Google Antitrust Notes
tbray.org
tbray.org
Also not a lawyer, but I listen to enough law podcasts to make a wild guess: if the lawyers ask for specific structural relief, they constrain what the judge is allowed to do. For example, let's assume they ask for Google to be broken up... If they can't build a case strong enough for that to happen, they get no relief and they wasted the case. Or if they don't ask for Google to be broken up, but they make a strong enough case that a judge could decide that's appropriate, it doesn't matter; plaintiff asked for something else, therefore plaintiff believes that something else makes them whole, therefore that's all the judge can and should do to resolve the case. Remember, the whole point of courts is to keep society humming along by exploring how people have been wronged and righting that wrong, and the plaintiff telling you something rights their wrong is a good primary source.
Asking for "Structural relief as needed to cure any anticompetitive harm" leaves it in the judge's court and grants the saddle-point of maximum potential for disruption to Google and maximum potential for some positive outcome.
The most educational aspect has been their dives into the point of the law itself, i.e. the underlying theory of why we approach guilt, innocence, findings of fact, etc. the way we do.
What happens when the plaintiffs have 5 secret ideas in mind and the judge comes up with the worst one in his independent process?
How often has a judge given a longer sentence or larger layout than a plaintiff asked for?
Maybe so, but judging by my search results most of the fruit is still on the tree. Natural language understanding is an active field of research.
Turning Google into a search utility is a bad idea. Splitting off the crawling part and turning that into a utility would be a lot more interesting. It would allow search engines to compete on search quality while sharing the capex required for crawling.
Apparently there is some effort in Distributed SE coordinated by blockhains. I think this comes close to the idea of "Google as an utility".
https://hackernoon.com/is-the-concept-of-a-distributed-searc...
"...does away with every form of a single point control in all of the tasks relating to indexing, querying, data mining, and crawling."
Yeah mostly that's distribution. How, though, will something that does indexing in 10 different locations ensure that I get results including something that was indexed in Japan 5 minutes ago?
And Blockchain. I get that one might like to mitigate fake search entries (as that post notes). How though, using a blockchain, does one manage someone's right to be forgotten, when the blockchain's sole raison d'etre is to make all the things indelible?
Such a blockchain would not be good for current news, obviously. But even google takes days to update the average blog.
Right to be forgotten requires the ability to edit a blockchain, which is contrary to the entire point. A judge is unlikely to say you’re compliant with a voluntary flag on a distributed network so you’re either editing history or restricting it to trusted parties with legal obligations, at which point you’ve just invented a centralized system with more overhead.
Similarly, that hypothetical team of genius developers working to solve all of the hard problems hasn’t shown up on cryptocurrencies in the last decade despite enormous amounts of money being plowed into the field, and this would have orders of magnitude less funding. Basic economics suggests that you’re far more likely to see a few ideologically committed people drowned out by the SEO types (who have far more money) trying to manipulate things to their advantage.
The underlying problem is that almost nobody thinks they should pay for search and the minimum quality bar is quite high if you want people to switch from Google.
Case in point: image search for "shirt without stripes":
https://github.com/elsamuko/Shirt-without-Stripes
HN comments: https://news.ycombinator.com/item?id=22925087
Edit: I'll also point out that it may not even be a modern PageRank fault but a consequence of learning to rank or some other major signal but it's in line with what I would suppose to be the culprit.
The lawyers were right not to talk about remedies. Really, just getting the court to agree that google is abusing it's power is enough to stop them. Fool me once, as they say.
That said, something has to be done about pushing more advertising rev and therefore power/influence back to alternative publishers (wpost/wsj/nyt are alt publishers now). This monopoly/duopoly on 'thought' is insane and could very well be driving a great deal of ignorance / conspiratorial thought / etc in society. Trump loves twitter!
Trump is crazy, but crazy is what you need to get your message out these days. Just look at Obama and his tirade yesterday. I think it was ill-advised, the value prop of biden is dialing down the crazy, but that could just be the world we have now without more long form / substantive dialogue.
They're not crawling the web frequently enough. They don't execute JavaScript (as far as I know) and I'm not sure they have access to everything that Google has access to (such as paywalled sites).
