A Post-Google World?
thebignewsletter.com
thebignewsletter.com
A bit over the top with some of the things google has done in the programmatic space, but aligns with reality. I disagree with the display/programmatic space innovations being held back by google - there are an insane amount of small players who are doing different things and it’s easy to integrate them into the existing space. IE if I want to measure foot traffic, there are like 6 vendors who all do it slightly differently, 1 huge and 5 small companies, some super anti-privacy and some very pro privacy.
Its is called out in the article but not made extremely clear - the vertical integration google has is insane. They own the sell, intermediary, buy, measure, and operations software for a large percentage of the space. Imagine if NYSE owned Fidelity, SP500, and the SEC.
But avoiding the web or mobile apps is becoming nearly impossible. Google tentacles are all over this stuff. No escape. For example, one visits a US government website and it needlessly sends data to "google-analytics.com".
The blog tends to be a bit hyperbolic but the author does do the research. If his facts were wrong and he was notified, I suspect he would publish a correction.
It eventually becomes impossible to avoid doing business with them.
Google owns the infrastructure of search, maintaining them and showing ads has a finite cost. How does piting ad buyers in an auction is different than price gauging?
Imagine apple every year introducing the new iPhone without a price and when you go to the cart it asks for your credit limits before it creates a unique price that magically matches exactly your credit limits.
Isn't it kind of like real estate?
In theory, it should cost roughly the same amount to build a house in the middle of nowhere versus the center of Manhattan. But more people want to live in Manhattan, and there's a limited amount of space, so buyers effectively bid up the price.
On any given website, there is a limited amount of ad space.
Search is a bit different than display which is covered in the article, but imagine 2 jewelry companies that are exactly the same. A user searching for “necklaces” you’d think each company would bid the same amount to advertise there. But say one company is trying to raise a round of funding and needs to drive sales, they’ll bid higher to increase sales but at a lower roas. The other company is satisfied with their revenue and doesn’t bid higher.
In display each bidder has different information about a user (they may have their own data about a users behavior on their website, or it may be provided by a vendor that other bidders aren’t using). So the difference in price is largely a difference in information.
Same thing with real estates, or stocks.
However the iPhone example you mentioned follows a different scheme. iPhone is not something people competing for, as long as stock lasts.
Simple: the 'free market' exists to maximize the price for a given good or service with respect to the demand for such. That's really the whole story here.
> Imagine apple every year introducing the new iPhone without a price and when you go to the cart it asks for your credit limits before it creates a unique price that magically matches exactly your credit limits.
This is how sevreral 'free' markets work already. Anything involving credit or finance, real estate.
It is a grave mistake to assume that the actions of a free market have anything whatsoever to do with what is morally right, or even what is legal.
The market makes line go up and that is the only thing it cares about. Capitalism does not give one good goddamn about who or what it may hurt, it can only ever consider the bottom line.
They can set arbitrary prices and have customers pay anyway, with little fear of competition or market loss.
Giving up on that is stupid, and would set society back. We need more companies and organisations doing it. AI has brought a few new startups into the search space, but yes, splitting the companies up may make more opportunities for them.
That said, if you can raise the capital you can get into search, because of opportunities presented by AI, and google has been sacrificing quality for optimising ad revenue. Email has also seen some startups making big leaps by adopting AI. I wouldn't touch video sharing platforms, or ad networks.
It's almost more useless than Yahoo was after Google came along, which is crazy.
ASEAN and India however rely on google heavily. Googles ad based subsidies for their other products mean that less well of consumers can have access to technology as well. I worry that the freebies will be very hard for many people to ween of.
Google already gave up on this long ago. Their mission now is to print as much money as possible (late stage capitalism).
It's unfortunate, because 2004-2014 was unlike anything we'd seen before, even in the dot-com days. Compared to anything before or since, ElGoog was Incredibly good for the 10's of thousands of rank-and-file employees.
Thanks Ruth Porat! (For serving as the instrument that ended the party.. at the board's request, ofcoz.)
The current issue IMHO is having a single for-profit company doing it at such scale. They're inherently motivated to keep competition away and keep themselves relevant.
