Posted by donohoe 14 hours ago
My belief is: legislation needs to either make it just as hard to merge two companies as it is to unmerge them, or make it just as easy to unmerge two companies as it is to merge them.
It's insane to me that for how often companies merge and cause competition issues, we effectively never see the opposite happen. I know there's a ceremonial approval for merging two companies (at least in the US), but it's just impossible to undo or prevent the damage.
Aswath Damodaran, who teaches corporate finance at NYU Stern, has a bunch of great talks and content about this where he discusses how companies should act their age. When older companies that are no longer paying dividends are moving into a divestiture phase, they restructure and split up. What we are seeing with some of these big tech companies is that they are transitioning from an exciting growth story into an extractive dividend story in a way that's becoming harmful to consumers. That's why we're talking about breaking them up, and thats the line their management will have to walk if they want to maximize the value of the firm to shareholders through the decline.
There's no shortage of obvious problems that can arise from concentration of influence and control over a market. So just try to prevent it. Trade some max efficiency for redundancy + anti-corruption/pro-competition/pro-consumer market forces.
Alphabet is rolling out autonomous driving vehicles in many cities around the USA, potentially getting rid of the need to drive a car in 90% of today's use cases.
How is this not still an exciting growth story? My kids might never need to drive a vehicle, even though we live in a car dependent suburb.
[0] https://batesonlaw.com/divisional-merger-tactic-how-liabilit...
[1] https://www.creditandcollectionnews.com/u-s-supreme-court-de...
[1] https://www.investors.dupont.com/news-and-media/press-releas...
I intended it to point out that it’s too easy to merge companies together but it’s effectively impossible to stop someone from becoming anticompetitive without a real consequence.
If we can’t make it easier to rip them apart then we shouldn’t be so slap-happy about approving them in the first place.
> legislation needs to either make it just as hard to merge two companies as it is to unmerge them
"The laws says X" doesn't change whether X is possible or not.
In a merger you can take as long as you want to go from
1. Two separate companies except at the end of the quarter we add their revenue and expenses together in a spreadsheet to transition to
2. One fully integrated organization
And usually you are becoming more efficient and saving time and money as you integrate.
Splitting a company needs to happen quickly or you'd get all sorts of weird effects where coworkers are ostensibly competitors whilst sharing resources during the transition. And you have to expend a huge amount of effort. Just a couple random complex systems that need to be untangled off the top of my head: physical property and leases, IP space for every IT service you run, multi-year contracts with every vendor from janitorial to SaaS, multi year contracts with customers depending on how the split goes, and of course all the intermingled finances and HR and spreadsheets every company in the world lives on. I'm sure there's thousands more considerations.
I agree antitrust is a big problem that needs to be solved. But "it should be the same amount of effort to merge and split a company" is just fantasy.
What makes splitting a company out difficult isn't (directly) a financial or paperwork burden - it's that tightly integrated systems are very difficult to untangle. There is nothing analagous that could be introduced in the merger process. You could add a mandatory delay, but that's not making it "as hard", it's just making it slower.
what is the purpose of regulation except to resist entropy in such strategic places?
Or we could just roll antitrust policy back to what it was before Ronald Reagan and Robert Bork installed the Consumer Welfare Standard, the idea that companies must be allowed to merge if they can scribble a tall tale with crayons on butcher paper about how the merger will benefit consumers, for sure, pinky promise. This is obviously mega-rigged, it comes from the Robber Baron era, it was defeated before (look up Louis Brandeis) and it can be defeated again (look up Lina Khan). They didn't even change the talking points (dontcha know, the Standard Oil monopoly reduced the price of Kerosene by 70%?!) -- time is a flat circle when it comes to anti-trust policy. Let's spin it back to the part of the circle where we win.
I think that despite this reform being a blunt instrument it would work surprisingly well.
It would allow companies that should have declined to decline and it would give massive incumbents a major incentive to innovate in-house.
Phillip Morris / Altria / Kraft / Mondelez
HP / HPE / Agilent / Keysight
[Disclaimer: I work at Google, all words my own and not representative of anyone, etc.]