I could be wrong though as I'm not following the project very closely.
Which, by the way, is another hindrance for potential competitors.
So is the Stratechery guy right? Is dis-aggregation the only solution?
Most top results for my searches are quite literally gibberish.
On a given Google search, you have one to four ads, maybe an AMP carousel, and a knowledge graph box. Somewhere is the end of what fits on your device screen. Past that is the first real search result.
Google is making money because everyone thinks the first ad is the first search result (They're nearly indistinguishable, visually, and studies have demonstrated non-technical users can't tell them apart) and then everything below the ads that fits on screen is packed with links to other Google properties.
We are at the point where organic search results are nearly the "page three" of Google Search: Somewhere nobody goes anymore.
Meanwhile, Google search traps me in a loop of the top ~5 ad spenders in a given industry VERY frequently.
Used the phrase "public records" anywhere in your search query? Now the first page is all click funnel blog posts by the same 5 background check companies and ancestry.com.
Mentioned the term "VPN" somewhere in your 10 word query? Clearly you wanted a page full of sales funnel blog posts by the same 5 consumer VPN providers with big marketing budgets.
Want to disable Google Search ads? You can't. You can only make them less targeted, until the ToS changes in a month, and only if you log into a phone verified Google Account.
Want to use an ad blocker on Google Search? Too bad. Google's native ads continuously define the state of the art in evasion of end-user controls through endless evolution and complex adversarial misdirection even within the same browsing session.
Just use DuckDuckGo, and disable ads, and donate. It's a much better user experience (even with ads).
A significant part of Mozilla's revenue comes from their deal with Google. I don't think a project as big as a browser could survive without that revenue.
If Google wins, Google has a search monopoly. If Google loses, they will have a slightly smaller search monopoly and a browser monopoly.
While you could be cynical, I think people using Wikipedia are way more numerous than people using Firefox.
I think it's inevitable that the future will be dominated by a Blink based browser. It's omnipresent on mobile and the desktop will probably just be Edge (a Blink derivative).
> inevitable
Well now that that's settled...
It's my understanding that this case will continue to evolve as it goes forward and so will the possible remedies. I think you are assuming the best case scenario in a loss for Google.
Mozilla doesn't factor into this for me. They have been making missteps for years. The fact that they have become totally dependent on Google rather than finding other partners and business models is on them. Live by Google, die by Google.
While Mozilla has made some progress on the browser quality, the reason they are loosing the browser market is 1) the browser just isn't as good as Chrome (I've tried to love Firefox) and 2) Google is spamming Chrome on all its platforms. This lawsuit may resolve #2, but only Mozilla can resolve #1.
How do you figure? Firefox is my browser of choice on both desktop and mobile.
Outside the work context, leisure time is regularly frustrated by FireFox struggling to render pages or load videos on Google owned YouTube.com as well as they do in Chrome. This seems to happen a few days each month for the past 8 years or so.
Therefore Chrome is clearly the better browser. /s
However, I don't think many would care if the world ends up on a single common browser as long as it doesn't consolidate power to a single entity. Microsoft has their own fork of Chromium, and if the possibility of building a browser business makes a comeback (thanks to any changes in Google), I'm sure many other companies will follow suit.
Of course, this all depends on what actually happens to Google as a result of this lawsuit. So we'll have to wait and see.
I'm not so sure weaponizing privacy against Google by breaking a good amount of websites embedding Youtube videos is really the best way to get market share though, so by that metric, Google losing would hopefully mean Mozilla could work just as well as other browsers by default and actually gain market share.
I think people forget how terrible browser compatability used to be. You had to use jQuery and worry about what you could and couldn't use. Pax Chromana in comparison is a dream.
It's possible that chrome stagnates like IE did. But I don't think so. Google is primarily a web company and so it's in their interest to keep innovating on the browser to make their websites better.
https://blog.samuelmaddock.com/posts/google-widevine-blocked...
And this bullshit even before "Pax Chromana". Imagine the things to come. It's like we've learned nothing.
Besides, "freedom activists" don't need to watch news clips or youtube?
Or are they so short-sighted they think DRM will remain limited to entertainment services like Netflix and Hulu?
