Hacker News new | past | comments | ask | show | jobs | submit login
Google says it doesn't monopolize digital ad market – senators don't buy it (npr.org)
338 points by belltaco 13 days ago | hide | past | favorite | 217 comments

What so many people fail to realize about these issues vis-a-vis Google, it’s not so much the issues of Google having a monopoly or not on the ad market...in either case google is a dominant market incumbent that uses its position (in this case web traffic, data and online ad platform) to unfairly compete and stifle competition.

Google has often used data acquired through their market position to start subsidiaries to unfairly compete with their ad customers.

The reason it’s unfair is because Google has not just the ad data but the web search data, this often results in a Google ad customer going from #1 google Organic search result for key terms, to #2 to Google’s competing subsidiary.

Worse from going from #1 to #2 to a google product, the natural instinct to save the business is to increase ad spend to be sure you are still the #1 ad to our place the google at the top of organic search, of course ad spend goes right into the pocket of your new competitors core business anyway, and in many instances googles subsidiary will start bidding for your same keywords so they literally can’t lose competing with your business rather the have basically acquired an off form of rent seeking equity or they kill your business and become the market incumbent.

Of course the kicker are those instances a google subsidiary gets a custom tool at the top of google results (such as Flights) above both organic and ads.

Words get abused a lot, and for some of them, they do not have a clear-cut, universally agreed upon meaning. "Monopoly" is one of such words (others candidate include: freedom, democracy, justice).

When random person X complains about, say, Google having a monopoly, there are two possibilities:

- Person X means something along "Google has a dominant market position and abuses its power", slightly abusing the meaning of "monopoly".

- Person X means Google is literally a monopoly, that it is not possible to get online ads otherwise and does not know that firms such as Facebook exist.

Somehow a lot of people choose to believe interpretation #2 is true, and spend a lot of time debating whether this or that company is a "monopoly" as if it is somehow more important than the substantive issues.

> Somehow a lot of people choose to believe interpretation #2 is true, and spend a lot of time debating whether this or that company is a "monopoly" as if it is somehow more important than the substantive issues

Not to accuse anyone here of this, but intentionally doing what you describe is an disingenuous but effective way of shaping the conversation to discourage scrutiny of more general kinds of anti-competitive behavior. I'm sure it's part of any monopolist's PR strategy (or large business engaged in anti-competitive behavior, to be more precise).

Are you saying OP is a Google PR insider accusing Google of abusing market privilege while ensuring the perfect definition of “monopoly“ isn’t marred by inaccuracy?

No, I'm pretty sure they're saying that debating the semantics of an argument rather than the merit of the argument is a tactic often used by bad-faith actors who seek to reshape a conversation to their advantage by discouraging deeper scrutiny of their position past "yeah, well, they keep using that word and I don't think they know what it means."

The problem with this analysis is that “monopoly” is only one part of antitrust law, and there are plenty of other abuses of dominant market positions that don’t require a monopoly to be present.

Furthermore, in the legal context the meaning is more clear-cut than you make out, though there is significant room for debate. I would not put it anywhere near “freedom” in terms of loose definitions. There are rigorous tests for determining whether a company should be deemed to be a monopoly within a specific market (which requires both a method for defining the boundary of a market, and what “dominant” means.)

This is also different by country, see

https://en.m.wikipedia.org/wiki/Sherman_Antitrust_Act_of_189... Vs https://en.m.wikipedia.org/wiki/European_Union_competition_l...

I think your general point is to stop focusing on monopoly and get more familiar with all of the different antitrust provisions, to which I would wholeheartedly agree.

This doesn't clear anything up though. Who gets to define the boundary/scope of the market being measured? It's pretty easy to manipulate this to make it seem like a company has a "dominant" share even if that's not the case from the consumers point of view.

Take the anti-trust case against Microsoft, for example, the market was defined as that for computer operating systems for stand-alone personal computers using microchips of the kind manufactured by Intel. This left out not only operating systems running on Apple computers but also other operating systems such as those produced by Sun Microsystems for multiple computers or the Linux system for stand-alone computers. In its narrowly defined market, Microsoft clearly had a “dominant” share.

Also lots of people use Amazon to search for products so it's not clear to me how there is a market failure here.

If your point is that "monopoly" has a different meaning in a colloquial/political context and in a legal context, that it relies on non-obvious definitions of what the relevant market, what dominant is, and that it varies by country, I do not think it strongly supports the assertion that it does not have a loose definition.

I was mainly arguing against

> "Monopoly" is one of such words (others candidate include: freedom, democracy, justice).

-- there is a much more rigorous definition of Monopoly than the other ones you listed. You're claiming that it is some undefined concept that varies from person to person, but I think that's only true in the sense that "encryption" or "space-time" has a loose definition; most lay people don't understand technical jargon, and often misuse the jargon when they think they know what they are talking about. But that doesn't mean there _isn't_ a well-defined meaning that is agreed upon by the specialists that actually work in the domain in question.

The other words you listed do not have clear-cut legal definitions in the same way; they are not technical jargon defining criteria for enforceable regulations with tests and case-law backing them up.

Fair point.

I am the CMO for a company.

Google is the ONLY ad provider that give results, all others don't even get any clicks.

How that is not a monopoly on the classic definition? Every time Google has a bug and screw with my ads, the revenue of the company I work for tanks hard, and I am yet to find any solution for it.

That depends heavily on your industry and obviously isnt true as a rule. Google gets 37% of ad spend, Facebook ~20% and then everyone else combines for the last 40%.

It's true that if your looking at targeting users searching for something that it's basically just Google. But people are increasingly using apps and specific sites over general surfing. Reddit, tiktok, snap, etc. all have their own ads and collectively represent major competition.

In our tests, we get results from Google but ads on Facebook produce trash results.

You bid on ads so if the results are poor on fb they must be working for the competitors.

You are assuming things that aren't true. There is no competition for us.

Google is the ONLY ad provider that give results, all others don't even get any clicks.

Being effective doesn't make you a monopoly.

I don't think he meant that other companies are bad at getting clicks. I think he meant his company relies on search engine driven ads. Google has a monopoly on the search engine. There are many companies that need to rely on search engine traffic, not social media, for online advertising.

Keep in mind that search engine traffic is enormous, and can fairly be called a "trade" or "commodity/service" that can be monopolized in it's own right.

I don't think anybody is saying that google has a monopoly on all advertising.

But I don't think this tracks unless you treat search as something special. If I own a billboard then I have a monopoly on that billboard. Presumably I built/bought that billboard because I knew that the ad space would be valuable and that companies would need to pay me for it.

Google owns extremely valuable ad space but that to me doesn't make a monopoly just because Bing or Facebook's ad space isn't as good for certain companies.

And if only one company owned all the billboards in the world, that would be a monopoly too.

Monopolies are traditionally assessed in terms of product categories, not entire industries. Nobody ever said "standard oil isn't a monopoly, because you can also buy coal fuel"

And practically any company can build as many billboards as they want. Online ads aren't restricted by physical space. Google just happens to own most web traffic because they're really good at search.

Maybe a better analogy is Google owns the biggest road system (search) and all land beside it for billboards (ads). Very few people use other road systems.

Now, one might say "But users can choose other search engines." But users are the resource, not the customers. Right now, Google controls most of these resources. And the very nature of search means they're not sharing access with competitors.

> And if only one company owned all the billboards in the world, that would be a monopoly too.

You mean clearchannel? (Now iHeartRadio)

It makes sense to treat search as something special; when people are looking for a service, they search for it. They won't see ads for that service on, e.g. social media because social media won't know they want that service.

Google search ranking is critical and Alphabet knows it.

If I google search for "koop ebike" and you're not on the first page you might as well close your shop.

They do have exclusive access and control over the source of data that bring you that effectiveness though. I'm not saying anyone else should have access to Google data, but the fact that this is what make it effective and that essentially no one else can reproduce it, that kind of make it a monopoly.

At a point when the cost of entry is too high and it's not humanly possible to beat it, isn't it essentially impossible to compete, thus make it a monopoly?

> They do have exclusive access and control over the source of data that bring you that effectiveness though

And what data would that be? All search engines crawl the same web.