There's no such thing. Let's stick to software since it's what HN knows best, but it's a universal truth.
Merging two companies: you take the two sides and they keep running. You probably need some extra accounting work to make sure the mandatory reports from each side get combined, but that's all you have to do.
Splitting two companies: oof. If you split ChildCo out of ParentCo:
* You need to hire all the roles that weren't part of that organization before (HR, legal, compliance, etc.)
* You need to register this new entity in all the countries it operates and/or has employees in
* Technologically it's somewhere between messy and impossible. Now you no longer have Google's build stack or monorepo. You have to rewrite everything. You can't just move some VMs to a cloud provider of choice
Off the top of my head, something along the lines of having a mandatory period of time where the company being acquired must be kept separate enough that the merger can be dissolved in a timely manner should it be deemed necessary. How long that window should be would be up for debate but personally I'd argue the window should scale somehow based on things like the valuations of each company involved, headcounts, and maybe even competitive landscape.
An example would be requiring both companies to perform a pre-merger assessment to determine positions/roles, technology use and regulatory requirements that must be maintained for a clean separation and creating a plan that gets filed with the relevant agencies to be used if/when the merger needs dissolving. If the time comes that they need to dissolve the merger but they don't maintain that separation, they still have to do it and you slap a fine on them equal to some percentage of that mergers cost to be paid by the parent company.
Hell, that idea alone would accomplish both making it harder to merge and easier to dissolve the merger.
This is similar to people who think we should just legislate secure encryption that has law enforcement backdoors. It's not possible, and people demanding it without an understanding of the area they're trying to regulate is lazy and insulting.
Read it more like a poem and less like a proposed bill.
You could give a copy to both successors. Probably would want to excise some of the repo on both sides, but build tools for sure would be fine to have a second copy. There's no need to rewrite everything, especially if Parent and Child are barred by court decree from entering overlapping businesses.
For things that really need a single corporate owner, you could potentially spin that off as well and have both successors contract from that one or depending on the terms of separation have a joint partnership.
Hosting would be messy, yes. But somehow the tech world outside google manages to run systems without Google tools. It would be an adjustment, potentially a large adjustment and take a lot of hours.
> You need to hire all the roles that weren't part of that organization before (HR, legal, compliance, etc.)
> You need to register this new entity in all the countries it operates and/or has employees in
This costs money, people, and time, but it's straight forward. HR and legal would likely need some additional people, but likely you assign the people who work with the spinoff business to the spin off corporation and then back fill as needed.
Spinning off companies happens all the time.
I then worked at FireEye during it's split.
Fun times.
Chrome is one of those nebulous areas. If they're dictating web standards they deserve at least an antitrust glance
It's hard isn't really a blocker, it's just saying 'this isn't convenient for us', which is why it would have to be forced by legislation.
No CEO wanted to merge when it would turn two 5mil a year jobs into one 5mil a year job. Cutting taxes on the rich enabled profiteering by CEOs.
Can someone closer to GOOG explain this? The phrase "ad tech" seems to have a very specific meaning here. Does this 1% include all advertising around the Web? Basically all ad revenue outside of Google's own properties? The number is surprisingly low.
The lawsuit is specifically about googles ownership of the auction and the ownership of the relationship on either side of the auction. The AdX auction also contains demand (ads from places besides google ads) and inventory (ad slots outside of ad manager). Website publishers allege that Google uses this ownership to artificially deflate the value of ads on third party websites.
https://www.sec.gov/Archives/edgar/data/1652044/000165204426...
On page numbered 60, you can see the Disaggregated Revenues.
The $30B quote seems to be referencing "Google Network" revenues declining from $31B in 2023 to $30B in 2025. "Google Network" is grouped with "YouTube ads" and "Google Search & other" to comprise the "Google advertising" category of revenues which increased from $238B in 2023 to $295B in 2025.
So yea, "ad tech" does not represent the whole of Google's advertising revenue.
