Live data from Hacker News

How much longer will we trust Google’s search results?

theverge.com

251–260 of 284 posts

Re: How much longer will we trust Google’s search results?

#251

The more intelligent they try to make it, the more stupider it gets.

Their attempts to make their search more intelligent ends up keeping the user from using theirs. Which is practically always a bad thing: Regular users tend to suck at searching (e.g. formulating queries) because they haven't learned it, not because they wouldn't be smart enough or because it's something you need a special talent for.

Re: How much longer will we trust Google’s search results?

#252
post #207
post #103

Earlier quoted context omitted.

After all this time watchin me, everything I ever asked it, all the ads I never clicked on, all the videos it watched me watch, all my emails it read, heck even after I worked for the dang company for nearly ten years, Google does not understand a very basic, essential thing about me. I hate advertising, damn it! I am sick of looking at it, I am sick of it looking at me. I am sick of hearing about how advertising is…

They understand that you are part of the cohort that doesn't want to be monetized. Which is the reason they don't change it for you, because the efforts required to remove this cohort from tracking and still serve useful search results are a lot higher than what they think people would pay for. But let's surprise them. I personally would be willing to pay up to 10k per decade of my remaining life expectancy to opt ou…

That's nowhere near enough. Sorry. You do want to see ads, you just don't realise it.

Google did actually run this experiment. They aren't stupid and lots of people say they'd rather pay money directly rather than see ads. Because of Google's place in the ecosystem they can actually implement that.

So they did. It eventually became Google Contributor and never really launched properly. The early version of Contributor was a lot more pure than what eventually launched. It let you set a budget and compete against yourself in the ad auction, so publishers were getting the same amount of money for their content as before. I tried it out when I worked there. You could set a fake budget and Google would pay for you to see empty boxes or pictures from your own photo albums instead of ads. A few things became apparent.

One is Google doesn't serve all or even most of the ads on the internet. So I still saw a lot of ads. This made the experience feel pretty pointless.

Another is that this model is completely infeasible. The internet economy is huge. Just a week or so of browsing rapidly blew my test budget of a few hundred dollars. To see no ads on the internet served by Google I would have needed to pay a large multiple of what I paid monthly for TV, mobile and Internet connectivity.

You say you'd be willing to pay $10k per decade? Forget it. If you're a heavy internet user that'd buy you less than a month, not a decade.

Finally, this didn't opt you out of search ads. That's because Google has strong experimental evidence that ads on search are good for users. One of the longest running A/B tests the firm ever did was selecting 1% of users and disabling search ads for them. What they found was a long term trend where those users simply used Google less than normal. Without ads the results were less useful so they did the obvious and searched less. Why would Google let you pay to make their product less useful for you?

Re: How much longer will we trust Google’s search results?

#253

Earlier quoted context omitted.

> Your hard-line position that all ads are bad, always, is laughable. This doesn't seem especially civil. > If you search for "cheapest gasoline" it's not unreasonable that an ad for an electric car might be interesting. Interesting to whom? The parent comment doesn't want to see ads. So no it's not likely to be interesting in this context. > Or if you search for "Orlando hotels", an ad for generally less expensive h…

OK Mr Literal, you're saying when you search for "Las Vegas hotels" you want the results to exclude the entirety of the Strip, "Dallas football stadium" to exclude where the Cowboys play, and "American airlines reservations" to provide the web site or phone number for Delta (which is, technically, an American airline.)

I'm not sure I understand your response.

Discerning the intent of a search is completely orthogonal to whether to serve advertisements.

Search engines are generally not great at discerning the intent of a search. For example they frequently return irrelevant results that don't match key words even when you ask for a literal search. This has been discussed a lot.

But that's an unrelated issue to whether the search should return advertisements in addition to search results.

Re: How much longer will we trust Google’s search results?

#254
post #21

Does anyone here even really use Google's search engine anymore? I do search for stuff a lot, but to be honest I know perfectly when it will be a stackoverflow page I want, or a wikipedia page, or it will have that link to maps. I can't remember the last time I searched for something and I didn't already know what domain I actually wanted to end up on.

You need to keep in mind that HN is a bubble. The responses here do not represent the masses. Yes, most people will search for 'insurance quotes' and click an ad.

Yes? Why are you asking us to keep that in mind?

Re: How much longer will we trust Google’s search results?

#255

Earlier quoted context omitted.

In theory websites could kill off googles dominance by blacklisting their crawler and only caring about traffic from DDG and other lesser known engines

and probably kill themselves off the internet at the same time

If enough websites did it, google would become just a giant resolver for corporate content that everyone already knows about. It would effectively fork the internet into google and non google. We just need a few search engines that can compete of quality.

Re: How much longer will we trust Google’s search results?

#256
post #226

Earlier quoted context omitted.

I've both read trials of 3rd parties and run campaigns where ads are a significant boost to the site already ranking #1 in organic results . The price of AdWords has certainly gone up, but the value hasn't declined.