Companies shouldn't own the core data (yes, even when they farmed it themselves running cars through millions of km of streets), perhaps we could come up with a fixed price they have to abide to to sell their data to any eligible entity, with governing bodies being a potential reseller of it.
Meteorology wouldn't be what it is today if some single US company owned and organized the world's info. We need a similar model for the rest of it.
This is incredibly wrong. We've had mountains of innovation in all of these areas under Google's leadership and from competitors like Apple, Mapbox, Microsoft, AWS, etc. When Google ends up costing more to use, competitors will _not_ need to innovate to attract customers to their platforms, and the technology will stagnate.
Aside from buying Waze in 2013?
I believe that using wifi to improve localization on phones also happened within the last 19 years—quick Google found me an article from 2010 about how they started gathering data for this in 2007. I definitely remember this having a real improvement on how accurate my phone's location could be, which made using maps much nicer day-to-day.
I also occassionally enjoy the AR features for walking in cities.
[1] Yes this is a thing. A "maps SEO consultant" popped up offering me his services a few months back, and after a bit of digging it's clear this is a new and hideous barnacle of a subsector has appeared to make money out of making maps and everyone's lives worse
I'd sooner think hell would freeze over than Google would ever 'give up' on its business model of its own accord. I can hardly imagine half of its services being viable as separate businesses. Which might well be a boon for competitors (especially in adtech), but not for any of Google's current employees or stakeholders.
Late game capitalism applied to Google's portfolio would be catastrophic.
Obviously this would be solved if we had a way to centralize our identity and file storage and such across platforms and systems. Yet at the moment the silo is what makes it seems painless and increases its value beyond the fact there’s a substitute to each platform and product.
Tech folks love to say "X is is perfect substitute for Y" in the same way they say "what's so hard about X, I could write that in a week" by considering only simple ideal workflows.
The real value, and the real moat for these products, is that they're tied in with SSO and other products, easy for organizations small to huge to manage, have dozens or hundreds of specialized features crucial to large organizations. They comply with legal regulations, accessibility standards (which ARE legal regulations in many jurisdictions such as if you want to sell to the government), and more.
Writing an email server is perhaps 2% of the work involved in building an email product.
All of our core service and software is locked up in a handful of companies.
For humanity and the future of software’s sake we need to go back to users of software owning their software, preferably by being free and open source.
It hurts me as somebody who wants to make money off software, but SaaS is just a bad deal for everybody. My compromise is in prim solutions.
And Google have provided tremendous value with their free tools. People forget what it was like before Gmail and Docs.
I sincerely hope we can move away from their business model, but looks to me consumers rather pay with anything but their money.
So anyone who’s 18 today has never known a world without it.
People don’t know what the save icon is in reality.
Paying for things is interesting as media streaming is frequently valued by paying for it, collaboration and things like Google Maps not so much.
I remember using a Palm Pilot with Tom Tom, and an old maps pack that didn’t have half the new main roads included in the UK! I’d honestly pay for Google Maps for the value I get from it.
You begin a trip with driving directions, but 2/3 of the way in your phone shuts off because you forgot to plug it in. When you get it going again you have no service. You open Google Maps and... nothing. Your directions are gone, and you can't load them again because you don't have network connectivity. I believe due to the restart this is even the case if Maps happens to give you the "download offline directions" option and you accept (which they don't always do). AFAICT there is no way to "reload" downloaded offline directions, only a search bar which does nothing without network connectivity.
Google only really makes things that serve Google's interests, that's why they'll never fix this. If you, the user, put yourself in a situation where you depend on them, you're the sucker. Google has no incentive to actually make the product good for users, they just have to make it good enough that a sufficiently large number of users look at the advertisements in it. I'd much rather buy software and services from a company that has better aligned incentives. That's why I keep a paperback road atlas in my car.
It's hard to beat Google Maps for POI discovery though
We had Google docs before Google released it and it would have prevailed as an independent company if the authorities had had any interest in preventing trust and oligarchy
I don't think you're correctly remembering Writely or early Google docs. Expectations were very different back when ajax was still a cutting edge buzz word.