Like, it's a pretty basic part of American law that courts only hear "cases and controversies" - that is to say, you need to have two parties who disagree before judges are allowed to do anything. If both parties no longer disagree, a settlement is how they call the lawsuit off. But if they're not allowed to settle... what happens? Are we going to force both parties into the court to argue positions they no longer hold? Like, what stops the government from just arguing the case really badly to intentionally lose? That would be worse: now, instead of a prosecution leading to a weak settlement, you have every weak case being carried to term, resulting in a miscarriage of justice as each judgment forms negative precedent against future antitrust action. This is a recipe for eroding antitrust faster.
What you're thinking is that government lawyers are just there to extract the settlement, ergo if we force them to go for the kill, they will get more kills. The reality is that the current social class of lawyers and judges hate antitrust. They successfully recast it from an argument about market power to an argument about price controls, and well, most monopolies don't actually raise prices[0], so building a case against them is very difficult.
Hell, in the Microsoft lawsuit, the reason why the government settled was because the initial judgement to break up Microsoft got thrown out, because the judge who issued it blabbed about it to the press. Keeping the case going would not have produced a better outcome than a negotiated settlement.
Furthermore, while the current "monopoly is about prices" argument is new, the judicial contempt for antitrust is not new. It took three tries and FDR threatening to pack the courts before we got judges consistently applying antitrust law as anything more than a mere ban on unions. While courts are ostensibly neutral arbiters of law, they bend to the fiscal interests of the nation and are inherently political entities, because courts have no inherent power but what they are given by the state.
Or, in other words, if you want more antitrust breakups you need Congress to start appointing more neo-Brandeis leaning judges.
[0] Strictly speaking, there are lots of businesses where monopolies are economically efficient. A monopolist that raises prices is
It's the government as a prosecutor. If current teams are compromised, they find a different team.
This isn't like some law firm with an interest to keep their own lights on. A government prosecution against a threat to labor and economic markets should be to close whatever loopholes lead up to this point.
>The reality is that the current social class of lawyers and judges hate antitrust. They successfully recast it from an argument about market power to an argument about price controls, and well, most monopolies don't actually raise prices[0], so building a case against them is very difficult.
Sounds like a compromised DoJ in that case. Which is definitely the case in the modern day.
But the monopoly argument doesn't hold weight. Some aspects of life are best run as monopolies. But such businesses are highly, highly regulated to prevent the downsides of unbridled capitalism. Big tech is not regulated at all in terms of pricings. And thus we're seeing the results of that in real time.
>While courts are ostensibly neutral arbiters of law, they bend to the fiscal interests of the nation and are inherently political entities, because courts have no inherent power but what they are given by the state.
Sounds like a fancy way to say "big companies always bribed the courts". It's in economic best interest to have a free flow of competition. But it's in individuals' best interests to have good relationships with the biggest winners. The claim of 'interests of the nation' contradicts the fact that being soft on anti-trust betrays such interests.
If they no longer own the adtech then their ad revenue suffers greatly.
Something can be unprofitable but still extremely valuable. These companies are playing strategy games at the geopolitical level and money is not the only resource they want to accumulate to win the game…
I used to work in AdX (and AdMeld that Google acquired) but not since 2014 or so. I believe the AdMeld acquisition itself was one of the things under the microscope here.
Its still a massive business that they make way too high of a margin on through some truly brazen monopoly tactics[1], so the takeaway there is fundamentally twofold:
1. Google's money printer is so much bigger than one can really even fathom. When you're searching, you have intentionality already; for this reason search ads are less like a new type of billboard, and more like a technology that replaces the racks at Macy's every time you walk in, depending on who bid the most for you in particular based on what you're there to buy.
One consequence we don't think about enough is the crazy levels of innovation that Google has gifted the world for free, from Gmail to Google Maps to Android (all problematic in various ways, sure, but still incredibly expensive products to make that we get for free). They don't do that because they're some weird company who likes giving away stuff -- they do that because that's rational behavior when you own a money printer.
2. Display ads are becoming more and more desparate as our online culture spurns them more and more. It's perhaps not obvious that this business is why they care about gathering your data -- specifically, because personalized Display Ads on the internet are so relatively inneffective at this point that they need the boost.
