This is a frank discussion about "search advertising", probably from someone at google. If it was not already clear to anyone, this explains it clearly:
"... we could mostly ignore the demand side of the equation (users and queries) ..."
In other words the people who use google search and what they want to know can mostly be ignored.
Instead the focus is on "advertisers" where advertiser is a euphemism for anyone who wants to target people for any purpose. This purpose can and does include disrupting our social order, promoting extremist causes, selling products that are unfit for use or products that are prohibitively expensive.
In case you have not been paying attention, welcome to clickbaitnet.
Seems to me like there are two disjointed teams worrying about two different things. Search org is worried about daily active users and user experience, and ads is worried about monetizing search. The reason finance doesn't have to worry about demand is because search org is worrying about it.
Did that person work on the Youtube search? You can search for any topic and videos from your feed keep appearing in the search results, no matter how unrelated.
Google’s lawyers should make the case that they can’t be trialled for monopoly on the search, because it’s one aspect that they don’t even do anymore.
What's missing is that they've exhibited monopolistic behavior toward their "supply" side (advertisers) as well. "Tear the economics textbook in half" is a weird way to say "follow exactly what the book says regarding monopolies".
"They're opposite" is the most incorrect statement in this thread. Yes, there is a difference, but the effects can be similar. Either way, market participants are denied a fair price because of their counterparty's dominance. This very case demonstrates it, as does Amazon's. That's why our laws treat them similarly, even using the word "monopoly" to mean both ... as do professional economists when both are under discussion. This "I know a word you don't know" game is tiresome, especially when people do know the words (and apparently better than you).
There’s nothing surprising here. We’ve known for two decades that Google and others treat users as the product being sold to their actual customers - the advertisers.
We need to stop being shocked that the companies themselves, addicted to ad revenue, did some truly despicable things and most of the people at them (yes, many on this forum) actively enabled these behaviours and acted as apologists for them.
Seriously, fuck Google and fuck Meta and if your company is ad driven then it’s on you too.
Ad blocking is a moral imperative. We should all do what we can to drive these advertiser profits to as close to zero as humanly possible. This will fix the web.
I spent too much time trying to parse the meaning of this headline. I wish HN would train a regressor for the difficulty in understanding a headline's meaning, and then automatically restrict them past some threshold.
It is a noun phrase (which means there's no main predicate verb) where 4 words in the middle could be read as verbs. So yes, it can actually take a few tries to parse the headline correctly.
What is the correct parsing of the headline? At first I figured it was a regular garden-path sentence [1] but after multiple readings I still can't figure it out.
[1] A garden-path sentence is one where you find at the end that the expected parsing is wrong. e.g. "The old man the boat" or "The horse raced past the barn fell".
https://en.wikipedia.org/wiki/Garden-path_sentence
New evidence was released in the DOJ vs Google case. The judge for the case called this evidence "embarrassing" to Google.
It's a bad headline. There's no a complete sentence here. "DOJ vs. Google evidence release" is a noun phrase ("release" here is used as a noun), modified by the clause "that judge calls embarrassing exhibit". In that clause, "judge" is the subject, "calls" is the verb, "that" is the direct object, and "embarrassing exhibit" is an object compliment, which refers to the same thing as "that". So basically we just have a big noun phrase.
Not having to worry about the demand side of the equation == no interest in gaining the consent of the advertised to.
The advertising industry has a problem with consent.
Q.E.D.
I will not accept anyone who fundamentally tries to use platitudes to go "but it's just ads". No damnit. It's not "just ads". My answer is no, not only do you not ask, but when you go and do it anyway, you get pissy when I start blocking everything to do with you.
Anything that can be destroyed by the truth deserves to be destroyed by the truth.
To hell with Google, and to hell with web advertising. Let it all burn.
I've actually already had that solution architected for years.
Web partitioning.
One is an indexed information retrieval system, purely intended to be a non-commercial/mercantile address space.