Do "freedom activists" want to live in a world where you need Google's permission to compete with their browser? Is that the kind of freedom they're activisming about? Freedom from the tyranny of non-Google choices?
Sure, but that's like being one of the tribes of Europe before the Romans. When the tribes got crushed and everyone became Roman peace and prosperity (mostly) reigned. Having to use chrome if you don't want to sucks, but it makes things way easier for developers to have one browser to target.
Well, it didn't really go well for the tribes. About half of the gallic population was killed or enslaved. As for the romans, that war was started to reinforce the personal power of Caesar, and the power imbalance it generated led to the fall of the republic. I'm not sure your historic analogy makes a great case for your point.
> it makes things way easier for developers to have one browser to target.
What makes things easier for developers is standards. As for targetting one product present in a walled garden, it may be pleasant in the short term, but we have many cautionary tales about it, and Google is responsible for more than their fair share of them.
on mobile the complaint says:
"The six “core” apps are Google Play, Chrome, Google’s search app, Gmail, Maps, and YouTube. Manufacturers must preinstall the core apps in a manner that prevents the consumer from deleting them, regardless of whether the consumer wants them. These preinstallation agreements cover almost all Android devices sold in the United States."
So chrome must be installed, prominently, and is not able to be deleted by the device manufacturers.
In absence of this, would device manufacturers install firefox and send them some support?
- It isn't actually that browser rendering anything, it's Safari under the covers (though Chrome does replace the networking stack with its own) - Every attempt to launch a browser other than Safari requires an extra selection on the user's part to approve it. Every time.
Don’t you find it a little strange that pro-business/anti-regulation conservatives are going after “big business”?
For example, apple has built deep linking into their browsing pipeline (allowing phone applications to snoop and intercept registered URLs). For example, if the amazon app is loaded, any web/email/imessage activity to an amazon url gets sent to the app.
Reread the relevant section: Chrome cannot be deleted by the device's user either.
Allowing users to delete "core apps" (and ensuring that doing so doesn't brick the phone) would go some way to leveling the playing field even if device makers were still required to preinstall them. It would either force many of the capabilities that have been migrated out of Android back into it, or would force releasing Play Store as open source to provide alternative implementations of those APIs. Maybe a bit of both.
This would undoubtedly complicate security and permissions, but it would be worth it.
There ought to be an exception, BTW: locking things down to prevent uninstallation should be acceptable in some contexts in return for restricting advertising and targeting, for example, on kids' phones.
I'm a fairly savvy user, and I am uncomfortable disabling anything that isn't very obviously carrier-added adware/bloatware.
Dark pattern to scare users away from switching browsers.
Mozilla has already written a blog post asking the court not to go after Google on this issue because it might hurt them. I am sure Google is all too happy to promote this angle, suggesting we should protect Chrome out of some misguided attempt to protect Firefox.
If Google is stripped of Chrome or prohibited from using various tactics to promote Chrome over alternatives, Firefox will benefit, massively. Revenue may drop in the short term, so it may not look great for Mozilla-the-company, but Firefox usage will soar, and revenue-generating strategies will follow thereafter.
It's sad Mozilla is hurting themselves here by coming to Google's defense, but that's how whipped Google has them right now.
Google save $10B per year paying to Apple for the default search engine spot. User continue to use Google anyway as 98% of user simply dont go anywhere else. Their Chrome remain dominant and Firefox is dead.
The only possibility this is actually worst would be the incentive of Apple creating its own Search Engine. Likely without porn or whatever content they deem unfit and enter Web Curation. Of course they would be highly unlikely given the current Anti-Trust issues.
Personally I think there's plenty Google is doing that's wrong † but I'm just disappointed the antitrust complaint seems to be all about the advertising industry.
As far as I'm concerned, the online advertising industry is already doomed. The first time there was a drive-by malware installation from an advert on a reputable website, the industry's fate was sealed.
When I heard there would be antitrust action, I was hoping there would be more attention to all the other problems, so hearing it's all about competition in a dead industry is a disappointment.