I'm honestly guessing that Sundar is sad about this decision, TBH; splitting off the whole "buy side" of their display ads business would win them pity for the next few decades from regulators, and it's just so clearly doomed, IMHO. Ads increase, blockers increase, paywalls increase, scrapers increase, ads increase, on and on...
e.g., to pull a random old quote:
Meanwhile, the economic engine that powers the long tail of the open internet — Google AdSense — has been quietly shrinking. The “Google Network” (which includes AdSense and AdMob) once represented around 20% of Alphabet’s revenue. At this pace, by the end of 2026, it could be less than 5%, with AdSense (web) potentially representing half of that. In Q3 2024, for example, Google Network revenues declined to about $7.7 billion — less than 8% of Alphabet’s $76.7 billion total revenue — continuing a steady multiyear slide.[2]
TL;DR: On Google's brand new, gorgeous Ads campus right up against the San Francisco Bay, giant signs were put up with "Powering the free and open internet," apparently an old informal motto from the AdSense days. I just can't help but think about that sign, today. I wonder what the people working there feel, now that it's mostly just powering the scammy mobile gaming market.[1]: Super basically, , they've spent insane sums of money to stay as the middle man for the split-second auctions that determine what to put on the Macy's shelves. This makes them some money directly, but it makes them way more money by unfairly propping up their own advertising companies, which are competetive customers of Google's own marketplace.[3] The details get boring, but I think it's obvious why running a fundamentally-opaque auction in which you are a participant creates perverse incentives.
Technically they run the automated marketplace that lives below the marketplaces that the advertisers and publishers work with directly, if that makes more sense.
[2]: https://www.thecurrent.com/opinion/opinion-andrew-eifler-vp-...
[3]: Some goats rants from previous stages of my DABDA process, if anyone is curious:
- From day 1 of this trial: https://news.ycombinator.com/item?id=41496923, and w/ some links: https://news.ycombinator.com/item?id=41496968, on a particularly damning memo: https://news.ycombinator.com/item?id=41497424, and my honest best attempt to characterize the exchange itself: https://news.ycombinator.com/item?id=41501491
For clarity: I'm just some fool, and there is absolutely 0 inside information I can/would share in any of the above.
Of course the best laid taxes always get abused or turn into something else entirely.
Can you explain this more?
“ Robert Jordan, wiping out the stew bowl with bread, explained how the income tax and inheritance tax worked. 'But the big estates remain. Also, there are taxes on the land,' he said.
'But surely the big proprietors and the rich will make a revolution against such taxes. Such taxes appear to me to be revolutionary. They will revolt against the government when they see that they are threatened, exactly as the fascists have done here,' Primitivo said.”
now there's the euphemism of the day. Alphabet derives 75% of their revenue from Ads. It is their business.
Read more about SCOTUS justice Robert Bork who ushered in the modern anti-trust viewpoint.
E.g. https://en.wikipedia.org/wiki/The_Antitrust_Paradox
> Bork argues that the original intent of antitrust laws as well as economic efficiency makes consumer welfare and the protection of competition, rather than competitors, the only goals of antitrust law.
Not very different from how federal protectionism keeps Apple's App Store centralized despite trillions of dollars in ostensible market damage. Anyone trying to loosen America's grip on software distribution is crossing an ugly national security line that we're all too ashamed to admit exists.
If you want to have some fun, go on LinkedIn and see how many former military people, proud of their service, boast how they are at x or y tech company in director esque roles.
Then bookmark them and see the pages disappear in the future. Boy I’m glad it’s legally 100% ok to scrap LinkedIn data because it truly is a revolving door.
We're all ashamed, apparently?? What do they call this, projections of a guilty conscience?
The whole system is now easily abused by huge mega-corporations. That goes against capitalism. Why did the USA suddenly abandon capitalism?
251 individuals in such detected instances just in google. And that report is 10 years old, I'd bet it's 10x that by now.
Google employees tens of thousands of people. The US intelligence apparatus employees hundreds of thousands of people. In all of that, over eight years, there are 251 people who moved between the two, and because of that absolute rounding error, the entire federal government is beholden to them?
