control over it as well, including separating it from the underlying Google product.
Yes, this would have to be done carefully, to avoid (say) exposing more sensitive data regarding searches to these other companies, but if it was done in a way that was transparent, and which the end user had control over, it could be really valuable.
Not surprisingly, Google is very, very upset about all these potential remedies. It suggests that if they were forced to share such things with others, it would lead to privacy and security risks:
Forcing Google to share your search queries, clicks, and results with competitors risks your privacy and security. It’s widely recognized, including explicitly by the DOJ in its outline, that forcing the sharing of your searches with other companies could create major privacy and security risks. The search queries you share with Google are often sensitive and personal and are protected by Google’s strict security standards; in the hands of a different company without strong security practices, bad actors could access them to identify you and your search history — as we’ve seen before. Additionally, while sharing Google’s search results with others might create a few copycats, it could also decrease incentives for other companies to actually innovate in search.
This very much depends on what information is shared, with whom, and how. I still think that simply giving the user more control over it, rather than just letting companies fight over access, solves some of Google’s stated concerns.
On the whole, the larger structural remedies (spinning off lines of business) don’t seem to target the underlying issue, seem mainly punitive, and won’t do much to help competition or users. But the idea of opening up access to search systems and data, especially if it gives more control to the end user actually seems like a really good way of increasing competition and improving the situations for users.
Google’s statements about security and privacy are still ones worth considering, but there are ways to deal with those issues, mainly by providing more power to the end user, rather than just opening up that info directly to other search engines.
Filed Under: android, antitrust, api, breakups, chrome, competition, data, doj, frand, remedies, search, search history, silos, structural remedies
Companies: duckduckgo, google
Comments on “Google Antitrust Remedies: Promoting Competition Without Punishing Users”
Would this actually create competition that people would use, though? It just sounds like Apple keeps getting its billions, Mozilla keeps getting its millions (which seem to not go toward anything), and we get search engines that receive a handful of improvements to their usability, but which people don’t use because they’re not Google.
I’ve seen it many a time before. People say that they love competition, but never use the competing services any any meaningful way even when it’s just as good, or even better in some aspects. The search results on Google feel like they’re going to keep getting worse from the bad state they’re already in, but people will refuse to use anything other than Google and they pay Apple and Mozilla to stay the default. Sometimes angry tech-illiterate users don’t actually know what they want or what needs to be done.
Did any of the remedies talk about how Google cut a deal with Reddit so that their search engine is the only one that can surface results that occur after a certain date? I feel like that deal needs to be scrubbed for the sake of competition. And yeah, I’m gonna be honest, I would love to see structural remedies with Google broken up. None of these large tech companies would have been allowed to get this big if we had real antitrust with teeth over the last 30-40 years, and we’d be better off for it, I reckon. What Lina Khan at the FTC is doing and what the DOJ is doing as well are moves to steer us back toward actual antitrust rather than “consumer harm” garbage.
I think one big element you’re underestimating is how something might ‘damage’ end users a little bit, but be beneficial in the long run. If a remedy makes search a bit worse, but enables competition in the long run preventing Google from enshittifying, that’s a net win for consumers. Of course, if you can avoid it entirely, that’s better. But if not, it can still be well worth it.
There’s some truth to this, but if it were that black and white, Google would just refuse to pay in the first place. This theory requires thinking that Google wants to just burn money for no reason. If anything, Google should be happy to stop paying, and endorsing this outcome. It’s free money (and with respect to Apple, a not insignificant amount of it).
Eh, kind of depends on why you see the underlying issue as. To the extent that more and better data makes Google’s search better (and/or more easily monetizable), those other lines indirectly contribute to the lack of competition. The reason Google subsidizes things like Chrome/Android is that it’s a massive firehose of data that gets monetized elsewhere- including search/ads. There’s an interconnection between data, search, and ads.
That said, search is one of those things where it’s a monopoly not just because it’s vertically integrated, so it’s harder to solve. Google’s search is legitimately better, and spinning off Chrome etc probably only really touches on the edges of that. I like the idea of licensing their database.
I think network effects get a good amount of attention, but they’re way harder to beat. Especially without being coercive, and especially under existing antitrust law (which doesn’t actually ban monopolies, just unfairly leveraging it). Case in point, forcing Google to license to their search is pretty coercive. And there’s a good chance this lawsuit never happens if they didn’t pay Apple (and to a lesser extent Mozilla). The benefit to scale by itself probably doesn’t violate existing antitrust law.