† Nonexistent support, low-privacy defaults in Chrome, perpetual "oopsies" breaking firefox support, over-enthusiastic anti-spam features forcing e-mail centralisation, iOS gmail constantly "forgetting" that I didn't want to install Chrome, reCapcha and DoH and AMP and GA all having the potential to act as centralised tracking, reCapcha being deliberately slow and difficult for people who disable browser fingerprinting, opaque and capricious treatment of youtube creators and android app developers, pushing sketchy browser "features" like "signed exchange" that badly solve problems of Google's own creation...
The DoJ focussed on search. Specifically, Google paying for the default slot.
The DoubleClick/ad business is not directly addressed. The harms in the complaint are more widely discussed than in this article.
That “dead industry” is worth many $100s of Billions. What are you on about?
You may feel like ads are a dead industry, but remember that via Google's Chrome monopoly, Google can harm ad blockers (Manifest V3) and protect their ability to track and monetize users ("privacy budget"), and that their decisions as such affect the vast majority of users on the planet. Everyone else might be stripping out advertising and tracking, but as long as Google's browsers are protecting it, it'll do just fine.
Breaking the back of Google's monopoly is going to take going after them in many angles, and their browser monopoly and search monopoly are very important pillars of that. But all those pillars are upholding one thing: Ads.
Do they now? You seem to know a lot about them.
YouTube Premium, Docs, entire GCP suite etc kind of hint that they don't have the same vision about their company as you seem to have about them.
> Search and Chrome and Play Services and everything are about protecting their ad revenue.
I might be mistaken but reading comments on HN from googlers, I remember Chrome being a docs outcome as a better browser can support web productivity software well.
A narrative Googlers might tell you, sure. But that has no historical reality. Chrome, is, of course, the brainchild of Sundar Pichai, prior to which was in charge of the Google Toolbar, a pack-in that Google paid Adobe and other companies to inject into their software installers to change the search provider on IE to Google.
Chrome was developed out of a (not unfounded) fear that Microsoft would intervene and block the Google Toolbar from IE. (Microsoft hesitated solely because they themselves were pretty scared of getting hit with antitrust violations again.)
The Google Toolbar, and it's direct successor, Chrome, came explicitly and exclusively out of a desire to set people's search provider to Google and to enable collecting user browsing data, which of course, leads to ads. Glowing biographies of Sundar Pichai that came out around when he became CEO all testify to this.
An example is https://marketrealist.com/2019/12/a-history-of-googles-visio...
There is a historical account of Google devs submitting improvements to Firefox which were delayed or not merged and that led to the desire to just create a new browser.
I will need to search for the Firefox PRs if they exist to verify that story.
Chrome really set the benchmark for browser performance and reliability when it came out.
I'm not sure I buy that line of argument myself, but I did expect it to come up more prominently.
During that 2005-2012 timeframe you had a US search engine becoming really global, a US online retailer almost single-handedly inventing a new industry (cloud computing) and a US social-web company crushing almost all foreign competition on its own home-turf (and which could only be stopped by explicit Government intervention, as was the case with China explicitly blocking FB and with Russia directly taking control of VKontakte as a better way to counter FB).
Amazon applied open source software to the hosting and specifically the network stack and security stack. The model was already there.
Should you not prosecute murderers because birth rates are down?
Also consider that mainstream androids are significantly cheaper.
I wont go into further comparisons, but just check out the p30 camera
“Winning” against China isn’t just some notional idea. Like it or not, being globally competitive matters for the overall health of the US economy.
In the end, if any of these companies are forced to change some policies or split from an acquisition or two, they'll still be massive institutions.
When I switched, about 2 years ago, DuckDuckGo was clearly worse. Now I feel like DuckDuckGo has gotten a little bit better but Google Search has definitely gotten worse.
I'm now using Pocket and Pinboard a lot more because I can't count on search being able to bring up something I saw a couple of months ago.
What does selling ads look like if search is not a monopoly? As an advertiser in the new world, I'd have to go around and buy ads on all the new search front-ends. I probably want to pay them proportionally to how much traffic they get (well, conversions, but let's say traffic). If there's a small number of large front-ends then maybe I go to each of them directly, but if there's a long tail of front-ends then probably there'd be a search middleman who the various front-ends contract with... I guess there's still Google Search, and Google AdSense, but now AdSense also sells ads on other front-ends, and Google Search works with other brokers besides AdSense?