Since you are making the accusation that this is a real conspiracy: statistically, how many people should have rotated between these two giant employers over an 8 year span? The answer isn't zero.
Also, this makes the "data" even more pointless:
"between Google or related firms, and the federal government, national political campaigns and Congress during President Obama’s time in office"
"were appointed to federal advisory boards."
Do you know how many of those boards exist? There are tons of them. I've known a couple of people who have served on them. They're incredibly boring and deal with things like "What accessibility standards should federal government websites support?". Of course you would want people from Google on that type of board.
> 251 individuals in such detected instances just in google. And that report is 10 years old, I'd bet it's 10x that by now.
"Detected instances"
It's so secret that I can name one of them off the top of my head: Matt Cutts. Is he part of this vast conspiracy too?
If you think it is "nice evidence", how many people should have moved between those two employers?
And you made the claim that the rate likely drastically increased over the past 10 years. Based on what, exactly?
> FAANG and the three-letter-agencies are one revolving door
Most of the NSA is active duty service members and former military. Did you mean other three letter agencies, excluding the most powerful one dealing with surveillance?
> Any judge that attempts to interfere with Google's data collection may well run afoul of the American surveillance system
The judge in this case put the final nail into Trumps slush fund. But now she's doing his bidding. Is that what you're claiming?
Not everything is a conspiracy. You don't have some special insight into this.
> The judge in this case put the final nail into Trumps slush fund. But now she's doing his bidding. Is that what you're claiming?
No, they're clearly claiming it's a non-partisan thing and the sitting president is irrelevant, as are the specifics of the virtue signalling happening. Yes, even though they qualified themselves with "this administration". Do you think that the Clintons aren't friends with the state department surveillance abomination or the nastiest parts of government writ large? That Bill or Hillary wouldn't fight tooth and nail to protect this terrible thing that's been wrought? Replace the context with some virtue signal about NATO intervention in Serbia 27 years ago, nothing changes. Partisan politics is for yokels.
> These days all three branches march lock step in their jackboots
Which is why the Judicial branch has been ruling against the Executive branch again and again? I don't understand what evidence you think you have to the contrary, other than vibes?
That's a very different, partisan effort. It's not bipartisan like the Patriot Act was, or domestic surveillance in-general.
> Not everything is a conspiracy. You don't have some special insight into this.
Not everything is as it seems. You can't meaningfully prove that the NSA sat on their hands in the 13 interceding years since Snowden. There's no evidence that US interventionism is on the straight-and-narrow now, you're wishcasting and calling it truth.
I mean, you're correct? It's impossible to prove a negative.
> There's no evidence that US interventionism is on the straight-and-narrow now, you're wishcasting and calling it truth.
Extraordinary claims require extraordinary evidence and all that. You're the ones making the extraordinary claims. The burden of proof is on you, not the people pointing out the flaws in these claims.
Please. The Androidfication of iOS with malware/scamware would cost quadrillions in market damage.
What about "is incompetent enough to press the buttons when the nice man from Microsoft calls about the viruses on her USB"?
Washington warned us of this threat.
The media makes sure to qualify any candidate threatening the interests of capital as a "dangerous extremist", while "pro-business" candidates receive the lion share of donations for their campaigns. And there are many other guardrails in place to ensure nothing impedes the slow erosion of our infrastructure and public services under the strains of austerity in profit of the few.
The key problem is that most people don't want to pay attention to any of it because it's either depressing or just not within there range of interests. Add to that the fact that most people vote based on their emotions and those are stupid simple to manipulate on the Right: anger at immigrants and "others" works all the time.
Democracy is the least worst option, but FFS, it would be nice if the electorate were actually properly informed when they vote.
Edit: I kind of contradicted myself on the "choice" thing, but this is effectively like herding cats.
My armchair assessment of the the electoral trap we're in:
* first-past-the-post voting
* dark money (money is not speech, it is power)
* campaign finance reform (current model is legalized bribery)
* eliminate the electoral college (popular vote for president)Edit: I don't know that there's an actual database served up by the US Gov, but governments control the naming of these things within their own jurisdictions. All of the big mapping products have the ability to serve up different names based on where you're viewing from.
