The Rise of Hate Search
nytimes.com
nytimes.com
Is it actually clear what the searcher thinks? I enter phrases I'm searching for into Google. I don't just type my current thoughts in there to let Google know.
In response to "are all Muslims terrorists" I would expect to get an informative page detailing that no, obviously they are not (I know muslims personally who are as western, educated, and civilized as any of my Christian friends) while also explaining why some muslims or muslim countries are uncivilized places that have not been exposed to basic things humanity figured out during the enlightenment. (In these places Apostasy is a crime punishable by death, for example.)
So it's a reasonable question with a reasonable answer.
The search "I hate muslims" is also quite reasonable, and I would expect a reasonable search result to be about how to deal with personal bigotry in the face of a mixed population. For example, if a shop owner has 17-30 cases of armed robbery occurring every year, and all of them are by African-Americans, I might expect this person to search for "I hate blacks" and get an informative page about why the fact that some African-Americans are thugs is no reason to generalize and be bigotted. At the end of the day, the shop owner's next hire might be an African-American, and one who wouldn't think of armed robbery any more than the HN readership does.
Here possibly I'm being hopelessly optimistic. I do think it's up to Google to return search results that are not hate-speech based. In the United STates with good reason, blind hatred advocating violence against a group is not protected political speech - you couldn't just publish a book advocating lynching every (minority group). But that doesn't extend to the whole of the Internet, and people can say whatever bigoted stuff they want.
That doesn't mean search engines need to return those pages. They can return reasonable pages instead.
Your search results there will already have been filtered by google to things you agree with based on your previous search history. So if you search for that kind of thing a lot, you'll likely see a lot of news items about Trump and rants by angry people talking about how all muslims are indeed terrorists.
Just one of the many ways our modern technology works to create divisions between us, all based on the best of (marketing) intentions.
EDIT: I see your response but I'm confused. did you go to http://www.google.com/custom ? there's no signin or anything like that, it's just an old page that's still up, from when they were cobranding the page or something.
EDIT in response to your edit: That page redirects to https://cse.google.com/cse/ for me, which requires a sign-in. There's a search box there, which only searches under google-owned domains, which I guess is not what you're talking about. Might be a regional thing.
To be fair, the link I posted is obviously a fluke - it could be removed at any point. it doesn't even have the current branding. But it's very useful!
[1] https://www.browserling.com/
[2] https://www.kproxy.com/ (after entering the URL then on the result page there's a popup banner you click to hide, it happens to cover the input field. click at the far left of the banner to hide it.)
Despite being a bias this feels like nothing, because your beliefs about reality are aligned with the paradigm. You can only feel the bias when your beliefs are very much not aligned with the paradigm.
"Are all white people racists?" You can feel that one, can't you?
Unless they're searching those phrases for a myriad of other reasons that have nothing to do with that person being a racist. In which case, your argument kind of falls apart.
>"Are all white people racists?" You can feel that one, can't you?
I'm not sure if that comment is more presumptuous than it is condescending, or vice versa.
The searches phrased as questions strike me as likely to have a more unified intent, but even then there may be a lot of variation.
Of course it is. Otherwise there would be little basis for the article.
It's not like these otherwise repulsive phrases may be (in part) the result of ironic or dark humor, or that they were searched out of curiosity—perhaps even concern.
That might imply searches have some sort of greater context beyond a mere string of text.
/s
Actually it does. I know I'm proably a minority, but I have typed such questions into google (well, DDG to be fair) to find such topics on sites likes skeptics.stackexchange.com or just when I'm bored and feel like reading some of the huge crap which is out there on the internet.
It really is amazing how different the internet looks if you do searches in other languages, or phrased in such a way as to imply you're from a specific country, a certain race, gender, hold a certain political belief, etc etc
It is sad that NY Times had turned into yet another "news generator".
> The search data also tells us that changes in Americans’ policy concerns have been dramatic. They happened, quite literally, within minutes of the terror attacks.
I don't know how they got this "within hours" and "literally, within minutes" data, when they previously say the data was weekly
> Yes. Using weekly data from 2004 to 2013, we found a direct correlation between anti-Muslim searches and anti-Muslim hate crimes.
http://www.express.co.uk/news/uk/597254/ISIS-Map-Europe-Terr...
That sounds far too low. Since most people don't search like that, I'd imagine "hate boss" probably has a far higher amount, with the collective tally for queries that have near-identical semantics being astronomical.
https://www.google.com/trends/explore#q=ISIS%2C%20i%20hate%2...
I hate it that we're so spineless to venture an actual judgment without the support of some "study" or "statistic" stating the obvious.
Of course tons of people searching for "kill Muslim" matters.
In fact, it would matter even if it wasn't correlated with increase with hate crimes (though, of course, it obviously is).
Yeah, but my point is we could just use mere prior knowledge of how the world works to rule that out. I wouldn't expect many people to be typing "kill muslims" in order find muslim related hate-crime reports, especially in the aftermath of an terrorist attack attributed to muslims (and I wouldn't expect reality to be counter-intuitive to such an expectation either).
Yes, and my point is that presumption is enough in lots of cases -- and especially enough to be sure that "kill muslims" as a search term shows negative "emotional involvement" and not a desire to find out about muslim hate crimes...
And yes, it can be a real problem if people can't reason out the simplest things without assistance from "experts" and "research".
Especially since ideally they should be able to judge those experts and that research too, not take it as a gospel.
Common sense can get you a long way. And if you lack common sense, then the finest and more interest-free research wont really save you.
Maybe the future of poll is a kind of a global real time sentiment analysis. Imagine someone delivering a speech or participating in a political debate and having a feedback in real time.
PR companies for politicians etc have been doing these sort of things (with Twitter etc) for several years...
> are all muslim extremists
> are all muslim marriages arranged
There was a pretty bad "accidentally" racist search correction of Google. If you searched "white people stole my car" it responded with "Did you mean: black people stole my car". Apparently they fixed it by now.
Of to Google "Are all The New York Times articles shitty journalism?" Our research shows that in the months after these searches, the number of shitty articles on The New York Times seems to rise. We know this before the FBI does so, because we wield the power of Google Trends.