Checkmite
Skepticifimisticalationist
I recently watched a video on YouTube called "The Dead Internet Theory 2". It is as you might expect from the title a continuation of a longer thesis centered around the notion that most of the internet isn't "real", in the sense that most of the content you can find that presents itself as the interests, opinions, and decisions of actual people using the internet was actually produced by machines for various purposes - chief among them the ubiquitous "public manipulation" by governments and the corporate sector of course, but also for less personal purposes, like fabricating entire customer bases for products and services that aren't actually consumed by anyone.
In this post I'm not really interested in the larger thesis so much as the part of it that is the focus of this second video, which you can watch here if you want:
It focuses on the possibility that the claimed power and reach of Google's internet search capability is a deliberate lie.
As a supporting argument for this claim, about five and a half minutes into the video the author conducts an experiment by doing a Google search for the generic term "pizza". As shown in the video, Google initially reports at the top of the results page that its engine has found 1,050,000,000 results in just over a second for "pizza". The author proceeds to click through the results pages, each of which contains 10 results, save the first page which only has 9.
As he proceeds through the pages, at certain points the number of reported results at the top curiously changes, increasing or decreasing by hundreds of millions at a time, until he reaches the last page of results - page 35, where the same message that a page ago had said "Page 34 of about 1,340,000,000 results", now says "Page 35 of about 348 results".
He notes at this point that at the bottom of the final page is a disclaimer, where Google says it has "omitted some entries" that were "very similar" to the 348 already displayed, and provides a link whereupon he can "repeat the search with the omitted results included". The author proceeds to do exactly that, and during the new search, a whopping 1.9 billion results is initially reported this time; but he is only able to browse the results up to page 53 - for a grand total of 529 actually-displayed results. There is no means to proceed further - Google simply will not give any more pages of results, even though it claims there's nearly 2 billion more to be seen.
The author of the video concludes that Google's reporting of billions of results, which certainly sounds reasonable for a search query as broad and basic as "pizza", is an utterly baseless and false claim, intended to make users think the search engine is expansive, when in fact it can only provide a little over 500 actual results for this search. Of course, central to the author's overall thesis is that it isn't merely a case of Google's search engine being far weaker than advertised, but rather that, contrary to what most people expect as a consequence of what they've been led to believe about the internet, the content actually isn't there to be found - it doesn't exist at all..
I submit that the author's conclusions are misguided.
I decided to replicate his experiment for myself. I chose a different, but still very broad and generic term, "snow". Like with the pizza search, no further (manually-submitted) context. Google, find me everything you can on the internet that has to do with snow!
For my search, Google reports 8.86 billion results, and that report doesn't change at all until we reach "page 28 of about 276 results", along with the link to repeat the search including omitted results. With the "omitted" results "included", Google reports...8.33 billion results, until it cuts of my search off at "Page 47 of about 462 results".
This validates the search done in the video; and like the search done in the video, the result is absurd and dismissable on its face. It is self evident that there are more than 462 pages on the internet that have the word "snow" in them.
We can verify this even on Google itself; I took a step the author did not and did a new search, this time for "snow Cleveland". Google sensibly reports 60.9 million results instead of billions, and initially puts me at "page 10 of about 100 results" with omissions, "page 43 of about 426 results" without. Compare this to the 462 results returned for just "snow", hardly any of which mentioned Cleveland, and it becomes obvious there are a lot of results in my second search which were not included in the first, even though they logically should have been.
This does not prove that Google's claims about finding billions of results for certain searches AREN'T lies intended to impress users. But it does prove that Google isn't stopping at only 500 or so results just because that's literally all it can find or access; the truth is, Google WILL NOT display all of the search results it finds for your query over a certain amount, omitting the vast majority of them even when you've clicked a link telling it specifically not to do that.
The author's conclusions are wrong-headed; but his observations are valid. Why does Google bother to tell you that it has found millions or billions of results from a search query when it is clearly programmed to never actually return more than a few hundred, and to cut any search off that exceeds that limit?
In this post I'm not really interested in the larger thesis so much as the part of it that is the focus of this second video, which you can watch here if you want:
It focuses on the possibility that the claimed power and reach of Google's internet search capability is a deliberate lie.
As a supporting argument for this claim, about five and a half minutes into the video the author conducts an experiment by doing a Google search for the generic term "pizza". As shown in the video, Google initially reports at the top of the results page that its engine has found 1,050,000,000 results in just over a second for "pizza". The author proceeds to click through the results pages, each of which contains 10 results, save the first page which only has 9.
As he proceeds through the pages, at certain points the number of reported results at the top curiously changes, increasing or decreasing by hundreds of millions at a time, until he reaches the last page of results - page 35, where the same message that a page ago had said "Page 34 of about 1,340,000,000 results", now says "Page 35 of about 348 results".
He notes at this point that at the bottom of the final page is a disclaimer, where Google says it has "omitted some entries" that were "very similar" to the 348 already displayed, and provides a link whereupon he can "repeat the search with the omitted results included". The author proceeds to do exactly that, and during the new search, a whopping 1.9 billion results is initially reported this time; but he is only able to browse the results up to page 53 - for a grand total of 529 actually-displayed results. There is no means to proceed further - Google simply will not give any more pages of results, even though it claims there's nearly 2 billion more to be seen.
The author of the video concludes that Google's reporting of billions of results, which certainly sounds reasonable for a search query as broad and basic as "pizza", is an utterly baseless and false claim, intended to make users think the search engine is expansive, when in fact it can only provide a little over 500 actual results for this search. Of course, central to the author's overall thesis is that it isn't merely a case of Google's search engine being far weaker than advertised, but rather that, contrary to what most people expect as a consequence of what they've been led to believe about the internet, the content actually isn't there to be found - it doesn't exist at all..
I submit that the author's conclusions are misguided.
I decided to replicate his experiment for myself. I chose a different, but still very broad and generic term, "snow". Like with the pizza search, no further (manually-submitted) context. Google, find me everything you can on the internet that has to do with snow!
For my search, Google reports 8.86 billion results, and that report doesn't change at all until we reach "page 28 of about 276 results", along with the link to repeat the search including omitted results. With the "omitted" results "included", Google reports...8.33 billion results, until it cuts of my search off at "Page 47 of about 462 results".
This validates the search done in the video; and like the search done in the video, the result is absurd and dismissable on its face. It is self evident that there are more than 462 pages on the internet that have the word "snow" in them.
We can verify this even on Google itself; I took a step the author did not and did a new search, this time for "snow Cleveland". Google sensibly reports 60.9 million results instead of billions, and initially puts me at "page 10 of about 100 results" with omissions, "page 43 of about 426 results" without. Compare this to the 462 results returned for just "snow", hardly any of which mentioned Cleveland, and it becomes obvious there are a lot of results in my second search which were not included in the first, even though they logically should have been.
This does not prove that Google's claims about finding billions of results for certain searches AREN'T lies intended to impress users. But it does prove that Google isn't stopping at only 500 or so results just because that's literally all it can find or access; the truth is, Google WILL NOT display all of the search results it finds for your query over a certain amount, omitting the vast majority of them even when you've clicked a link telling it specifically not to do that.
The author's conclusions are wrong-headed; but his observations are valid. Why does Google bother to tell you that it has found millions or billions of results from a search query when it is clearly programmed to never actually return more than a few hundred, and to cut any search off that exceeds that limit?
Last edited:
