Reddit reviews Everybody Lies: Big Data, New Data, and What the Internet Can Tell Us About Who We Really Are

In the book Everybody Lies: Big Data, New Data, and What the Internet Can Tell Us About Who We Really Are, Seth Stephens-Davidowitz uses Google Trends extensively to research social shifts. It's not a perfect method and it can't paint us a perfect picture, but it's generally a good indicator of how a social trend has progressed.

A search on Google Trends for "jewish jokes" shows a general downward trend since 2004. "jew jokes" went up around 2008-2012 but then fell and is currently at the lowest point recorded by Google. More blatantly anti-semitic searches like "jews in the media","jews control the media","jews did 9/11","jewish elite", and other similar searches are all on a steady downward trend as well.

I'm not an expert at this stuff but it seems to me that anti-semitism is on a general downward trend overall. There are alternative explanations for these trends, like anti-anti-semitism becoming more common and having a chilling effect on peoples' searches, but it seems to me that the simplest answer is the most likely.

No offense taken. ☺️

You’re right. Secrets are not safe. Gay, straight, bi, cis... whatever it’s being labeled as these days is not the culprit IMO.

As a recovering/recovered addict, the most damaging secret is hiding a drug addiction.

This book was a fascinating read (well, to me), and it just goes to show that everybody lies.

From the NYT bestseller "Everybody Lies'

https://www.amazon.com/Everybody-Lies-Internet-About-Really/dp/0062390856

The error comes from focusing on the wrong filters. The "liberal media" is a thing (in two ways, really, but let's focus on the one people generally refer to) when you look at rhetoric. This can be determined through linguistic data analysis, which Seth Stephens-Davidowitz did in his book Everybody Lies. One of the interesting things he shows there is that the rhetorical bias varies not by ownership, for instance, but mostly through the dominant political leaning of the area in which the paper (which was the focus of the study) is sold. That is, a news outlet's rhetorical bias depends on its audience, not its owners.

This analysis is useful, but there is a glaring problem with it: In focusing on rhetoric, it ignores actual policy advocacy and, importantly, publication bias. And that's where the owners have influence. As long as the policy advocated agrees with the owners (and the media's inherent structural biases, re: the Herman-Chomsky Propaganda Model), how it is presented (the rhetoric) only matters to the extent that it influences revenue. And anything that is counter to these interests, will be ignored.

So, yes, there is a "liberal media" (and they're actually fairly dominant). Problem is, they are liberal in rhetoric only (and sometimes in actual policy, depending on what you mean by 'liberal'). What the media doesn't tell you is usually much more important than what it does.

There's a really good book that delves into this issue, it's called Everybody Lies: Big Data, New Data, and What the Internet Can Tell Us About Who We Really Are. The author talks about how when someone grows up can effect their favorite sports team, and what political party they identify as. People who came of age during Watergate grew up to be more Democratic leaning, while those who came of age under Reagan were more Republican leaning. Basically who was president when you were growing up, and what that president did, will give you either a positive or negative impressive of that president and the party they belong to.

It's a really interesting book, touches on a ton of subjects. Worth reading.

Can you explain how I'm being divisive?

I really do appreciate solidarity. My biggest take-away of the election. I'm uncomfortable with it, but I recognize how powerful something like a >50% swing in Republican views based on a changed party stance is.

I voted for Hillary and told friends to. Over 90% of Bernie supports did, too, and considering some of that 10% were probably never-Hillary or conservatives dipping left, that's some pretty good solidarity.

I'm not personally upset about stuff like super delegates. We wouldn't have Trump if the Republican's used them, and Bernie isn't a Democrat. I don't think some of the reactions of Hillary supporters to people that felt disenfranchised by their party are helpful, though. And if things like the leaked "pied piper" strategy are real/impactful the DNC fucked up real bad.


You're right, "neoliberal" isn't a great term. It's got a good amount of drift from its usage in other areas/historical (not that we aren't using roughly the opposite meaning of "liberal" as the rest of the world). Unfortunately I don't have time to write anything with more depth. The flawed shorthand would be the intersection of Bill Clinton/Obama policies and protested Bush policies, primarily the pro-war, pro-big business, pro-surveillance state policies at the cost of social programs.