And site operators presumably have to deal with many more ad platforms, which is likely to be a complication for them. Theres a certain logic to the way it works now, even if it is (arguably) anti-competitive and undesirable.
Today other adnetworks mostly serve sites that are shady, controversial or too small for Adsense since they can't compete.
It's basically the case of skyscanner, kayak, etc. and the air travel companies.
Click fraud is pretty real and I’ve had contacts say it’s 30-50% of actual ad revenue. Currently, this is just “factored in” but how neat would it be if this could be prevented. Google has no incentive to fix this as click fraud is profitable for them.
For ad auditing, Google frequently shows ads to people outside my geolocation and doesn’t audit how they do this. Would be nice to fix this as it’s a decent number of botched customer interactions.
Finally, pricing could be so much lower. I’ve advertised for low frequency terms where the minimum bid is $1 even though there’s only 1 or zero current advertisers. So the “auction” is against a floor set by google, not the market.
If no one else is buying a keyword, the minimum should be very low and climb as multiple ad buyers drive the price up. This is probably the most annoying as it would save businesses so much money.
Things have changed. Google repeatedly has modified the delivery settings of their ads for active campaigns, just as you complain about the geolocation issue. My guess is that if someone did a proper investigation Google would have to back pay advertisers many billions of dollars. This is fraud anyway you cut it. They have done the same thing with mobile ads. You disable the mobile app inventory and they modify the UI and re-enable it.
Showing excessive ads wastes users' attention and teaches them not to pay attention to other higher value ads. A price floor seems like a very reasonable way to handle this?
(Disclosure: I work for Google, speaking only for myself)
Second, when there’s a single ad shown and the floor is $1 or $2 or $3 even that makes no sense. Having no bidders with a floor and a single ad vs 2 or 3 is done for a single reason- revenue. Any customer benefit from missed ads is wiped out by wasted costs. It’s rent seeking because Google has a search monopoly.
I'm not sure I understand what your objection is. It sounds like you're saying that you'd like Google to remove price floors and show many more ads? Is that right?
What do you mean by "wasted costs"?
But what I’m saying is that if there’s only a single ad and they are paying $3 because google set that limit, they could lower that to what the market will bear and start at .05 or something.
If they are worried about ad space they can limit to 1-2 ads and let people bid on them. They already do this as there’s not an infinite scroll of ads, they only show 1-3 and then more on the next page. Setting floors on ads doesn’t reduce the number of ads I see, it just makes more money for Google.
This is only possible as they are a monopoly.
Really? If you go to https://myactivity.google.com/item?product=19 you should be able to see all your recent google searches. Here are my queries today:
* myactivity
* somerville trick or treating 2020
* Lilywise dinner
* jefftk mask usage
* How far around the world is it at the longitude of NYC
* wbz coverage map
* wbur coverage map
* wbz frequency
* radio station coverage area
* us radio station map
* time zones
* telegraph
* atlas of the historical geography of the united states
* isochrone
* travel time from dc to nyc historically
* nyc longitude and latitude
* philly longitude and latitude
* dc longitude and latitude
* wikipedia time zones
* wikipedia solar time
* fivethirtyeight
* who won the debate
Going back and searching these again, only "atlas of the historical geography of the united states" showed ads.
> I’m saying is that if there’s only a single ad and they are paying $3 because google set that limit, they could lower that to what the market will bear and start at .05 or something.
If they did that, then every query would have an ad at the top, generally a pretty irrelevant one. People would learn to ignore the ads. This would then make ads on searches like "buy insurance" much less valuable.
> Setting floors on ads doesn’t reduce the number of ads I see
Huh? Of course it does. If you set a floor, some bids are going to be below the floor. For some queries, in my experience a large majority, all bids are going to be below the floor and you see no ads.
> This is only possible as they are a monopoly.
This has nothing to do with whether Google is or isn't a monopoly. Even a very small search engine would make this kind of tradeoff, choosing not to annoy users with low value ads.