What data? We are talking about ads. It's not the crawling data which help in targeting, it's what people click on, it's what people read, it's what people do.

Being the search engine, if I search Asus computer, do you think it's the fact that there's Asus computer in the crawled content that will be used for targeting, or the fact that I searched Asus computer? When I click on the first link that say "Asus computer for cheap", do you think it's the existence of the page "Asus computer for cheap" that will help in targeting my ads, or the fact that I clicked on a page where it's about the value of Asus computers?

What matter isn't the content, it's knowing what each individual click and read. That data come from Google Analytic and Google Ads.

Why do you think Android exist? Why do you think Google Chrome exist, why do you think exist. You are really mistaken if you believe they are for the good of customer (even though it's true they do bring some good), they are there to get that data and be the one that exclusively get that data.

If you want to compete with them, you'll need to beat Android, beat Chrome, beat their search engine, beat EVERYTHING that make Google what they are. Not saying it's not possible, but the cost of it make it nearly impossible.

You realize that search engine itself drives massive traffic right? Even with the dumbest adds, that is billions of eyeballs concentrated on one page multiple times a day. That already makes a very steep barrier of entry.

> You are really mistaken if you believe they are for the good of customer

You are responding to a bunch of strawman arguments I never made. All those data you mention is plausibly increasing their targeting and revenue, but claiming that it is the biggest differentiating factor is a stretch and I don't think is likely. Facebook has more ground truth demographic data that Google has to infer or outsource for their billions of users, yet they weren't able to dominate competition solely on that basis.

> If you want to compete with them, you'll need to beat Android, beat Chrome, beat their search engine, beat EVERYTHING that make Google what they are. Not saying it's not possible, but the cost of it make it nearly impossible.

No, if you want to compete with them, you just need to have a damn good product that billions of people use multiple times a day. Turns out search engines are a pretty good fit for that. You can have a platform that can profile users to their genes, you still need raw the engagement numbers to make money out of it. At the end of the day it is accumulation of commissions for bits and bobs sold over internet ads.

> No, if you want to compete with them, you just need to have a damn good product that billions of people use multiple times a day.

* using billions of people clicks to infer what the best results for every single search query

If crawling was enough, we wouldn't be all here arguing that Google is too big. It's not the tentative that are missing.

You may want to contact the antitrust regulators, testimonies such as your are highly relevant to this case.

I'd love to see the specifics of these players. Facebook targeting is actually more effective in many cases then google (I don't even use facebook personally or have the app myself). A lot of the claims by folks fall apart when you look deeper (ie, scam review sites complaining about google rankings etc). The CUSTOMER should benefit - I don't give a crap if some scammers get ranked down.

they are also untrue in general. in some markets social networking ads work, in some search ads work, that's to be expected - you don't expect eg. ad placements by the train company to be always effective regardless of what you're selling.

Don't worry Google does not need your defense, they'll have plenty of very highly paid lawyers and lobbyists ready to do that.

I don't care about Google. I simply know from experience this is untrue. If anything, I am defending Facebook here.

I'm surprised. I thought Facebook would have been somewhat effective. Can anyone point out any use cases of when Facebook and Google work and don't work?

FWIW my dad ran a few FB ads for his SME and it was alright, although nothing stellar either. He didn't run on Google though so I can't really draw a comparison.


Please don't cross into personal attacks in HN comments.


Meaning isn't dictated by a language academy and may change over time. Debating about the precise differences between monopoly and monopsony serves more to distract from the core issue: abuse of their market position.

If the allegations are true then I can't see a solution that does not involve breaking Google up. The incentives to cheat are likely just undeniable over a time scale of years.

I think the argument that Facebook and Google compete in the ad space is tenuous at best. Just ask a digital marketer: almost all of them consider FB and Google to have separate purposes and markets.

So fine, Google doesn't have a monopoly on online ads, but they do have a monopoly on search ads.

From the government itself[1]:

> Courts do not require a literal monopoly before applying rules for single firm conduct; that term is used as shorthand for a firm with significant and durable market power — that is, the long term ability to raise price or exclude competitors. That is how that term is used here: a "monopolist" is a firm with significant and durable market power.

[1] https://www.ftc.gov/tips-advice/competition-guidance/guide-a...

I've seen this quoted a lot lately but saying that a literal monopoly (as in, a single company with 100% market share) is not required doesn't really say much about what is required, which depends on the specific violation being alleged.

Some violations (price fixing, bid rigging) are considered so egregious that they are illegal regardless of market share.

Other violations (tying) require some degree of market power.

Violations related to monopolization or attempted monopolization do require monopoly power (or the dangerous probability of obtaining it it).

Monopolies are a well-defined theoretical concept, but in practice it's about market power. When one entity has overwhelming market power it's afforded with options that harm consumer welfare.

Anti-trust isn't about defeating theoretical monopolies. It's about protecting consumers, new businesses, the market from abuses through overwhelming market power.

So your first possibility seems right on the money and appropriate.

You may be over-indexing on a prescriptive lexicon:


To be clear if anyone has any doubt, antitrust law talks about firms having "market power". That's all that is required; the ability to move the market in bad ways. A monopoly (strict definition) is not in any way required to violate antitrust law. You could violate antitrust law with a 5% market share.

People wanting to nitpick about the word monopoly should be educated (first offense) or mocked and shunned (repeat intentional offenders).

Google has a 73% share of online advertising, according to a random study from 2019.

People who abuse the word monopoly should be educated (first offence) and mocked or shunned (repeat intentional offenders).

It's almost impossible to have rational discourse these days because accurately describing things is no longer required (and if you care to try you are "mocked and shunned") so everyone is arguing without a common set of understandings or facts or even language that is consistent.

Kind of pathetic.

Your water company may be a monopoly. You may have no choice but to use it (in cities you cannot even drill your own well or in some cases are banned from collecting rainwater). That is a monopoly. As someone who grew up using a well, this is a clear difference and impact.

People need to use the words antitrust or anticompetitive more often, just because you could claim there are competitors, doesn't mean the market is fair to competition or consumers.

If someone wants to make a charge against Google, then by all means do it. But the people making the charge have a responsibility to communicate their position clearly. Complaining that the the wider population is not following along with an intentional abuse of language sounds like a bad faith argument, especially when we're talking about a well established term like "monopoly".


Again, mindless arguing what exact word fit or does not fit the situation detracts from the real issues.

If the only thing the word "monopoly" does is spark a discussion about the word then why use it? Why is it so important that Google be a monopoly for their behavior to be bad? Do we really have no other words to describe business practices that we don't like?

> or we really have no other words

There are other words, but it is a bit of a mouthful.

Because of these exact issues, the way that I describe the situation is usually as follows "Company X has significant market power, and is engaging in anti-competitive actions".

I have yet to have someone object to that definition.

Because it is generally accepted that monopolies are bad and that not-monopolies are ok. It is much easier to sell the simple idea that "X is a monopoly, so split X", than to make a long case that "X does Y and Z which meet ABC definition of unfair use of market position, bla bla bla, so split X". Conversely, for a company in the wrong, it is easier to say "Of course monopolies are bad, but we are not a monopoly!".

>n either case google is a dominant market incumbent that uses its position (in this case web traffic, data and online ad platform) to unfairly compete and stifle competition.

It's not that clear how much that helps. The places where Google won (Maps, Gmail, Chrome) they had a pretty clearly better product. Other places where they didn't (Duo/Hangouts, Plus, Gsuite, Chromebook, GCloud) they don't dominate the market.

> It's not that clear how much that helps.

It's clear that it helps enough to subsidize the entire rest of the Google operation (enough to run a more computationally and sophisticated business like search). AdTech is all about targeting by extracting context and minimally about tracking events.

Leveraging search to power ad targeting is the most efficient means. I'm not sure how this is a topic of debate other than to split hairs on what measuring stick to use.

Translate at least seems to have some people ignorant of the alternatives convinced that they have a clearly better product — “no competitors that even come close in terms of quality” when they objectively don’t. https://news.ycombinator.com/item?id=23975001

The claimed “objectively” better translator is DeepL.