One is designed as essentially a business registry. You must be a business (in the legal fiction sense) to be a part. You can advertise to your heart's content in the mercantile web. In the non-mercantile, all toxic/polluting forms of SEO will be pruned with extreme prejudice.
The problem? There's no way in hell that there will not be low-key infiltration efforts by the mercantile side of things into the non-mercantile, simply because that'd be leaving money on the table. Further, the non-mercantile one will probably be so absurdly resource constrained as to eventually be subsumed by the mercantile Net in one way or another.
There's also still the problem of administrative authorities being pressured into intentional distortions of search indexes.
Pretty much. When I was younger, modeled it quite literally off of "neighborhood grids". The idea being communities + their networks mirror one another. Grids would end up becoming infrastructure items maintained on however granular a municipality decides it wants to. Instead of a wild hodgepodge of differently financed networks (some bad because not wealthy, some good because wealthy), we'd focus on codifying and rolling out "this is the minimum quality of service to be expected period".
The reason I architected things that way was because I strongly believe that the best way to bootstrap network savvy for the widest possible audience is to bootstrap off of ways of life everyone no exceptions can relate to. Road systems are the prototypical network we all end up learning from.
By mirroring civil infrastructure, you can also cultivate an appreciation and awareness w.r.t network traffic, and propagation in and out of a space. You essentially end up, with local groups of network operators able to coop to keep their surrounds operational.
If this is sounding a lot like municipal networking, that's cause it is. Networks should be the realm of communities, and not seperable by mere financial accident.
The bloody business only net would reside in what I call "the remote net" which is basically the net we have now, completely divorced from any type of local geographical constraints. The reason for it not to the necessarily "default net" is to specifically allow for things like a community engageable desync from the web at large if necessary, because back then, I was far more interested in empowering groups of people living together to coordinate amongst themselves uninterrupted rather than faciliatating top down control by "remote zones" of local nets, like you see starting to materialize today in the form of the practice of deplatforming.
A new arrival (child's) onboarding process to the world would basically simplify down to one simple practice: learn to navigate and travel the real physical world, adding on increasing scopes and breadths of access as they can handle it. The shift toward algorithmic thinking, and getting an intuition for navigating these networks would just materialize in the same way that one learns ones neighborhood, local library, local municipality, next town over, etc.
The final net gain, har har, is that one could over time basically decentralize every form of networked interaction.
Yeah, it's pretty naive as *f given today's world. But the idea of Nets of data ever widening in scope, preserving data associated with the local inhabitants over time at least provides a framework of organization for any locality, or handful inhabitants to organize themselves with, and most importantly, express their own digital sovereignty with (which includes individuals setting up straight to remote grid connections, bypassing the local grid systems entirely).
It was really more of a passion of getting the onramp to global networking a bit more navigable to the less tech inclined, and creating abstraction layers whereby municipalities can more effectively protect community subnetworks in some fashion.
Agreed. Our attention is ours. It doesn't belong to these people. They are not entitled to it. They do not get to insert little brands and taglines into our minds without our consent.
I loot authors I read for grammar, and I picked this and “nb” up from David Foster Wallace, who used both all the time. This usage of “which” is really handy for achieving clarity in certain situations.
If you wanted to rewrite it as a sentence and weren't constrained by conventions used in headlines, you could rewrite it as, "Evidence released in DOJ v Google case includes an exhibit that the Judge called embarrassing [for Google]"
Now I'm wondering what that noun phrase would be in Dutch, since that would make it clear it's one thing and not a phrase with a verb inside. Maybe "DoJ-versus-Google-bewijsmiddelpublicatie gênant genoemd door rechter"? Any Dutch in here that can say whether that's easier to read than the English title?
The problem is that all of "release", "calls", "exhibit", and even "judge" could be verbs, making this rather difficult. Especially if the title of this news turns out to not be an action ([This PDF is] ...)
Even now it doesn't really make sense. This PDF is not an "evidence release", it is either evidence or a statement about released evidence. Unless that is a term of art in law?