Similarly, if you try these searches on Bing or DDG (which both show ads from Microsoft Ads) you will see they are using the same approach, with many ads on highly commercial queries and no ads on most queries. They even published a paper about it: https://www.microsoft.com/en-us/research/wp-content/uploads/...
It's not clear how this is better. The gas company still exists and takes the bulk of my payment. The 3rd parties tack on their service fee. And they basically all run a bait and switch where you sign a contract for a fix term and then the price gets jacked up if you forget to sign a new one.
It's the worst of both worlds. The gas company still has a delivery monopoly. And now we've got a bunch of scummy 3rd parties running what are basically scams.
- Crawling and document parsing. Imagine something like archive.org where all search engines would get the content from. This could also be used to enforce copyright legislation (e.g. deletions of documents, etc...). Today, it's nearly impossible to collect the data as google does, because most website owners won't even allow that you crawl them (e.g. distill networks anti bot protection, etc...)
- Metrics (e.g. data from google analytics, dns resolvers, browser toolbards, browser statistics, etc...). Again this data should be available to everyone from a central authority.
Then there could be more independent search engines building on top of that data. If that doesn't happen, then there will never be any competition. Todays copyright and data privacy laws even further strengthen a monopoly position, as smaller players will not even get access to some data (or at different pricing then google does)
Regarding those other data sources, the more privacy conscious engines would perhaps object to particular kinds of data but agreed that the kind of scale and various sources that Google has at its disposal, does give it an advantage when it comes to search.
There are alternatives beyond Google and Bing who do index the web albeit at the moment are smaller in scale, and likely outnumbered by all the Bing clones. There is a high barrier to entry wrt cost but the technical/capital hurdles of crawling are not so high as compared to ranking and indexing.
I wonder how true this is these days?
I understand that the "long tail" of web queries is very important, but if you were to skip that for version 1, and concentrate on crawling the top 1M English language sites how much would it really cost?
Would there be value in a search engines for specific topics? For example, movie reviews and blogs. Assuming you had a way to determine if a page is a review or blog, would it make indexing easier to ignore the rest? Could a tailored search engine be tweaked to give better results?
I kind of think it would be useful, because some websites I use have really great search tools (e.g. ones with tags). But those are only for the website itself, so there's no crawling.
Actually it doesn't.
That's why Apple has been building out their search engine for the last few years.
Just like with Maps at some point it will be good enough and they will push an update to Safari to use it as the default.
"I recommend reading it; It’s not that long and I certainly learned a few things about the shape of the search and advertising business, and you probably would too."
Maybe it is too early in the procedure to ask for specific measures? I know that in criminal law, the state doesn't suggest specific punishment until after the verdict; and in civil law in many countries, you may ask for some specific relief to begin with, but it is very customary for the scope, nature and extent of that relief to change and be fleshed out as the procedure approaches its conclusion.
This approach would have different consequences to the DOJ's approach. It would leave intact google's ability to provide useful services that Tim Bray and others like, whilst redressing the economic tidal wave that flows google's way, for the better.
How would this work? There are many possibilities. Here are just 2, one of them individualistic and consumer oriented, the other social and community oriented:
1. Google gives coupons to users as compensation for harvesting their metadata. So if a user performs say 5000 google searches per year, google gives them coupons for say $500 which they can redeem on purchases made with vendors who advertise through google.
2. Google funds journalism by paying the metadata harvesting fee into a trust administered by an independent org that facilitates journalism. Like PBS or the BBC but not owned by the British Government, owned independently by a worldwide org that is affiliated with journalists associations.
The logic behind #2 is that pre-google, advertising revenue funded the news business. Post-google, the news business has shrunk by more than 50% of journalists and so the quality of the 4th estate has suffered.
This. Youtubers get compensated, why not ordinary users. Search, shopping and social media is work... It's the current year and billions are stuck home , many without even jobs.
> How would this work?
First you need a legal framework that allows users to sell their metadata. GDPR makes it illegal, but that's wrong. If this data has value (apparently it has a lot of value) , it should be tradeable for services.
Incidentally, i think one of the benefits of the DoJ action is that it gets people talking about solutions that people called 'crazy' until yesterday.