You need to read like 8 paragraphs deep to get to that and there is no real explanation as to why it is objectively better - just a link to an article where some journalists apparently thought it was better.


The bottom of the page you linked shows ten articles, six of which praise DeepL as better than the competition and four which praise it without comparison. The least positive is probably Wired’s: “WIRED's quick test shows that DeepL's results are indeed in no way inferior to those of the high-ranking competitors and, in many cases, even surpass them. The translated texts often read much more fluently; where Google Translate forms completely meaningless word chains, DeepL can at least guess a connection.” — which is a long-winded way of saying that DeepL’s worst translations are as bad as Google Translate’s, but many of them are better — making DeepL strictly better overall. Of course, any individual review must be made by subjective humans, but you can find as many corroborating comparisons as you like and every review I’ve seen has either described DeepL as better or at worst as bad as Google Translate, and the former is exactly my personal experience after using each for years. We can also compare the results of sending your comment on a round trip through the top language on each:

DeepL German: “The allegedly "objectively" better translator is DeepL.”

Google Translate German: “The alleged "objectively" better translator is DeepL”

DeepL Spanish: “The allegedly "objectively" better translator is DeepL.”

Google Translate Spanish: “The best translator "objectively" that is claimed is DeepL.”

As a native English speaker, the DeepL results seem clearly clearer and cleaner to me. If broad consensus isn’t evidence to you, you could also compare quantity of languages instead of quality of translations, in which case you’d find that Google Translate is missing languages that Microsoft Translate has. Either way, Google certainly doesn’t have “a pretty clearly better product.”

I experimented with testing with this line of Japanese.


DeepL: It is known to be the most ground-breaking of all the bovine cats

Google: Known for being the most remote of all quolls

Excellent Comment

Let's consider a few more examples of abuse of market dominance

1) Retail Me Not

Google through Google Venture secretly invested in them and didn't disclose that

Then for next 3 years, Retail Me Not was #1 for EVERY SINGLE deal and coupon search term on Google

Is that an abuse of market dominance? Pretty Sure

Is it illegal? According to US Anti Trust Law - not sure. A lawyer can step in

2) A market we are in

The largest 2 players (us and another company) are #40 and #41 when people search for the biggest key word

Not #2, Not #10

At #40 and #41

Who is in the Top 10?

companies that advertise A LOT on Google, even though they are distant 4th and 6th in the market

Not only that, they do it in blatant ways

Last year, they pumped up organic traffic to one of our competitors THIRTY TIMES over night

So, the top two companies (another company and us) saw ZERO CHANGE

However, a company that is a distant 4th in the market suddenly got its organic search engine traffic increased 30 times

Pretty sure in 3 or 4 years it will be revealed that Google invested in them or did some tie up


Google is basically using Search as a weapon

this comment acts like there is some person on Google's side manually curating the search results. why is it Google's fault that your competitors have better SEO teams?

I totally agree with the idea of your post. Just wanted to share a sort of funny tidbit about the specific example you chose: I work in fare marketing for airlines, and google flights is actually a bit of a boon for us because it drops you directly into the airline booking engine with pre-filled info.

Before Google Flights we were mostly competing against Expedia/Kayak/etc. Now Google Flights indirectly gives the airlines the top rank again, and we're competing against the OTA's for the secondary spots with ads and SEO. I can also anecdotally attest that Instagram ads work very well for airlines. (Well... worked very well, pre march 2020 and all).

I do agree with your main sentiment though, and the case could totally be made that google flights is doubly screwing over the OTA's now. Just wanted to share some interesting anecdata.

As detailed on the site css-tricks.com, a popular HN item, there was a time before "#1 google Organic search result" meant anything. That is, the best search engine at the time did not try to order the results. The search was accurate and comprehensive but there was no preferential ordering according to "popularity" or a "secret" algorithm. That search engine was Alta Vista. Raw results.

I can remember combing through page after page looking for the results I wanted. Try that today and Google is likely to temporarily block you, accusing you of having "strange activity" coming from your computer, or looking like a "bot".

The more Google has tried to "solve" the problem of finding the appropriate results for the particular searcher within the raw results returned from the query, the more they have manipulated the raw results to suit their own perspective. There is no option to get those raw results and search within them the way you want. There is only Google's way -- reorder them according to Google's "secret" processes. Do not dig into subsequent pages, stay on page one. This necessarily carries a Google bias. It also creates a "winner-take-all" contest to reach page one that Google's exploits for selling online ad services.

In other words, not only does Google manipulate searches through deciding the "1st search result", it actually created this concept of "1st search result" with respect to web search. The idea that Google could guess the "correct" search result, reorder the raw results and place the "best" one at the top of the first page of results was in fact Google's idea.

Thanks to Google, reordering the raw results has become commonplace so no one really thinks about this anymore, but it is this (biased) reordering that allows Google to profit.

> google subsidiary gets a custom tool at the top of google results (such as Flights) above both organic and ads.

Searching for flights is still a search. You wouldn't expect the first result of "cat pictures" to be Bing's search results for cat pictures, so it's also natural that the first result of "flight nyc la" would return actual flights as its first result would return actual flights rather than a link to Expedia.

That is illegal under antitrust laws.

Google was fined billions of euros for doing that.

Let me make it really simple: The problem with the tech industry is vertical monopoly, but many people want to just think about horizontal monopoly.

Especially where tech has the best margins, we end up with horizontal monopolies over time. eg In idiocracy, "Welcome to Google-Costco, we love you."

Not just web search, also android, google home, google/android pay, gmail (even if you're not on gmail you send to/from), youtube, 3rd party android apps running analytics, 3rd party websites running analytics, chrome... the list goes on. Far more reach than any other competitor could hope to have when making a competing service.

good point, when your ad funnel is almost the size the internet(search, gmail) ... and traffic gravity of a black hole.

So whats the solution? Make it illegal for any search company to build any other product?

Can you point out any specific examples, and why Google would forgo that ad money?

What ad money do you think Google is forgoing in this example?

There is an example at the end of the comment you are replying to, flights. Two other examples that took me only a couple of minutes to find: google 'smart thermostat' or 'corporate email providers'

Also Yelp with Google Reviews / Google MyBusiness

The UK recently evaluated the online ad market and found that Google and Facebook commanded 70% of all online ad spend in the country [0].

Their large pool of advertisers also helps them command a higher return per search compared to other platforms, allowing them to outbid its peers to become the default search choice on devices, e.g. Apple products. Their lack of privacy sensitive policy also helps their ad targeting.

Generally to me it seems hard for new entrants to make any inroads for market share when there's a self-perpetuating cycle like that.

[0] https://www.gov.uk/government/news/cma-lifts-the-lid-on-digi...

Fb could disrupt google because it had the impressions first, so that is the dimension to think about. Adspend and advertiser pool is a by product. Disclaimer: current fb, ex google employee.

I'd think that if someone has the intention to buy but doesn't have a site in mind, they'll usually use Google: like "best mortgage rates" or "american girl doll clothes". I can see FB ads being better for raising interest in a product to a targeted demographic, but I'd assume that the Google ads that are very close to the point of sale are much more valuable. What am I missing?

It's 2 different things to me. Buying an ad on Facebook is an attempt to get a specific ad campaign in front of specific users without the user specifically requesting it. Buying an ad on Google means getting your advertised item in front of a user as a result from a user's search. Either way, the user is now just one click away from getting to your advertised product, so I'm not seeing how Google is any closer than Facebook is. Facebook is passively placing your ad in front of a user, where Google's is active from the point of view of the user's actions. That's just for ad placement within the Google Search. I'm sure Google is doing the same passive ad placement in its other ad distribution arms.

Fb is also a problem, so relying on them isn't going to solve anything. A cartel or oligopoly is not a better situation than a monopoly. For there to be a healthy market, it needs to be possible for small business to compete. If that's not possible in a market, then it needs to be regulated like a public utility.

> If that's not possible in a market, then it needs to be regulated like a public utility.

What I don't understand here is does this not just hand Google the monopoly but now with government protection? Because they're already in place, so who's gonna offer this public utility?

We see similar situations with, e.g. cellular carriers in the states, where wherever you live, you only get not-even-a-handful of choices.