Unless you compare to an ad free version of search, you can't prove that the value of placing an ad is more than the negative value of an ad being sold to a competitor.

Haven't ever wanted to prove that!

Re: How much longer will we trust Google’s search results?

#257

Earlier quoted context omitted.

You gave examples of what I assume you consider "bad" ads. What would be examples of "good" ads in your opinion? I mean, examples matching the definition you gave: Anticipating needs of person X, (and serve person X an ad), without any foreknowledge of person X.

another poster gave a good example--offering related merchandise on a interest-based website/forum. a made-up example might be a customer at best buy spending time in both the tv and pc departments and then having an ad pop up on a nearby screen for an intel NUC and/or apple tv (making the assumption you want a tv-connected computer). the ad didn't need to know any personal characteristics about you other than limite…

To be fair, the example you gave require facial recognition wouldn't it?

You'd have to recognize the face and track it prior to displaying the ad.

Re: How much longer will we trust Google’s search results?

#258

Earlier quoted context omitted.

>quality ads I've seen this term for years but I still don't understand what it means. I don't want to see ads. There are no 'quality' ads. All ads are designed to sell me things period. I don't need anyone telling me about what I might want. I understand ads exist to make money and if they didn't we wouldn't have the internet etc. But there's no 'quality' to them, they're a necessary evil i've come to accept and ada…

You may have never seen a good ad then. There was a period of time I would actively look at ads on penny-arcade, because the owners of the site only ever allowed ads (mostly for games) that they themselves could at least tangentially endorse. This meant I basically got hand curated ads for products I was genuinely interested in, by people who’s taste I understood if not agreed with. That said this ended up being the…

Just like the guitar mags I grew up with. I looked at all the ads, and even bought stuff based on what I saw in competing ads along with gear reviews.

For that matter, the same goes for motorcycle magazines, and just about every other kind of magazine that caters to a niche market.

Those kinds of ads can be great, even.

Re: How much longer will we trust Google’s search results?

#259

Earlier quoted context omitted.

another poster gave a good example--offering related merchandise on a interest-based website/forum. a made-up example might be a customer at best buy spending time in both the tv and pc departments and then having an ad pop up on a nearby screen for an intel NUC and/or apple tv (making the assumption you want a tv-connected computer). the ad didn't need to know any personal characteristics about you other than limite…

To be fair, the example you gave require facial recognition wouldn't it? You'd have to recognize the face and track it prior to displaying the ad.

well, face recognition is more specific than you'd need in this case.

if you went that route, you could get away with simpler person recognition (height, width, clothing color, accessories, etc.). or you could just track phones and other devices via wifi/bluetooth (and presumably uwb in the near future).

but you could also just have salespeople on the floor doing the recognition & ad placement part of it, for a lower-tech solution.

i'd actually bet trained salespeople would have better hit rates over algorithmic solutions, where access to large amounts of dislocated data is the main advantage (at the cost of our privacy).

Re: How much longer will we trust Google’s search results?

#260
post #207

Earlier quoted context omitted.

They understand that you are part of the cohort that doesn't want to be monetized. Which is the reason they don't change it for you, because the efforts required to remove this cohort from tracking and still serve useful search results are a lot higher than what they think people would pay for. But let's surprise them. I personally would be willing to pay up to 10k per decade of my remaining life expectancy to opt ou…

That's nowhere near enough. Sorry. You do want to see ads, you just don't realise it. Google did actually run this experiment. They aren't stupid and lots of people say they'd rather pay money directly rather than see ads. Because of Google's place in the ecosystem they can actually implement that. So they did. It eventually became Google Contributor and never really launched properly. The early version of Contributo…

Good points, actually very interesting, especially the budget part.

But I wasn't talking about just the ads, I was talking about the tracking to optimize those ads. I don't care about the ads. Show me ads, I ignore them anyway. I just don't want to be tracked to build out my personality profile for the rest of my life, because I don't know who will get access to that data and what it might be used for. Show ads based on my search term, not based on my browser, email and location history. They will have the same effect as the optimized ones: none.

Just keep in mind, the cohort I was talking about, was the one that will not be monetized. It is the cohort that still sees ads, but will never convert. So I'm not sure why I would buy my own ad space for an amount of money that is based on the usual conversion rate, when my conversion or click rate for my sample of n=1 is zero all the time. My profile correlates to zero spending, it is not worth the average ad price.

Google Contributor kind of proves my point about the efforts required being too costly. It was easier to inject a person as a buyer for their own profile into the merchandising system, than to rework all the tracking tools and the analytics that runs on it to stop tracking that specific person. So yes, my $10k is still not enough.

> Why would Google let you pay to make their product less useful for you?

Because I value privacy as a principle more than a slightly less useful product, and I would pay for it. But apparently I can't afford it.

Post reply on HN