Where does the GDPR make that illegal?
It makes 0 sense for journalism to shrivel in the age of the internet, it's just dying because its revenue (selling ads, newspaper prices were never what covered expenses) was disconnected from the value it provides.
It's a public service, we should have a way to fund it somehow.
And I don’t like 2 because the news isn’t the only industry being negatively affected by Google’s monopoly.
Interesting question, with a lot of scope for answers. Different people have different feelings about privacy. The Nobel prize for economics this year was awarded for auction theory.
Google's ad revenue model implements a form of Vickrey-Grove auction.
https://www.adviso.ca/blog/tech/google-adwords-what-kind-of-...
One possibility for "pricing privacy" is for google to compensate users for their metadata using a price that is derived from the commercial gain yielded from harvesting the metadata. Google already has a vast IT infrastructure for managing every individual's metadata history. The price dividend could simply be an output of that IT infrastructure, paid to users in some form.
But then the compensation users receive depends on whether Google gets a good deal or not, and it still leaves Google with immense power and control.
And what about the users who value their privacy higher? They might decide not to use Google, but that's only useful if there is an alternative that can actually compete with Google. Maybe the compensation thing will increase Google's costs and make it easier for others to compete, but that seems unfair; why should Google have to compensate users, but not others? Any ruling would have to apply to everyone, because that's how common law systems work. So if everyone has to compensate users, then no one will gain an advantage over Google. So the end result is just that Google makes less money while still maintaining their monopoly, and innovation doesn't happen.
But note that you can not sell your organs, because that will (and does on the black market) quickly create highly coercive contracts between individuals, or worse organ robbery. You can, however, donate your organs.
I would see a similar scheme for your digital data. That, most of it cannot be sold [1]. It can be donated by your positive assent, but the receiver (businesses and such) are highly regulated to ensure they don't exploit you or society using that data.
[1] I concede that there might be certain types of data that could be sold, though I cannot think of an example.
I find it greatly detached from the reality of what is known about the historical personalities that framed the constitution to imagine anything other than them being absolutely horrified.
Privacy has been an assumed right for a LONG time. And if you, right now, went up to someone on the street and offered them $20 to learn their GPS whereabouts for a day, to learn what their voice sounds like and record it, to learn what their face looks like and be able to search it in photos, to learn when they are menstruating, what kind of condoms they like to buy, who all their friends are, who they hook up with, and where they live, hiw many kids they have and where they go to school, I assert that you would find the vast majority of the population would be deeply offended you even had the audacity to ask such a thing, and a few might even call the police.
Privacy is deeply ingrained in our basic social interactions and assumed by most. I think most people just don't have a clue how much it's being invaded because the collectors of such information are careful to hide it behind free gifts, and bury the truth in long, overly broad, complex, highly veiled and obscured EULA's. Then they are more careful still to try and never quite reach the level of using that information in a way that creeps people out by carefully hiding relevant advertisements in enough noise so you won't notice, trading and selling all that information in secret, all while carefully trying to systematically normalize ever more collection and otherwise creepy uses of that data.
Eh, I don't think that's a reasonable take. They're much more rooted in property rights. I guess you could argue they're just restatements of each other. Privacy applies to your person, while property rights apply to your stuff.
But notice the 3rd amendment specifically says "owner", not resident or something similar. For example, if a apartment building owner wanted to allow a soldier to live in your apartment, you couldn't stop them. Obviously any sane lease wouldn't allow that but my point is you couldn't make a constitutional argument against the government saying it's a violation of your rights.
Also, SCOTUS didn't extend the 4th amendment to imply privacy until 1967. Before then it only applied to property. https://en.wikipedia.org/wiki/Katz_v._United_States
(needless to say, I completely approve of the idea of government controls on what other companies can do with my data without my consent)
On the other hand, the main value of your meta data and social media content comes not on it's own, but as part of the larger database. Each individual person could be locally happy with the sale of data (as are sellers of organs on the black market), but globally, you quickly land in a very suboptimal nash equilibrium, which is plagued by coercive relations.
[1] Though recent trends in the reality tv market or adult streaming market are worrying.
Currently, users give it away for free, or else they can't access google's services.