And how does a country regulate an international "utility"? For Google to maintain its global position, it needs to consider not just local regulations, but also those in international markets. Public utilities so far have always had a limited area of operation.

It really depends on what the regulation is. If it's something to the effect of making a siloed company for AdWords that needs to offer FRAND terms to everyone, plus Alphabet is bared from doing its own ads, and has to contract it out, then AdWords Inc could gather more of the market and become even more dominant. This would still be a win, because the new, more dominant, advertising platform would not have any other businesses to anticompetitively promote. Maybe you would get competitors offering different terms, but that's probably wishful thinking.

A company that operates in a jurisdiction is regulatable by that jurisdiction. In this case, if Google weren't physically present, you sanction advertisers and local sites that use the ad service. In cases where there's no economic connection, the option is to force ISPs not to carry that traffic (and all the cat and mouse that entails).

Also, a lot of mobile carriers are international companies. In each country, they follow the local regulations. They have some economies of scale that may make it easier to compete with a network that's only in one country, but it depends on the size of that country how big the effect is.

Regulation like a utility doesn't mean their has to be only one gatekeeper. It could just more rules once one becomes so dominant they are a defacto gatekeeper.

Doubtful this would be done in the same way everywhere. Though big markets like the EU or US may regulate at the federal/union level to ease the burden on interstate commerce.

I could imagine ads being run like a utility from a purely technological pov. I'm not sure I like that image however, nor do I think it has any chance of happening as all major companies would sabotage it.

but back to the original idea of it being a utility: Ads are imo user hostile and harmful to a society. They do work, which make them invaluable in capitalism... But they're effectively mass manipulation. I don't think we will ever be rid of them, but a world in which this is considered a utility is extremely dark imo

Facebook has the impressions because in many countries they have a de-facto monopoly in social networking, which has become a large part of all web trafic. Google will cover the rest of the net almost unchallenged. So it's not even a duopoly, it's a segmented monopoly.

Every company is a monopoly based on how you define the market. Why should we focus only on online ads ? Why not look at the entire ad spend and then see what % is controlled by Google ?

By some logic the coffee shop in my block has a monopoly on coffee if their market is the block. They are irrelevant if you look at the entire city.

Which is why market definition requires you to look at substitution. If there is another coffee shop two blocks away and your coffee shop raised prices significantly, people would start going to the other one, because they're strong substitutes for one another. Whereas if the nearest coffee shop was twenty blocks away, likely they wouldn't, because that's too far to walk and it doesn't make sense to pay $2.75 to ride the subway to save $1.50 on a cup of coffee. So the block is too small but the whole city is too big.

> Why not look at the entire ad spend and then see what % is controlled by Google ?

Because they would be reaching different customers, such that if the advertiser moved a dollar from e.g. online to print, they would lose customers, because the print viewers have already seen their ad from the existing print advertising they do, and the online viewers (who are different people) haven't. Which means they aren't adequate substitutes for one another.

That's too relativistic. If only AMZN, GOOG, and FB gain during coronavirus, this clearly means there's need for action considering the perspective of online-only advertising going forward. Especially if the success of TikTok as the only competitor in the space has resulted in government influence.

That sounds like an attempt to maintain "never my fault" consistency akin to stating we need to beat all people at traffic stops instead of admitting that beating people at traffic stops to look good on crime in the first place was criminally wrong.

Not sure if I understand what you're saying, but it's not about criminal law, whether anyone did something wrong, or punishment; rather it's about maintaining market access.

What I am saying is that Tiktok's treatment was clear capricious abuse of power and shouldn't be taken as a standard or baseline.

Being a monopoly is actually irrelevant. The anti-trust law in the US focuses on regulating anti-competitive business practices. You can implement anti-competitive business practices with any portion of the market, and you can have an outright monopoly without implementing any anti-competitive practices (and therefor without breaking the law).

But to successfully implement anti-competitive practices, you do usually need a significant portion of market share. Which is pretty much what’s being discussed here, even though it’s not a very accurate use of the word. You can also do it by acting as a cartel, which is another situation commonly (though not very accurately) described as a monopoly.

Bizarre that you are equivocating a local coffee shop to online advertising; the emergence of which gutted the magazine industry, severely damaged and forever changed print news, and was single-handedly responsible for pushing Alphabet and Facebook into the Fortune 50.

Even in that extreme hypothetical I’m not sure it would be considered a monopoly - is there a convenience store on the block? A grocery store? A bakery? The market for “coffee on a given block” isn’t limited to coffee shops.

The only way to have competition is if FB started a search engine or if google bought instagram

FB may already be backing their way into search:


backing into? They've had a search box for ages and they've consistently tried to make sure all content is on their platform (rather than hosted elsewhere).

Same senators who thought Google made iPhones? Not defending Google but I am saying that most of those senators are not the holders of truth when it comes to technology.

That's not what happened. The question was asking why a child saw a family member in an ad while playing a game. The Congressman just held up an iPhone while asking but it had nothing to do with the hardware.

Of course there are numerous factors that may have contributed to the ad (and maybe Google wasn't involved) but they went with the smartass answer and the media ran with that narrative instead. It's not the greatest question but it's much more valid than what people are assuming it was.

Video of the question (and notice the title): https://www.youtube.com/watch?v=wmuROTmazco&t=60

He does specifically say "how does that show up on a 7 year old's iPhone" at 1:30, and those words will be entered into the record, which is probably why Sundar responded with that.

iPhone claims aside, that's just a terrible question. Doesn't know what type of phone it was, doesn't know what type of game they were playing, doesn't give any details on the sort of photo shown or the ad copy shown. By all likelihood it sounds like this was a Facebook ad because of FB's "Social Ads" policy does use profile pictures of people on your friends list in their ads. On top of that, kids are not the smartest or most honest. It could have been a picture of any old man that looks similar and she assumes it's her grandpa. It could be a made-up story entirely.

No, it's not a terrible question. It is a general question about the business of online advertising, something that Sundar Pichai is unquestionably an expert about since he is the CEO of the largest company in the world that sells online advertising, and for all you know it could have been a web game that the grand daughter was playing in a chrome browser.

A reasonable, non-combative answer to this question may have been something along the lines of:

Senator, ad providers have little control over what the content of the ads that are going into these applications is. While I'm not sure what the specific ad that your granddaughter saw was, I can tell you that at google, there is a system where a user can report an ad as inappropriate. Again I'm not sure what the specific ad that you saw was, or if it was part of google's advertising platform, but if it was, you could have reported that to google for review, and I can understand why that would have been an uncomfortable experience for you.

Strong disagree, it's a terrible question, most likely not entirely factual, and there's no reason to dignify it with an in depth answer as it would require breaking down multiple company's ad ecosystems for one question. You're re-interpreting the question as something he didn't actually say. The question, as it is in the official record is "how does that show up on a 7 year old's iPhone".

> Senator, ad providers have little control over what the content of the ads that are going into these applications is.

This isn't really true, and also if interpreted in the most negative way it can be, this is opening up a big can of worms that isn't true. To say that they have little control over whether an ad somehow finds a picture of your grandfather and shows it to you is just straight up false. The only ad network where this is possible is Facebook and the advertisers do not get access to those images or know what names they are using to affiliate with their ads. Only that they can show names from a user's friends list.

I still think the most likely assumption is that the ad showed an old man and the 7 year old assumes any elderly man is her grandpa, or that the congressman made up the scenario based on something similar happening (seeing an ad in a facebook feed with a photo of the grandpa). It's a combative question and deserves a combative answer.

It's like when my mom calls me to say her iPhone is showing an error on the screen, but she can't remember what it was. So I ask her to write it down on a piece of paper and then call me next time it happens. It is always

"A new version of iOS is available for your device. Press OK to upgrade to iOS x.x.x."

Edit: If they were chatting at the bar about how ad networks work, it's one thing. But this is a congressional hearing and there's no reason you should need to potentially implicate yourself on a situation that doesn't even exist.

Let's look at the possible interpretations of the officially recorded question and see if we can figure out what the congressman meant.

Question: "how does this show up on a 7 year olds iPhone".