And Apple just renamed: https://apnews.com/article/apple-lake-ontario-america-google...
GNIS, Geographic Names Information System
(Which still shows Lake Ontario for me as I write this. The body of water along the US's southern coast is Gulf of America however.)
https://edits.nationalmap.gov/apps/gaz-domestic/public/searc...
GNIS renders their own static map tiles, so it may take a while for the update to be reflected on the map itself.
I see this on the top bar
I am curious though - is it convention to leave the data as is, or is it illegal to modify it?
Regardless, itll just be changed back in 2 years; as long as I can type "lake ontario" and get directed to Lake Ontario, I do not care at all what the gov database calls it currently.
Why do you expect meek reaction in a rename like this? What is eye rolling over reacting to this particular rename?
Google should override it but I don’t expect them to, because theyre selfish and suck. Only thing to do is win elections.
No they don't. People can, and should, call things what they want without deference to government overreach.
Apple updated it yesterday. If I were Google I'd turn down that refresh rate and let someone else take the bad PR next time
In this day and age, GPS mapping tools have major implications for societal function, safety, economics, national security, healthcare, etc. I'm struggling to think of an industry where a sudden absence/subtle changes of named GPS mapping data would not have an effect.
If everyone relied on the same data base which could push changes to the map in the pocket of every resident of a nation that would be the first thing I would think to hack. Probably easier than actually jamming or spoofing GPS/GNSS over a huge area.
Don't you understand the horror of what you're suggesting: "I want Google to impose it's own personal value judgments over the government's."
Yes, I think it's incredibly stupid, 3rd grader childlike to call this "Lake America". But the problem is not Google, the problem is the ass hats we (collectively, even if not you/me personally) voted for. The problem is especially with our Congressional reps who have just neutered themselves in service to this stupidity. And I get it, our federal institutions have definitely showed their frailty lately, but I can't fathom then why people think the proper solution is to give more decision-making authority to Google.
I can understand Google somewhat here (if I was running a business I wouldn't be happy wading into a political issue like this) but the idea that they are above criticism for just falling in line still doesn't resonate with me.
Corporations should not be absolved of criticism. But at the same time it makes little sense to put a hamster in hamster wheel and then blame him for running.
People holding power themselves is called self-governance (in the Jeffersonian way).
People voting for representation is called a republic.
People voting away their own power to another is called democracy.
The only people calling it Gulf of America or Lake America are hardcore MAGATs in my social circle. Even former Trump voters in my social circle are annoyed and not using the new names. So it's an effective way to figure out who is or isn't going along with the Administration today.
This is the same logic that the right wing used when they disagreed with Obama on some issues, so they blocked his Supreme Court nominations just because anything Obama does must be bad.
Sounds like your social circle is in denial. There are certainly plenty of things Trump, buffoon and graft-a-holic he is, has done that are corrupt. Renaming bodies of water doesn't look like one of them. (Obviously, because he would have renamed it after a donor, or at least after himself.)
I mean, sure, we can all call anything whatever. I can call New Mexico "Best Mexico" or call Washington, DC "Bad Washington" and that's my right. But pretending that there aren't official names for things because we don't like the current guy seems immature. It's clearly within the authority of the government. Talk like this makes me think that people want to just live in a reality defined exclusively by party politics where we just pretend every law or policy made by The Bad Party doesn't exist, rather than try to win elections so we can shape the one actual reality.
They have the ability to change name services in some database somewhere. But it's not actually an official name change.
So all this hand wringing about official names or whatever ignores the reality of the situation. This is just another temper tantrum by Trump that we reasonable people can ignore.
The easiest way to do that is to just say Gulf of Mexico.
Is it Congress who officially holds that power, then? And are you saying that if Congressional Republicans passed a bill renaming whatever, and Trump signed it, you'd then acknowlege it? Given how vanishingly little they accomplish in Congress already, I'm not sure we'd be better off distracting them from their semiannual debt ceiling/budget hostage situation in order to debate these inconsequential things.