Well, try searching via Tor sometimes. Your search queries will mostly go into a black hole, because Google refuses to serve users whom they can’t identify. If that’s Google’s idea of “protection”, I don’t want it. Pretty much no other search engine is so adamant about blocking anonymous use.
By contrast, DuckDuckGo explicitly supports Tor (give them the search query “duckduckgo onion” to see an alternate URL), ensuring that search queries can’t be linked back to users—unless they’re searching their own names or something like that.
Re:
Brave Search is pretty good as an alternative, and I recently discovered Startpage, which seems to have been offering the same service for at least the past 12 years. I used to recommend OneSearch, but stopped when it first announced its impending closure (two days to go).
Re: Re:
Brave is even more shady than Google.
Re: Re: Re:
On what basis? I do find them to block Tor connections occasionally, but nowhere near as thoroughly as Google. Results seem decent.
Re: Re: Re:
{{Citation needed}}
Uh, well if i decide to share my data with another party the way i share it with you, you shouldn’t have anything to say about it.
Privacy risks and open data
It’s true that releasing detailed click data (or even mere URL statistics) presents significant privacy risks. Google is well positioned to do it with proper guarantees of differential privacy, just like the Wikimedia Foundation did with far more limited resources (https://meta.wikimedia.org/wiki/Research:Wikipedia_clickstream ).
The problem seems to be that Google’s search engine has reached the scale where nobody else can afford to build out a new search engine to the point it can compete, while simultaneously the flow of advertising revenue from the use of Google’s search engine results is so great nobody can afford to out-bid Google for default placement in a browser (plus the fact that for the dominant browser, Chrome, no other search engine would even be permitted to bid for it). So perhaps the solution is to sever the search engine from the advertising business?
And is now copyright infringement, thanks to… Google. Up until about twelve years ago, there was a website called Scroogle that scraped Google’s results so it could serve them without all the bloat and tracking cookies, so Google sued it for copyright infringement and won despite the fact that if there is any actual copyright in mere search results, then it would surely belong to the person using the search engine, who is thus authorizing the copying of their IP, therefore no infringement.
Re:
There’s more than one way in which they’re promoting the idea of searches as copyright infringment. They also eventually gave in to the people sending DMCA notices telling Google to stop linking to certain sites, because those sites (and not anything Google did) allegedly had infringing data.
The DMCA, as bad as it is, doesn’t have any such requirement I’m aware of. Phone companies don’t have to remove the numbers of businesses accused of copyright infringment, for example. But we seem kind of stuck with this idea for search, and as Mike says, sometimes a site like BestNetTech.com just disappears from some engines because of some bullshit like that.
This isn't a solution
Every search company has mimicked all of the terrible, useless Google “features” in a race to the bottom. Making them all even more google-like isn’t a solution.
There was a time when both Bing and DuckDuckGo were superior to Google in functionality (ex: not changing my spelling, and respecting booleans). Then everybody decided to be just like google and treat all search terms as suggestions, so all started sucking equally.
This “solution” would just create more sameness, without offering users literally anything, and without punishing google at all. Demonstrably, people don’t care whether search results or functionality are good; they’ve been trained to use google, so they use google.
FRAND = fair, reasonable, and non-discriminatory
Assuming you arent recommending Google just give away for free the use of their API to competitors, how could a “fair and reasonable” price for access to their API be determined?
Re:
Fairness is a whole can of worms I’m not getting into because it is about as bad as trying to get a whole country to agree upon pizza toppings.
But, there is the natural monopoly precedent in place for natural monopolies like power plants of cost-plus models. That is charging of the expenses plus a fixed margin of profit. For power plants that margin is about 10%. Not saying that it would necessarily be without flaws or even a good solution. The perils of cost-plus models are infamous.
Making it workable would get slightly weird in Google’s case because of prioritization and their ad auctions. I suppose making the remainder of the over allowed profit margin revenue could go to the ad-host would make it workable. For context, the existing percentage is already 80% revenue to the ad-host.
Re:
You can do some back of the napkin math based on what came out in the trial about what the actual cost to Google of maintaining search is each year, divide that by the total number of searches currently being done on Google, and you come up with something that’s pennies on the dollar.