Interpretation 1:

>"How does this show up on a 7 year olds iPhone"

Senator, the iphone has a component on the front of it called an LED screen. LED stands for "light emitting diode", and the screen itself, which is similar in composition to a piece of glass, contains millions of these LEDs. The software in the phone selectively turns these LEDs (called pixels) on and off in patterns which your eyes interpret as images and text.

Interpretation 2:

>"How does this show up on a 7 year olds iphone"

Senator, there is a complex system of interactions between the user and the device they are using which influences the things that show up on the phone. I'm not sure what specific actions your grand daughter took to cause that specific ad to show up on her phone; one of the goals of interface design is to allow the user to precisely indicate to the device or software system what it is that they wish to see, or have the system produce. Likely there were some actions that your grand daughter took that caused the ad you are concerned about to appear, and some of these actions were likely tracked as a part of an advertising network which takes into account previous actions, location, browsing histo..

<Speaker the senators time is expired>

Interpretation 3:

>"How does this show up on a 7 year olds iphone"

Senator I am not sure how the age of your granddaughter is effecting that advertisements that she is seeing in this specific case, however I can tell you that at google, age is certainly a part of the algorithm we use to select which ads we show to children.

Interpretation 4:

>"How does this show up on a 7 year olds iphone"

Senator my company doesn't make the iphone.

> possible interpretations of the officially recorded question and see if we can figure out what the congressman meant.

This is not how things work legally. If they want better answers they need better questions. It's not about interpreting what they meant but never actually said. Especially when what they said it accusatory of Google and very likely false. He asked a terrible question, and you're bending over backwards to interpret it in the most charitable way possible. That's fine on HN, or in a group of friends. That is not how things work at a congressional hearing. If I was Sundar, my answer would be that without photographic proof I don't believe this ever actually happened.

Cheers to whomever is downvoting my responses. It doesn't change the fact that it was a bad question and it got a deservedly snarky answer. I'm leaving this thread, it's clear most people here don't want to understand ads and just want to be angry at the big corp.

I do make a habit of trying to understand what somebody meant instead of trying to twist their words into something inflammatory.

It’s worked out really well for me so far.

Yes it's lacking too much context. Google's CEO probably didn't intend to be so dismissive but stalled since there wasn't a clear answer, and the statement ended up being snarky enough to move focus from the question to the "old people are bad at tech" bias, which was then reinforced by the media.

It's unfortunate but it shows just how bad the understanding is on all sides, and why these corporations can get away with so much. The most recent hearings this year revealed how much training these CEOs get and how adept they are at maneuvering around any real investigation.

Watched the video, and there was just a random guy dressed up with a mustache and top hat sitting behind the camera during the entire hearing?

How does this happen two different comments about one incident captured on video.

Because people don't watch the original video, they read spin articles about the incident.

Read the comments on the video. People don't really comprehend much of the content they consume and are easily steered by headlines.

The kid dressed up like the monopoly man in the background was a welcomed surprise.

No... If I recall right, that was Steve King. Steve King was a congressman, not a senator. Maligning a single congressman does not invalidate the basic point that a lot of senators are making.

Or the entire Senate calling the wrong CEO in to testify?


Can you name one positive law with regards to tech that the government has passed in the last 20 years?

During the last hearing, many of the representatives were more concerned about Google and Facebook being “unfair” to conservatives. Well that and grilling Zuckerburg about Twitter’s practices.

That's not what the article you linked says. To the question though there really haven't been many federal laws regarding tech at all. It's kind of hard to find an example of anything that directly focuses on tech and has been around long enough to make a judgement whether it was good or bad.

There have been plenty of laws. The two most famous ones are the DMCA and COPPA.

And then there are the ones that were only stopped because of massive protests like SOPA and PIPA.


Not to mention regulations that killed the concept of net Neutrality.

There are also constant mumblings about passing laws against encryption that the government can’t decrypt.

The Snowden leaks should scare anyone about giving the government more power.

It was in 1996, but there is also the CDA.


> most of those senators are not the holders of truth when it comes to technology

They don’t have to be. Their staffers do.

I’ve worked with a number of staffers on these questions for about three years now. They’re some of the smartest people I’ve worked with in my career, which has centred around Wall Street and Silicon Valley.

So why aren't they the senators instead of the morons they back?

Because a senator should know how to listen to others. They cannot be an expert on everything. Heck, these folks are making rules about medicine, pollution, car safety, tech issues, teaching, school lunches, interstate highways, commerce, military, child toy standards, what sorts of meat to allow in, and lots of other things I can't think up off of the top of my head. No one can be an expert on all of them - why we would suddenly expect them all to be an expert on this is beyond me. IIRC, an education in law is pretty common, which would be expected.

(I know the system isn't perfect and not everyone listens to educated voices and in a perfect world, there would be a range of knowledge in the elected body itself).

Edited for a missed comma.

I think you missed the point - representatives should be expert (at a minimum) in HN interests or come from the same demographics (better) as HN commentators. Anything else means they are "morons", to use the word of the previous commentator.

You need to be a certain type of animal to be a successful politician. Especially in the US where the power per person in the senate especially is enormous.

The house of commons has 600 more MPs for 5x less population.

Because being smart is not a golden ticket or the only quality required to become a senator/politician.

The staffer is a specialist. Intimate knowledge of highway infrastructure, or technology issues alone do not make a senator. A senator is someone who manages a small cabal of specialist experts who collectively hold more knowledge than any one head could ever fit. The job can’t be done by one person, its a coordinated team effort.

would people actually vote for them?

Combining brains and charisma is very rare in a politician. Obama being the textbook example - senior law school lecturer but also has the emotional depth and natural charm.

I’d think the list is shorter of charismatic politicians who hadn’t gone to a top notch law school. Plenty use brains and charm. Even spineless hacks like Mitch McConnel know how to be charismatic around wealthy republicans, and are brilliant, just brilliantly evil.

And he was accused of being a “secret Muslim” trying to bring Sharia law by a large percentage of the population.

That and trying to use the military to “takeover” Texas.

Ironically enough, I think that's the problem that lobbyists were intended to mitigate. You can't expect legislators to be experts on every technical topic, but in an ideal world you could retain subject matter experts to give good information to any congressperson who needed it.

It's not a bad idea in theory, but you run into the classic problem of deciding how to define "good information".

I don't think lobbying is bad, the problem isn't lobbying, it is the money required to make senators listen to them. That is the element that is corrupting everything.

Also senators can, and often do is have experts and have them provide sworn statements about given subject.

Congresspeople do have access to experts. Their staffers are brilliant. The issue is that ultimately what motivates any politician is moneyed interests, and not collective good.

So the reasoning here is if there are dumb legislators we should just defer to Google on how they should be regulated?

No, we should defer to the career antitrust experts at the DOJ and FTC as well the academics who study anti-trust and competition (a mix of mostly legal and economic professors).

You mean the “career” experts that are just as beholden to the whims of the President?

Or in the case of the FTC that gave the cable companies more power and upheld states blocking municipal broadband and got rid of net neutrality?

Not defer to Google, but a smarter legislator.

I don't understand the issue completely, so the case may have merit but it's not like the current admin doesn't have an axe to grind with them. Might be just electioneering antics but look at this tweet: https://twitter.com/EricTrump/status/1303288007357796353

I'm pretty sure all of your politicians have axes to grind, for different reasons.

Google does make a smartphone, and if my parents are any example, every smartphone is an iPhone just like every gaming system is an xbox.

I don’t know, some people here seem pretty confident in the knowledge of senators etc. when it comes to calls for investigations into Apple.

EU courts have arrived at similar conclusions re: Adwords. Specifically, they found that adwords does this sort of thing to its "search partners," competitors to google monetising via Adwords.

Google fought it, lost it, and then took its puny (in google terms) $1.5bn fine as a (affordable) cost of doing business. They certainly would have be happy to pay more than this for the gains they made by operating this way.

The issue with Antitrust as it is, is that the "crime" requires active proof and specificity. Essentially, the (practically) inevitable side effects of market dominance. You might need to prove specific violations like price fixing or (like with Adwords, Amazon and other current cases) using market data to gain an edge as a participant. You might need to prove market dominance by "proving" price effects and other microeconomic effects.