This is a pretty mushy area, philosophically. Is a name what you think something is called (I think our planet is called “Bloop”, but nobody cares) what a lot of people think something is called (“Indian” in reference to indigenous Americans is a famous example of that going poorly, as are many racist epithets), or what an official body says it should be called (“Palestine”, “New Zealand”, and “Taiwan” have entered the chat)?
I don’t think popular usage is any more “truly authoritative” for a name than other sources.
The biggest factor is that tech giants now 'pre-game' major anti-trust relevant actions through an internal compliance team staffed by former regulators. The result is the slam dunk smoking guns of 20 years ago are rare. Today's monopolists artfully push right up to the line between "red-handed" and "arguable". Then take measures to fuzz that line and kick up clouds of plausible deniability in the record. Imagine how hard it would be to convict an embezzler who was carefully coached during the crime by expert former prosecutors and forensic accountants.
Sounds like it's not nothing, but also not much.
That must have been in drafts for a while.
How is wielding this weapon against our most innovative companies going to help the United States compete at scale against China? Scale is sort of required to do so.
So if the government isn't going to be in the business of massive capex investment (the kind China subsidizes) well then you need massive private companies.
The American peoples response to a future dictated by the terms of Chinese hegemony is break up the centers of American innovation? What do you think is keeping the capex cost manageable if it's not the government? How does Google work if you take away its cash cow?
So many of you (I assume despite the complaining that > 99% of HN readers can afford it) need to just pay the $20 or whatever for YouTube premium if it's so terrible and get over yourselves. The tax your children will have to pay if the United States is reduced to a second rate power will be far greater. What's with people wanting free access to services and simultaneously offended by ads? I don't like them either but I understand it would be silly to think they should be free. They can be regulated or reformed in many ways, but the government should be highly restrained about cleaving apart companies. Just remember the powers you give the government today the next government will inherit, and it is not so easy for a party to maintain power for more than eight years. Meanwhile China is operating on government initiatives and sustained strategy on a timeline that spans decades. Companies need to be able to match this horizon without fear of whatever the new normal in Washington is every four years. The trend for the United States is not looking good.
Here is an hour long video where Lina Khan makes this point to the Council on Foreign Relations much more cogently than I can: https://www.youtube.com/live/L_QaZk5iJOA?is=rk192CuSIBHLemsi
By "compete with China" I mean at least have parity of competition with China in the essential technologies of the future, rather than falling behind into some sort of obsolence or inability to shape the global economy we all on the whole benefit from. If China becomes the dominant economic hegemon without a viable American competitive check, I think the world (and certainly Americans) will be worse off.
Meanwhile, a balanced competition between the US and China I think will benefit almost everyone on the planet, for reasons likely not that unlike what I expect Lina Khan will say in that video. There is no perfect solution, so I will look forward to hearing more of her perspective on this issue, thanks.
> Meanwhile China is operating on government initiatives
china fosters intense internal competition though; they don't just create and then subsidize some monopoly... avoiding breaking up mono/oligopolies feels the opposite (and a loosing strategy long-term imo)In the United States we don't have the same reliability of public initiative. That's the point I'm making. Therefore, in such a context breaking up and weakening the companies that do makes us competitive doesn't strike me as a good strategy. And I wish more people would acknowledge that tradeoff in their anti "big tech" rhetoric.
Meta and Google gobble up an absolutely gobsmacking duopolist's share of western world's economy each year.
Meanwhile, two other American companies without monopolies are ahead of them in the frontier model race.
Why do you think that is?
If you think Google (or any large firm) is going to be by necessity be a loser in the AI race relative to smaller firms, you should go ahead and explain why you think that is.
My point is that Meta and Google have had an almost comical resource advantage when compared to all of the rest of the firms in the AI race.
Anthropic raised $65 billion this year, meta is spending double that on capex and alphabet 3x.
That's just their capex.
If you also factor the deadweight loss the duopoly imposes on the entire economy, it paints a grim resource efficiency picture.
Not to mention the negative externalities caused by the decline of journalism. This one is hard to quantify but I believe is the worst.