Well it’s true, they don’t have to worry about amount of demand because everyone needs search and Google dominates that market, with no one challenging their >90% market share or the use of that kind of search to find things on the Web.
I don’t think ads are really the “supply,” though.
This seems like one person writing down their personal observations, which are not super deep.
What is your definition of challenging? There are multiple other search engines that make up your admitted <10% share. That sounds like challenging to me. Your definition must be closer to neck-to-neck competing which is unrealistic expectations
At a feature level Bing is super competitive, if not outright superior with their AI search features. It hasn't proven effective yet in becoming people's idea of the defacto search, but it is competing (see Googles humiliating AI launch).
I mean, it should be a pretty easy case, Google search is objectively terrible and their other practices are extremely shady as shown repeatedly. Obviously law isn't that simple but at least the judge appears to be seeing through their behavior!
Demonstrably terrible results for the simplest of queries, returning drive by malware as preferred results etc, there is nothing subjective about that.
a query for "why does my heater take a long time to turn on" will turn up many first-page results like this one: https://candcheat.com/10-reasons-why-your-heater-fails-to-tu... which is shallow SEO content aimed at getting you to pay for a fix rather than answering the question.
I run into these kinds of pages constantly as I try to debug common household issues (HVAC, electrical, automotive, etc.)
Good example, thanks. SEO spam like this is indeed a problem. Perhaps the underlying issue is that GSearch was originally designed to "find webpages," not "answer questions."
(I'd expect that high-quality, official webpages exist for queries like "Nest Thermostat G4CVZ operating manual" or "HVAC enthusiast forum." But that's not the case for a question like yours, and SEO spammers have fooled Google to become the best result.)
Adblock doesn't hide search results, child comment gave an example but in many cases it's worse, worthless SEO crap is annoying but not quite as bad as the ones that clone genuine content to obtain higher rankings and then attempt to deliver malware...
Except their results are still sub par, which is the whole point of the case - Google has abused it's position to make competition difficult to impossible.
54 comments
[ 3.4 ms ] story [ 290 ms ] thread"... we could mostly ignore the demand side of the equation (users and queries) ..."
In other words the people who use google search and what they want to know can mostly be ignored.
Instead the focus is on "advertisers" where advertiser is a euphemism for anyone who wants to target people for any purpose. This purpose can and does include disrupting our social order, promoting extremist causes, selling products that are unfit for use or products that are prohibitively expensive.
In case you have not been paying attention, welcome to clickbaitnet.
> probably from someone at google.
Specifically, it's "meeting notes that Google’s vice president for finance, Michael Roszak, created for a course on communications."
I’m guessing the actual scraping occurred much earlier than that though
Google’s lawyers should make the case that they can’t be trialled for monopoly on the search, because it’s one aspect that they don’t even do anymore.
Which is why it doesn't make much sense to call a middleman either one, when they're both.
We need to stop being shocked that the companies themselves, addicted to ad revenue, did some truly despicable things and most of the people at them (yes, many on this forum) actively enabled these behaviours and acted as apologists for them.
Seriously, fuck Google and fuck Meta and if your company is ad driven then it’s on you too.
Is it really that difficult?
[1] A garden-path sentence is one where you find at the end that the expected parsing is wrong. e.g. "The old man the boat" or "The horse raced past the barn fell". https://en.wikipedia.org/wiki/Garden-path_sentence
It's a bad headline. There's no a complete sentence here. "DOJ vs. Google evidence release" is a noun phrase ("release" here is used as a noun), modified by the clause "that judge calls embarrassing exhibit". In that clause, "judge" is the subject, "calls" is the verb, "that" is the direct object, and "embarrassing exhibit" is an object compliment, which refers to the same thing as "that". So basically we just have a big noun phrase.
Not having to worry about the demand side of the equation == no interest in gaining the consent of the advertised to.
The advertising industry has a problem with consent.
Q.E.D.
I will not accept anyone who fundamentally tries to use platitudes to go "but it's just ads". No damnit. It's not "just ads". My answer is no, not only do you not ask, but when you go and do it anyway, you get pissy when I start blocking everything to do with you.