The problem is, these are hard to prove. You certainly can't prove all of them. But, they are a problem to the extent that monopolies exist. An occasional, multi year case ending in a fine <5% of annual revenue... this is a farce. Antitrust law is having almost no effects.

It doesn't help that software economics are very different, and that a lot of both theory and legislation practically assumes that the "market" is commodity-like. It just so happens that dominating literal "marketplaces" like adwords, amazon marketplace, etc. is an area with precedents.

Meanwhile, it's not like we particularly care if the adwords marketplace is efficient. There is simultaneously a massive public interest in 2020 monopolies, ooh. Otoh, the cases that prosecutors are making are very disconnected from the actual public interest. The public isn't all that interested in ad marketplaces being "better," cheaper or whatnot.

You can't count on senators to know what they are talking about. This article from yesterday I found quite helpful in explaining the details of the situation.


Thanks, I learned much about the workings of ad space from reading it.

I like the definition of anticompetitive behaviour that revolves around using power in one market to squash competitors in another.

Example: back in the 90s when Microsoft found itself at the pointy end of an antitrust effort, they did things like:

- Restricting OEMs from installing non-Windows OSs

- When they didn't restrict it, they effectively did by having more favourable pricing per license if Windows was the only OS you shipped

- They did similar things with Office but it was less egregious

- Restrictions on preinstalling Netscape

- Restrictions on default browser

- Using Windows to hinder competitors by breaking existing applications, giving a head start to MS software or moving core functionality into the OS (as they tried with IE)

Here are some theoretical examples of what anticompetitive behaviour might look like:

- Making AdWords usage exclusive (ie you couldn't also use Bing's version)

- Making Doubleclick exclusive

- Making either effectively exclusive by providing preferential pricing for exclusive use

- Making use of one ad platform require use of another

Antitrust doesn't exist to protect competitors from competition. People forget that. There are (many) competing display ad platforms. There's less competition in search (and thus search ads). But you can create your own search engine as Bing and DDG are witness to (plus all the dead ones).

It's just that all the other search engines suck. Or, perhaps more accurately, Google is seriously good at search in a way that's hard to replicate.

I personally don't think we should go down this road of dismantling companies just because their competitors suck. That's the Yelp business model. Yelp hasn't changed in 10 years. They have so many missed opportunities. But no, "Google is stealing our content".

Microsoft has the financial muscle to fund a search engine. They do of course. But Google is much better. Just because Microsoft sucks it doesn't (necessarily) mean that Google is a problem.

> Antitrust doesn't exist to protect competitors from competition

That's in America. In Europe, it actually does.

The headline doesn't quite convey what the argument seems to be: Google is involved in nearly every step in the chain between advertisers seeking to place their ads and the publishers selling space on their websites. Hawley pointed to findings from the United Kingdom's antitrust regulator showing that Google has dominant positions in various parts of the ad technology market, ranging from 40% to more than 90%.

There could be some merit in that argument.

But the way it is usually made ("Google (or Google+FB) monopolizes the digital ad market") seems easy to refute: just look at the demand for TikTok, and Snapchat is pulling in over $500M (revenue) a quarter and back to robust growth after some missteps a year or two ago.

And Twitch ads & sponsorships are huge.

Even the UK investigation found Google+FB only controlled 70% of the market (which I believe is less than the major newspaper publishers controlled in the newspaper heyday)

My impressession is that politicians prefer Facebook over Google.

Google don't seem to be as interested in playing the game as facebook.

They are all playing the same game, but the niches they found success force them to play differently. Google tried to take over fb.com with g+. FB tried to attack android with fb phones. They both are fighting over direct direct messaging with Hangouts/messenger/whatsapp. There is just as much politics in Youtube as there is in Facebook.

No one seems to agree on what the problem is or what the solution is. Any "solutions" are complex and it's hard to predict the outcomes of implementation. The motives of politicians implimenting these solutions have never been less reliable or more corrupt or short-term-ist as far as I know. And in the background we have problems 100 times larger than whatever Google is meant to be: the collapse of democracy, climate change, Chinas rise in power and fall in decency etc.

If Congress cuts Google into pieces, they should be consistent and set some sort of maximum market value (as a percentage of US GDP).

That way, once companies reach a certain size, they can expect a split.

I actually think this would be good for society, because it would shuffle things up consistently to force new ideas. Also, smaller companies can grow more effectively, so the stock market should enjoy that as well.

And then how do US companies then compete with overseas companies that don’t have this setback?

Other than the US regulating the shit out of overseas companies like China does, they can't.

make targeted advertising opt in. Make the opt in scary like how android enables the installation of apk's from unknown sources.

Just ban targeted advertising, but allow websites that ask the user to tell about themselves in order to get tailored product recommendations.

I recently listened to a podcast with Jason Furman and he made an interesting distinction. He distinguished between companies that grow to be very large through organic growth vs companies that grow very large through acquisitions. He used Walmart as an example of the former. They grew by being really good at logistics and having a really efficient supply chain, so they could offer lower prices and ultimately open a lot of stores. He used Google as an example of the latter. Obviously the search tech was their own but basically all of the ad tech (what they actually generate revenue with) was cobbled together through acquisitions.

Or educate users to pay for quality and make quality websites/products so we don't have to depend on advertising. This would be insanely difficult to pull off, but it is at least worth a try.

The incumbents have trained users to expect everything on the internet for free. People think they pay $75 (or whatever their internet connection costs) and that should cover everything - news, search, email...

How many people even pay for super important stuff like email? No wonder ads are everywhere, like bad smell.

That sounds like you are asking the customer to serve the market instead of the market to serve the customer.

$75 should cover the basics, including a sub to Facebook/Google if you want it. It's far too much to pay for only the internet pipes.

Advertising is a huge reason why the US economy keeps on churning. I think there could be several knock on effects to something like this, even though in principle I agree.

Despite what Government thinks of Google, I think holding antitrust hearings is just a political theatre. At the end of the day they might slap Google with a penalty which by all accounts will be just a margin of error tax write off. If the government really wants to encourage competition, they need to enable creation of new businesses, plain and simple.

Google isn't a Monopoly as much as they just have the most eyeballs because they built a great company. But Facebook, Pinterest, Snapchat, Reddit, and a few dozen smaller players, are all the answer to Google's monopoly.

I've just spent a few days testing different ad segments, and for the one I was doing, it turns out Reddit was 5x cheaper than the next cheapest source. Cool right?

What we need is more companies that are able to capture eyeballs, and an efficient mechanism to distribute ads among them.

> Reddit was 5x cheaper than the next cheapest source

I applaud anyone that can monetize the reddit audience. I've dabbled in their ad platform and had no luck at all.

The internet the way is currently design will resulted in centralisation and end up with monopoly or oligopoly.

>The internet the way is currently design

It's not the design but the users and to a extend Europe itself. The US is one problem, but that Europe is not even capable to create it's own "cloud" is a sign...just look at Germany they really think 4600 euros is a good payment for top-IT personnel...that says everything. Having Internet-speeds like in the year 2000 too. It's impressive how slow, complicated and lame Europe is (in this sector).

Depends on where in Europe you live. The more east you go, the faster it gets. Fiber and 1 Gbps is standard and cheap here. I pay €15/month for that (TV included). 4G is everywhere. I pay €8/month for unlimited data usage and phone calls.

It's not about in which part the net is fast, it's about that the biggest economy in Europe is sleeping.

Any system in which network effects and winner-takes-all exist result in centralization and monopoly. I don't think you could design the Internet differently to avoid this.

Ban indirect monetization of private data if it is not an opt-in feature separated from the rest.


I am completely confounded that I have not heard of any real legislative proposals to establish a clear separation of duties in regards storing data and selling it.

There is an obvious conflict of interest if you both have my data and are granted free reign to do what you want with it (via click through agreements that modern society is now dependent upon) you have unfair leverage over me.

If you are able to do this en mass without much competition... I'm thankful things haven't gotten worse than they are.

Something similar to Glass-Steagall is desperately needed in the digital era. Hopefully somebody is working on it...


The winner-takes-all positive feedback loop exists because social networks are held in silos built on IP regulations.