That ship has sailed in my opinion. I am under the impression most enterprise and public facing open source projects are fauxpen-source and use FOSS as bait for commercial support and freemium plugins. I am at the point where between two solutions, one fully commercial where I don't even host but can export data out and has strong GDPR compliance and an open source one with a locked-in scheme then I'd rather go with the commercial one because at least I am not under the delusion that I am financing feature-parity open source alternatives to commercial products. I draw the line at formats and protocols, maybe ? /rant
Lots of boutique stores in the city centre moved into the newly created mall (due to the economics of malls being as they are)- the mall was also centrally located so: nothing lost.
Almost immediately after this happened; another mall was built outside of town (requiring a car to get to), but as you might expect: rents were substantially cheaper.
All those stores (via peer pressure, cost saving and so forth) moved into the new mall.
Those boutiques never returned and the high-street is dead.
— we have this same situation in tech,
We have lost the ability to function without loss leading technologies (all US based, tying those needs to whatever economic and political situation is happening in the US to the entire world), and, at some point that will have to end, but the oxygen has already left the room for anything else to exist - everything else is starved.
And, Microsoft will not save us, they are leaning into these practices and will suffer the same if US tech is regulated (as it should be). There’s no space for competition.
I wonder if something like a NIH style grant system would make sense to keep opensource software funded and supported without needing corporate support. I have xz on the mind. Would that have fallen under the control of a bad actor if the original maintainer (or others) had the financial ability to dedicate their time to it's maintenance? Who knows.
Many countries have their own tech stack for what Google provides (think Japan, Korea, China, Russia). It's a shame how EU has completely given up its market for Google.
Reddit, newspapers, Facebook, Twitter, are all locked down or heavily restricted. I am sure there is still a burgeoning small web, but it is increasingly hard to find. Would many consumers notice/care if your search engine only surfaced pages from the top 1000 highest ranked domains? You would actually decrease the amount of SEO garbage you hit from Stackoverflow clones, Amazon listicles, or Grandma's cookie recipe #9381.
A bigger issue are the network effects - no website can afford to block Google; not so much for your startup search engine.
> For humanity and the future of software’s sake we need to go back to users of software owning their software, preferably by being free and open source.
This sounds like a decent plan. Why not start executing the first step of it today?
Or is that the actual first step to make a societal-level change? Maybe it's better to make millions of dollars under the existing regime, and then found or fund a think-tank to get certain politicians elected.
I'm in the process of setting up a Community Interest Company to facilitate this very thing. I spent a couple of decades in the world of Drupal, which is the largest open source community in terms of contributors, and have spent the last few years following and supporting the ceptr.org project which is rebuilding the tech stack aligning as to how nature works.
One of CEPTR's subprojects is Holochain.org, a distributed agent-centric open source language, and I'm using https://theweave.social/moss/ which is built on Holochain, to collaborate with my as of current one collaborator who is my support worker, funded by an Access to Work grant as I discovered and was diagnosed last year aged 50 as autistic and ADHD.
Free/Libre Open Source Software can work and be sustainable, it just takes more people getting involved in every aspect of it, and I find the biggest issue there is the majority simply don't know this stuff exists, let alone they can use it and adapt it to their needs.
So times are changing, we have the power, we just give it away every day by not making the most of what we have control over.
I like the idea of building communities and software that is inspired by nature. Btw homepage looks nice.
The most natural analogy for software is infrastructure. Lots of money to be made in building, maintaining, and supporting it, but as with any technology, the proposition of renting the right to use it is one that is obviously a bad deal for everyone but the rentier, and all the moreso for the fact that they can constantly make changes to the deal, including how the software functions, what one must pay for it, and what other benefits your use of it may extract from you on the behalf of its owner
I make money building software and think this feudal status quo is a terrible one that we should welcome the destruction of. Its benefits are ill-distributed and temporary, its drawbacks are dystopian, pervasive, and long-lasting, not only creating these awful deals for a wide swath of the economy, but also creating massive incentives for the rentiers in this equation to hoard competence in using these new technologies, stymying untold amounts of innovation and progress by making criminals out of tinkerers and discouraging even non-technical users from adapting them to their own needs
Saying that software is infrastructure only supports the renting model.