Anything that can be destroyed by the truth deserves to be destroyed by the truth.
To hell with Google, and to hell with web advertising. Let it all burn.
Web partitioning.
One is an indexed information retrieval system, purely intended to be a non-commercial/mercantile address space.
One is designed as essentially a business registry. You must be a business (in the legal fiction sense) to be a part. You can advertise to your heart's content in the mercantile web. In the non-mercantile, all toxic/polluting forms of SEO will be pruned with extreme prejudice.
The problem? There's no way in hell that there will not be low-key infiltration efforts by the mercantile side of things into the non-mercantile, simply because that'd be leaving money on the table. Further, the non-mercantile one will probably be so absurdly resource constrained as to eventually be subsumed by the mercantile Net in one way or another.
There's also still the problem of administrative authorities being pressured into intentional distortions of search indexes.
The first one kind of exists in various commercial forms but they are really expensive. Would your version be a kind of public utility?
The reason I architected things that way was because I strongly believe that the best way to bootstrap network savvy for the widest possible audience is to bootstrap off of ways of life everyone no exceptions can relate to. Road systems are the prototypical network we all end up learning from.
By mirroring civil infrastructure, you can also cultivate an appreciation and awareness w.r.t network traffic, and propagation in and out of a space. You essentially end up, with local groups of network operators able to coop to keep their surrounds operational.
If this is sounding a lot like municipal networking, that's cause it is. Networks should be the realm of communities, and not seperable by mere financial accident.
The bloody business only net would reside in what I call "the remote net" which is basically the net we have now, completely divorced from any type of local geographical constraints. The reason for it not to the necessarily "default net" is to specifically allow for things like a community engageable desync from the web at large if necessary, because back then, I was far more interested in empowering groups of people living together to coordinate amongst themselves uninterrupted rather than faciliatating top down control by "remote zones" of local nets, like you see starting to materialize today in the form of the practice of deplatforming.
A new arrival (child's) onboarding process to the world would basically simplify down to one simple practice: learn to navigate and travel the real physical world, adding on increasing scopes and breadths of access as they can handle it. The shift toward algorithmic thinking, and getting an intuition for navigating these networks would just materialize in the same way that one learns ones neighborhood, local library, local municipality, next town over, etc.
The final net gain, har har, is that one could over time basically decentralize every form of networked interaction.
Yeah, it's pretty naive as *f given today's world. But the idea of Nets of data ever widening in scope, preserving data associated with the local inhabitants over time at least provides a framework of organization for any locality, or handful inhabitants to organize themselves with, and most importantly, express their own digital sovereignty with (which includes individuals setting up straight to remote grid connections, bypassing the local grid systems entirely).
It was really more of a passion of getting the onramp to global networking a bit more navigable to the less tech inclined, and creating abstraction layers whereby municipalities can more effectively protect community subnetworks in some fashion.
an exhibit, in this context, is a piece of evidence
this one was embarrassing for google
"DOJ vs Google evidence releases, which evidence judge calls embarrassing"
found it!: https://news.ycombinator.com/item?id=37629780
"DOJ v Google evidence release" is a noun phrase.
If you wanted to rewrite it as a sentence and weren't constrained by conventions used in headlines, you could rewrite it as, "Evidence released in DOJ v Google case includes an exhibit that the Judge called embarrassing [for Google]"
Even now it doesn't really make sense. This PDF is not an "evidence release", it is either evidence or a statement about released evidence. Unless that is a term of art in law?
I don’t think ads are really the “supply,” though.
This seems like one person writing down their personal observations, which are not super deep.
I run into these kinds of pages constantly as I try to debug common household issues (HVAC, electrical, automotive, etc.)
(I'd expect that high-quality, official webpages exist for queries like "Nest Thermostat G4CVZ operating manual" or "HVAC enthusiast forum." But that's not the case for a question like yours, and SEO spammers have fooled Google to become the best result.)