I think FPTP is awful and we need to do something about campaign finance reform that both parties are disincentivized from doing (especially when in majority). I think (per this book) an unfortunate bit of politics is that when evaluating what platform to adjust to it is more effective to disregard people that are going to vote for you anyway, so you court people that skipped the previous election.

Sorry, won't be able to reply. Cheers

Okay, then how about you read the book the article was based on

One of my favorite not-super-technical books that can give some insights into the thought process and actionability of analytics and machine learning is "Everybody Lies". https://www.amazon.com/Everybody-Lies-Internet-About-Really/dp/0062390856

It touches on a concept I really like to rely on data for which is revealed vs. stated intent. People tell you they want what they wish they wanted. Data tells you what they actually want and how they actually behave. There are some good intuitive regression models in there as well.

I can't cite an exact source, but data scientist Seth Stephen-Davidowitz (author of Everybody Lies, one of my favorite books of 2017) has pointed out that the two most common Google searches coupled with "how to buy Bitcoin" are "how to get rich quickly" and "what is Bitcoin?." It seems to me that, while likely more valuable than a fiat currency inflated to serve special interests (rant, sorry), Bitcoin could have value as a market-determined currency like gold or silver, but not now. People are looking to get rich and that will cause it to crash.

Unfortunately, I don't have an easy direct source. The assertion was made by Seth Stephens, a former Google Data Scientist that had access to massive amounts of data. It was cited in his book Everybody Lies though if someone here knows how to pull Google search history by zip code/county it can be confirmed or discredited.

https://www.amazon.com/Barking-Wrong-Tree-Surprising-Everything/dp/0062416049/ref=pd_lpo_sbs_14_t_0?_encoding=UTF8&psc=1&refRID=7VVXAWNB2M8WP1R84JKQ

and

https://www.amazon.com/Everybody-Lies-Internet-About-Really/dp/0062390856

I thought the same until I read Everybody Lies which set out, for me, a convincing argument based on big data that people get more ideological diversity and come into contact with with more differing viewpoints thanks to the internet.

Fully 25 percent of female searches for straight porn emphasize the pain and/or humiliation of the woman,” he writes, citing search terms inappropriate to reiterate here, but featuring words like “painful,” “extreme” and “brutal,” and often focused on nonconsensual sex (depictions of which, he emphasizes, are not permitted on that site).

https://www.amazon.com/Everybody-Lies-Internet-About-Really/dp/0062390856?tag=nypost-20

I don't know why you said alt right, especially because it's been confirmed that Freudian slips aren't a real thing. If you want to have a conversation about an anti-fascist movement, how do you plan on doing that without mentioning fascism?

Although internet filter bubbles do exist, people are much more likely to see opposing views on the internet than irl. I don’t have a source but I read about it in this book everybody lies Think about the comment section of any newspaper.

I think an important point to mention here is that these viewers SAY they would cancel Netflix in this instance. It's easy to say whatever you want to a hypothetical poll.

​

This book about Google was an interesting read expanding on how people lie about what they're doing online because we're anonymous on the internet and it's easy:

https://www.amazon.com/Everybody-Lies-Internet-About-Really/dp/0062390856/ref=sxts_sxwds-bia?keywords=everybody+lies&pd_rd_i=0062390856&pd_rd_r=729c95c5-ba9c-4db4-b452-fa8c605a57e5&pd_rd_w=FB0Em&pd_rd_wg=IdYcs&pf_rd_p=f0479f98-a32d-45cd-9c12-7aaced42b1ec&pf_rd_r=5265D5ZMZNAX0FQW8AP5&qid=1557866692&s=gateway

Everybody Lies: Big Data, New Data, and What the Internet Can Tell Us About Who We Really Are

This book is really great for getting a surface level understanding of how big data can be harnessed. It also delves into how big data is being used as a buzz word to scam companies out of money.

I listen to audio books, so this might be a little dry for what you're looking for. The narrator was great. He definitely added to my sense or enjoyment