Your existence is contingent upon people the best of whom (ones who at least read) think 48 laws of power is a how-to manual. You live among people who know less than nothing.
Software is a land of idealistic, capable angels by comparison.
There needs to be a very strong case this time to go after Microsoft to argue for a break up of their business. Whoever is to bring the case (either the DOJ or the FTC), would have to go after Meta and Amazon first before targeting Microsoft.
It won't be easy, but Microsoft has been able to avoid scrutiny for years.
Many regulatory agencies would target Microsoft next, but they have played this game and are willing to do whatever is needed to avoid extreme measures like forced break ups and sell offs.
There’s nothing special about Gmail in 2024. But Gmail is totally free which prevents anyone from investing in a significant competitor.
Society will be just fine.
The irony behind Google is that Google in the early days was the antonym for Yahoo; just a simple looking search engine with a search box, minimalistic UI and a better search ranking compared to Yahoo which was crammed internet portal which looked and felt messy.
Nowadays Google still has minimalistic UI but lousy search and shit ton of web apps that try to capture as much attention as they can while sucking your data and doing God knows what with it.
>There’s nothing special about Gmail in 2024. But Gmail is totally free which prevents anyone from investing in a significant competitor.
Yea, Gmail is free and it somewhat feels OK to use so many people use it but I don't see how Gmail evolved substantially over the let's say last 10 years, there is still room for competition. And there are competitors, idk how "significant" they are but they do exist e.g. Proton Mail, Tuta, Fastmail, GMX Mail etc.
It will read your email for you, reply, make major life decisions (eg respond to marriage proposals, banking transactions, etc), and much more - all in the inbox you know and love!
If there’s one thing I know about Google under Sundar it’s that he’s a man of rare vision, foresight, and leadership. I’m sure things are going to go great!
This is probably my biggest fear about the eventual fate of Google. The amount of data they have is staggering and would be very short-term profitable for PE firms looking to make a quick buck.
If Gmail had a $5/10/20 per month fee, I would probably migrate to a more privacy/security focused provider.
I wouldn't pay much for YouTube, but video hosting is relatively easy and YouTube has degenerated so much that I'd like to watch content elsewhere.
I would pay for Kagi and ChatGPT each twice before I would pay the same amount for Google Search.
Good luck charging for Chrome and any of their developers tools.
I know there's a million services by Google and I would probably demean most of them and I'm not near the typical consumer. However, if the typical consumer at least had to consider whether to pay, there would be so much breathing room in many industries. Something as large as Google trying to explicitly monetize all those services could also tremendously change consumer attitudes towards being willing to pay for things thus making more businesses viable that do less of the shitty stuff that Google does. For a bit, there would be some bad will as people notice extra bills. It sucks that those extra fees would disproportionately affect the poor. The most likely case would be something like Google Prime with a monthly $15 or yearly $140, which I don't think is a gross value proposition for an individual of even low means.
They are a lot more replaceable than most people think.
SSO is also solved. People will survive
The primary reason Google has fought so hard is because it knows, even if left whole, its margins would drastically shrink once it has to compete on price.
Android users are all glancing at each other like John Travolta in Pulp Fiction right now.
Fundamentally I see no problem with multiple app stores, it will be a boon for developers as the stores will have to compete over how much of a cut they take. Or developers can sell direct.
* [1]: https://galaxystore.samsung.com/detail/com.amazon.kindlefs?l...
* [2]: https://www.amazon.com/Amazon-com-Kindle-for-Android/dp/B004...
I guess the only notable exception is the... YouTube app? So what are you talking about exactly?
Google's approach to sideloading is borderline inscrutable, and I say that as someone that will gladly sell them out for their abuse of the advertising market. This lawsuit exists to grant faux parity for the Apple lobbyists due in their government mandated reaming. As long as the AOSP exists, it is literally impossible for Google to abuse Android as an anticompetitive environment for other companies. OEMs control Android as much as Google does. The same cannot be said for Microsoft or Apple.