So I guess that's another reason the existing concept of intellectual property needs to be torn down and rebuilt. Would winner-take-all effect be so strong if all user-provided data was owned by users, and data processors (like social network) were forced to implement protocols allowing for easy and unconditional sharing of all data (including metadata forming a relationship graph) with any other service the user desires? That is, force services like Facebook or Twitter to be just UI frontends to what would become an implicitly federated social graph. Then, social media companies would hold no power over people.

Right now this can't happen, because the companies own the data on their platform, so competitors are not free to create alternative services for perusing that data.

If they depend upon another entity for doing the task they aren't competitors in the first place. That is like calling me a competitor to Exxon Mobile because I can get gas in my car at a gas station.

The data siloing concept doesn't make any sense on a networking and design level in the first place. It would call for the user to implement all data desired by the service in their form and pass it back and forth - which in practice would involve a lot of redundant dataschelping, especialally if everything was designed to be stateless by making websites not store any data permanently! Indeed the only point to the design is as a passive aggressive way to tell them to "fuck off, if I wanted you to have it I would give it to you".

On a technical level it is a bad reinvention/implementation of cookies and using them to store all state essentially.

> If they depend upon another entity for doing the task they aren't competitors in the first place. That is like calling me a competitor to Exxon Mobile because I can get gas in my car at a gas station.

I'm thinking more of "your car only accepts Exxon's gas", or even more accurately, "your car's engine only runs on a very specific type of fuel, and Exxon has a complete monopoly on producing and distributing it".

> especialally if everything was designed to be stateless by making websites not store any data permanently!

That doesn't follow. Websites should be able to store the data from the moment I give them consent, process it within the scope of my consent, and delete it when I revoke my consent. That doesn't imply they need, or should, be stateless.

> Indeed the only point to the design is as a passive aggressive way to tell them to "fuck off, if I wanted you to have it I would give it to you".

It's not about them having it, but about them taking ownership of my data and siloing it in.

And it's not (just) about telling them to fuck off, but about mitigating the strong positive feedback loop that causes centralization of the web - which is the topic in this subthread.

On a more abstract level, I'm thinking of ways in which "data intermediaries" could be made more transparent. For instance, I see no reason why Facebook should own the social network - which is defined fully by the graph of connections between its users. So e.g. if I friend you on Facebook, that edge in the graph is data that belongs to us both. If I share some photo to "friends only", you should be able to see it through an interface of your choosing. What happens instead is that it's Facebook that decides what interface you can or cannot use, and trying to disagree can cause them to terminate your account (and take away access to your data) or even sue you.

There’s a little truth in what you say but a very fatalist part too which is quite untrue.

We could say that roads are a natural monopoly (since they have to intersect everywhere it would be inefficient to have competitors fight and negotiate about every intersection) ...however, the consequences of that natural monopoly can be limited or extreme. For example: do we get to own private cars that drive on those roadways? Or must we all rent a national transportation vehicle at a non-competitive price? etc. where exactly does the “natural monopoly” begin/end and how natural is it?

I caution against letting your thoughts rest on the words “natural monopoly”.

(I think it’s natural for gravity to bring things down to earth. It’s physics. But the NTSB don’t seem to be satisfied with the notion of “natural plane crashes” just being unavoidable.)

One thing off top of the head is

For example: If the DNS was initially designed as such as https://handshake.org could be a better place to lay the foundation of Internet.

How would decentralized DNS and other decentralized protocols help with search, social, and recommendations? Those are fundamentally centralized systems.

What is the network effect for Google?

It already happened, but the fix is simple and remains possible at any time: repeal the CDA.

That would kill almost every internet company. For example if the CDA was repealed then this site would go away almost overnight.

There were internet companies before the CDA came along. I don't consider actively moderated, special interest forums like this one to be dependent on the CDA for survival. Facebook, Google, Twitter, and any company with a business model that precludes active content moderation would be threatened. Those are exactly the companies that everyone complains about.

Before the CDA came around there was real fear that Internet forums getting sued for libel as a publisher. This was before the internet was a mainstream thing. Any comment on HN could be libel and ycombinator would risk being sued over it.

Right, but we really don't know if that fear was justified or not. The CDA came very early on and created a legal shield around these companies that prevented these problems from reaching the court system. We don't know how courts would have dealt with forums like HN, nor do we know how forum moderators would have operated in that legal environment.

There's something slightly disturbing about the idea that a group of businesses who feel threatened by the public can rely on Congressional diktat to protect them from the law.

Speaking in 2020, I think we can say that giving all internet companies blanket immunity to lawsuits from the public was clearly not a good decision. Everything people complain about today on the internet, from censorship, surveillance, addictive design, and centralized control, is being implemented by companies that would never exist without the CDA.

With all the bitching and moaning people do about large internet companies, it's not a fair assumption that the internet we have now is the only internet possible or even the best internet possible. Sure, HN may not exist without the CDA. But we don't know that for sure, and it's shortsighted to limit our thinking to just one website or subculture of society.

Given both Cubby, Inc. v. CompuServe Inc. and Stratton Oakmont, Inc. v. Prodigy Services Co. we know that courts would have held moderated environments as publishers.

Do you think moderators could have adapted to the situation, by pre-filtering content instead of after the fact?

Also, weren't those cases about certain types of content (defamatory or libelous claims)? Don't you think AI technology could be developed to help moderators detect that type of content?

You haven't addressed my most important point: isn't there something wrong with the expectation that businesses feeling threatened by the public they serve can run to Congress for protection?

I don't think pre-filtering vs post filtering effects the decision. Basically the pre-CDA law was "If you apply any form of content based moderation then you are a publisher and therefor you can be sued for the libel people have posted". I was active in the anti-spam community at this point in time and one of the large concerns at that time was to build spam detection methods that where not content based in order to avoid the danger of losing the common-carrier status.

You said "Also, weren't those cases about certain types of content (defamatory or libelous claims)? Don't you think AI technology could be developed to help moderators detect that type of content?"

I do not think that you could build AI technology that would be able to do that. In US law (IANAL this is not legal advice) one of the things that separate libel from non libel is the truth of the statement. We are no where near the sophistication of AI to be able to determine the truth of statements. For example how would an AI be able to permit "Jeffrey Epstein is a child molester" but block the same statement made about another person?

Then you said "isn't there something wrong with the expectation that businesses feeling threatened by the public they serve can run to Congress for protection?"

I am of the view that one of the roles of a government is to provide a way when the rights and privileges of different groups come into conflict to resolve those conflicts. In a democratic republic the political process is entirely the correct place for businesses and other stakeholders to establish that balance.

I think that it is the correct balance that the actual author of the content should be the one who is held responsible for the content that they produce.

I find it ridiculously disingenuous of the DOJ to pursue Google for "monopoly" of the advertising market, while allowing Sinclair Broadcasting to do literally the exact same thing to local TV stations and the related local advertising markets. While this may not be a political witch hunt - it sure quacks like a duck.


The DOJ should pursue both companies if they think they are illegal monopolies. They don't need to do things serially. There may be a non-obvious reason that Sinclair isn't (visibly) being pursued.

Yep - these same govt officials falling over themselves to actually remove long standing laws and protections in this area. It's so political its horrible.

If Google buys them some more, then the senators might buy it.

Google doesn't quite have a monopoly, but Google, Facebook, and Amazon together have 70% of online ad spend.[1]

[1] https://marketingland.com/almost-70-of-digital-ad-spending-g...

Of course. There are some other players. Facebook, being one of the biggest. Twitter, Snap, Tiktok, to name a few.

Controversial opinion here but I support monopolies. Compare the innovation that came out of Bell Labs when Bell had a monopoly vs. the stagnation that exists with the current telecom oligpoly. Most tech companies are natural monopolies so they're never going to exist on a purely competitive market, so it's a choice between oligpoly and monopoly, in which case the latter is far better for funding research.

I think this is a good point that doesn't get discussed enough.

Was Bell bad for the consumer? Yes, probably. Was the benefit of the increase in innovation a net benefit for society when compared to deadweight loss incurred by monopoly pricing? I don't know.