Google uses the metadata to supercharge its advertising business, which outperforms all other forms of advertising previously invented, because google knows that ads users will click on as a result of analysing their metadata. William Randolph Hearst, Lord Northcliffe, Ted Turner, Rupert Murdoch and all the other media barons who went before google could only ever dream of this sort of control.
Google's advertising revenue is in the order of $162 billion per year.
More accurate to say that users give it away in exchange for Google's services.
Assistant knows a hell of a lot about me and does a pretty decent job keeping up with what I ask of it.
Seems that way. I pay around $0.30/click on keywords with no competitive bids. When I switched to paying per conversion value, I might get days with good value followed by days with terrible value, so that monthly it balances out to maximize my cost/conversion.
Thoughtful article, great writing but help me here: How can a natural monopoly (what ever that means) ever be a good thing?
Imagine if we'd let one company extract, refine and distribute all oil. Or even a monopoly for any one of those three.
For example, imagine operating the first power plant in a desert region. If it's expensive to set up power plants, then there won't be much point in encouraging competition; no one else will want to set up one. The operators of the power plant have a natural monopoly.
It would be better to spend the resources required for a second power plant on servicing customers elsewhere instead of fighting over the same territory — so the monopoly is "good" in the sense that it disincentivizes a bad allocation of capital.
Generally, natural monopolies get regulated (e.g. a power authority blessed by the local government operates the power plant instead of a private firm).
My argument against breaking up Google would be they face search competition via Amazon, Facebook, and even Apple. But I can see both sides here. That’s what makes it an interesting case!
How can a natural monopoly (what ever that means) ever be a good thing?
https://en.wikipedia.org/wiki/Natural_monopoly
> How can a natural monopoly (what ever that means) ever be a good thing?
Regulating the local phone company responsible for "last mile" wiring as a natural monopoly is why telephone poles no longer look like this:
https://i.pinimg.com/originals/ea/a8/84/eaa884f82bb111c77c4b...
Also, you probably don't want construction crews digging up your street every time a neighbor decides to switch to a competing water/sewer service.
edit: grammar
We can have an interesting debate about whether "capital costs predominate". Many have said here they do for crawling/indexing. I don't think that's so clear
The bigger issue, I believe, is that Google, in its present form in 2020, is not a search engine. It is an ad delivery engine supplemented by organic results which they really do not want users clicking on. The bulk of the searches normal users perform are served ads and those users click on those ads.
Really no analogy here close to something like power lines, a sewer system, or roads.
Well, I disagree.
1. It is quite difficult to determine whether a search engine is good or not - without massive resources for surveying. Also, remember that even if one likes the results - they typically don't know which potential results they are _not_ seeing, to compare against.
2. Google is known to suppress search results on political grounds. Specifically, Socialist and anti-Imperialist news and commentary is partially suppressed. This was covered here on HN a few times already, e.g.: https://news.ycombinator.com/item?id=14975338 ; Google officials mostly keep quiet about this of course, so the breadth and scope are not entirely clear. There's also an NYT story about this and other issues, but it's paywalled: https://www.nytimes.com/2017/09/26/technology/google-search-... . I would not at all be surprised if other non-mainstream political views are being suppressed as well.
3. Inherent, over-arcing Biases in Google's ranking: This is a different issue than direct suppression or blocking. Here's I myself am being very anecdotal and speculative, but I have the distinct sense that larger websites get ranked higher by Google, so that for most of a typical person's searches they are likely to stay within a the sphere of, oh, I don't know, a few thousands or tens of thousands of larger sites. Again, anecdotally, when I search duckduckgo I seem to get into the "tail" in terms of site size faster. I may be totally wrong about this but as I said in (1.) we can't really know.
Milton Friedman on Anti-Trust and Tech: https://www.youtube.com/watch?v=mlwxdyLnMXM
And going in second time around would be even more difficult
That is likely to cause significant issues in this process. For example, there is absolutely a political element that seeks to attack the market power of large technology companies because they want to force more control on public speech, and some who want to attack the power in order to explicitly prevent that.
I don't want tech regulation because I don't want to be ruled over by complete fucking idiots on subjects they don't understand.