How is "public interest" defined after Citizens United, https://en.wikipedia.org/wiki/Citizens_United_v._FEC?
some comparisons with past american monopolies were made, but there is a key difference between those and google - the direct impact to people across the world. to me it came across slightly myopic to look only inside one large economy, for things that impact everyone's lives but not directly.
we already have large portions of the world not impacted by google - just look at china. but even there android in its different form has been vital. it is not just a matter of "we got a disruptive new tool to replace the current choice". google is able to bankroll key internet services that could stand by itself even when broken up - android, search (people still use it to open websites they literally spell out), youtube, gmail to name the key few.
change is inevitable but i don't see it happening the way it is envisaged here. pretty much all the "disrupters" are backed the same way that made today's big tech. which goes beyond just google. the now-subsiding ai boom has shown that big tech is still swaying investors for immaterial promises.
Need to raise this issue with the California PUC, about Waymo wanting a Google login.
You wouldn't see e.g chatgpt having login with Google if it was such a threat.
My major concern with kagi is that its good now, whos to say it will ne good in the future. Plenty of SAaS have ebshitiffied. We NEED open infrastructure fpr search. What that lools like Im not sure but it may be possible in the far future.
And if google search collapses then the rest of the empire won’t be far behind
I wouldn't dare to speculate on the cause of this devolution. There are theories like becoming publicly traded for companies, and subsequently execs chasing short term profits to make shareholders happy, or elected officials chasing after votes for the next immediate election cycle. But it would be nice if there was some sort of a large scale in-depth study on the topic and research into solutions for it.
Honestly, the more I think about so many issues, the more I feel like there needs to be a dedicated field into the study of incentives and how they shape the systems and organizations humans build and maintain. Everything from google search being terrible to climate change feels like it all has a shared root cause of misaligned or defective incentives.
Perhaps existing fields like game theory, economics or systems design should cover this topic?
Either way, our systems design methodology is faulty and this fault is nearly universal and systemic, affecting all humans in almost every area of our lives.
One theory I have (full disclosure: subjects I am not too familiar with) is economics and game theory, specifically the "Nash equilibrium" game theory, which as I understand boils down to "everyone should participate in the system, only taking into account their own success and profit, and nothing more than that" might be at fault. Or maybe, it simply codified the innate faulty organizational instinct we have?
That currently falls under economics, sociology and game theory. It’s not really a dedicated field sure but it’s well covered
A Nash equilibrium is a state of affairs in which no single person can improve their position without the collaboration of others. A frequently cited example is a bank run: if the bank goes bust, you better get your savings out quickly, but if everyone does it, the bank goes bust. The solution is for everyone to cooperate and agree to refrain from a run on the bank. Such collaboration is not easy to organize and not easy to convince a majority to participate in; but the ego-centric philosophy of capitalism makes it infinitely harder.
Even among big tech, Google is an unusually meaty target because:
1. Literally all but two of their consumer product lines are acquisitions. A significant part of their ads network is also acquisitions. Most of those product lines and ad network divisions were preceded by internally-developed business units that were marketplace failures. Google does not build, it buys.
2. Android is FOSS, which neatly tidies away a lot of the "we have a right to monetize our IP" arguments that Apple loves to trot out.
3. They paid Facebook bucket-loads of money to keep them from launching a competing ad network under the "Jedi Blue" program. This is textbook illegal behavior and doesn't require any new theories of antitrust like, say, the Epic lawsuit did.
4. Google doesn't comply with litigation holds. Civil litigation is a Constitution-free zone, you can't take the 5th in this kind of trial, meaning that you can be forced to self-incriminate, and if you do not do so, the judge is free to make adverse inferences.
5. Google doesn't have a single charismatic(?) leader. Apple had Steve Jobs, who could invent the monopolistic business model and then conveniently die[0] before anyone with power started to question it. Google has a bunch of executives that all have to e-mail one another, which means that even with their cavalier attitude towards recordkeeping, something winds up getting recorded, making the missing information even more incriminating.
The one thing I can't answer with this set of answers is why Facebook isn't getting as much scrutiny as Google is, especially since Facebook has proven to be such a good conduit for scammers and disinformation. Facebook has even less ability to leverage "copyright disclaims antitrust" than Google does.
[0] Constitution or no, a judge can't compel a corpse to testify.