There is almost no way to answer this question in a rigorous manner, and yet it's extremely important for policy. The only thing I can think of is experiment over a long period of time between distinct geographies and see what happens.

If basic R&D is underfunded by the public, allowing monopolies to overcharge consumers and pass some of that money on to their proprietary research departments isn't a great solution.

We should guard against monopolies and increase public funding of researchers.

In what way was Bell Labs bad for the consumer? I don’t know the history.

Not Bell Labs specifically, but its parent company AT&T had a vertically integrated monopoly which gave them control over essentially all communications technology, and also virtually all telephone service, which it used to dictate (high) prices and constrain the growth of their competitors.

Now, the monopoly profits AT&T may have been worth it, because without them, AT&T would not have funded Bell Labs creating among other things C, Unix, the transistor, the laser, and photovoltaic solar cells.

I will just note that if you want to measure it was worth it, consider what the GNU in GNU Linux is commonly held to stand for.

Perhaps, in some cases, monopolies should be publicly-owned, rather than private enterprise.


State-owned monopolies exist in most capitalist countries, from operating prisons, to managing railroad or telecom infrastructure, to nuclear energy. The US may have fewer of those than say, the UK or France or Singapore, but I would not call those countries _communist_.

Not really. Different communist states have had different goals at different times. For example, nobody here is proposing collectivized agriculture (collectivized compute programming?), or purges of people perceived to be members of the bourgeois class.

God I wish at least 5% of people I see mentioning communism or socialism could properly define either of those terms.

It's a loose comparison, but a monopoly is like monarchy, competition is like democracy.

If a country has a king that is truly capable, caring and trusted, he can get a lot of things done that no other system can compete with. If he is Robert Mugabe, tough luck.

If the country has a democracy, there is limit to the amount of damage any one leader can do. And there is a hope that over time, the system will promote the "better" leader.

Similarly, when there is on a monopoly, and the leaders really care about developing the space, lots of good can come of it. Such as AT&T for a time, and IMO Google back in Sergey Brin's days as CEO.

But when they have gone south, it brings down the whole sector. Whereas with competition, there is a limit to the stagnation and anti-consumer behavior, and a hope that the favored companies will generally prevail.

Over the long run, I am pretty sure that democracies / competition is the better bet for progress.

<OT> Abrahamic religions pine for a king who will be perfect (a "Messiah"), since that has the benefits without the risk. Meanwhile, the religions compete with each other, which is a arguably better than if one had complete control.</OT>

Google X / Moonshots seem like a viable model, though I can't comment on how well they've done. I think the economic incentives are too different for the likes bell labs to exist anymore. It's hard to justify basic R&D to investors. The days of Bell inventing transistors only to have intel benefit is over, research need to be justified and monetized now. Basic research at Bell / Xerox / IBM labs existed in a time where there was some sense of charity.

> Most tech companies are natural monopolies...

What? Why? ISPs like Bell are "natural monopolies" because last-mile right of way is expensive and digging up the same street 10 times to lay 10 different lines is hugely redundant. There's nothing natural about Google's monopoly in advertising, Google just acquired most of their major competitors.

Here's a simple heuristic for determining if a company has a monopoly: if they say that they have a monopoly, then they don't; if they say that they don't, then they do.

Original author unknown.

My app with 0 customers is not a monopoly. Am I now a monopoly?

This isn’t what the expression is referring to

It sounds like Peter Thiel’s idea (though who knows where he got it from)

I don't buy it neither

What did they think was going to happen when Google bought the largest mobile ad network by far for both Android and iOS a decade ago? (AdMob)

Antitrust agencies and governments in general need to set more pro-active ground rules that attempt to prevent monopolies from even happening in the first place.

Punishing them after the fact is both ineffective and too disruptive. I'd rather they didn't allow Google to buy AdMob instead of forcing them to spin it off now. I mean, sure, I'll take it. Still better than nothing. But this shouldn't have been needed because AdMob should have remained a competitor to Google in the mobile ads space, not a division of it.

If large corporations think they have made so much money from their core market that now it's time to spread to new markets, then they must also have the money to start their own competitor in that market from scratch. They shouldn't be allowed to just buy everyone up from that new market.

I think we'd have WAY healthier "free markets" - and more importantly, high competition - if this one thing was implemented.

As much as I agree with the core of what you are saying, it isn't AdMob that is the issue. It's that Google controls both the marketplace, the pub-side and the buy-side, as well as the browser and the devices used to experience ads. Apple is now aiming for a similar level of control in the ad space very blatantly using their ownership of the platform to do what they want to power themselves up.

This vertical integration in the space is what drives the anti-competitive behavior, if the Google buy-side were a separate company that happened to acquire AdMob there would be no real issues as they don't control the whole stack.

FB on the other hand sells a service and lets companies use their APIs to integrate as well as competing with them via their internal sales team.

These companies can do this only because you are forced to work with them because they own such a large part of internet traffic, not because they offer particularly good services or have better technology or prices.

Traditionally vertical integration is fine and horizontal suspicious in antitrust law. Why are ads different?

I’m not sure how you can say that. The first examples of anti trust action have been about vertical integration and ownership of critical infrastructure by a market player not allowing entrance by others because it’s not easy to lay down more railway or more phone cables or telegraph cables. That’s the case for the web advertising situation.

My guess would be: because a single company owns a significant chunk of the market share of each segment of the vertical they're integrating.

Also because our pyramid shaped tax structure favors vertical integration.

Can you explain/expand your reasoning behind that a little more? This is the first I've encountered such a point.

Tax pyramids occur when the total value is taxed at each point of sale instead of the value added. This results in taxes being taxed which effectively increases the tax paid on earlier stages. For example, very few would find it reasonable for a 100% excise tax to be applied to alcohol or automobiles but less would object to a 30% tax and legislating a certain structure for the sale of alcohol, for example farm -> brewery -> distributor -> retail or manufacturer -> dealership, that effective results 185% or 69% whilst only having the appearance of charging 30%.

The easiest and most effective way to reduce this tax burden is simply expanding your vertical. This is why tesla is so staunchly against third party dealerships or why hyperscalers and apple, google, microsoft, and amazon are all starting to make their own chips. Breaking up big tech won't accomplish anything in the long term without fixing the broken tax system and the incentives it creates.

Progressive corporate taxation?

I'm curious about how "pyramid shaped tax structure favors vertical integration". What is a pyramid shaped tax structure? And what about it favors vertical integration? Casual searches on it don't yield much great info atm.

To my knowledge, the US government has wide ranging powers to prevent and break up monopolies. Thanks to standard oil and at&t.

The acquisition needed approval: https://www.ftc.gov/enforcement/cases-proceedings/closing-le...

It's just that the current political climate (including democrats) seem to favour non-intervention.

Not just current, more like "in general since the 70s". This was mostly due to the legal doctrines of Robert Bork, who was Solicitor-General during Nixon and held great influence from the late 70s, which stated that the only good reason to engage in antitrust was to increase consumer welfare (and implicitly, if antitrust would reduce consumer outcomes, antitrust is wrong). The idea basically boils down to "the customer is always right."

Now of course, many arguments being made today are that where the negative effects of a monopoly are felt aren't primarily with consumers, but with smaller enterprises being squeezed out, and with workers at a monopoly deprived of choice, both commonly cited with Amazon. Bork's stance was that these aren't legitimate reasons to engage in antitrust and really end up protecting inefficient existing enterprises.

Anyhow, Bork's Antitrust Paradox (https://www.amazon.com/dp/0029044561/ref=cm_sw_r_cp_apa_fab_...) was and is one of the most cited books in antitrust law, mostly to make the case as to why antitrust in a particular situation is not merited, and the Supreme and lower courts have adopted views from there directly dozens of times.

So it's not just a recent thing, it's a long running, over 40 years now thing

> The Competition and Markets Authority has completed its market study into online platforms and the digital advertising market in the UK.


See the analysis on DoubleClick etc. there.


I work at google. My grandfather spent a couple years in a Gulag. We both got free food at work, but I’m allowed to leave at the end of the day, and my job is writing code instead of counting dead people.

Counting dead people should be quite easy with HyperLogLog

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact