Silver as an investment

London Mayor Tries To Justify Mass Censorship By Reading Mean-Tweets

London Mayor Sadiq Kahn read a list of half-dozen racist tweets about himself to a crowd at the annual SXSW festival in Austin, TX Monday.

as

I say kill the mayor of London and you’ll be rid of one Muslim terrorist,” Khan quoted to the audience. “I’d pay for someone to execute Sadiq Khan.

Kahn, the first Muslim mayor of a Western capital, implored tech companies to censor “hate speech,” – which has recently become a catchall from everything from death threats to opposing longstanding symbols of actual oppression, such as hijabs. 

The London mayor said he didn’t want to be “portrayed as a victim” or “ask for sympathy.” Instead, he wants to tech companies to police people’s feelings by going further in “making the internet free of hate speech.” 

“But ask yourself this: What happens when young boys and girls from minority backgrounds see this kind of thing on their timelines or experience this themselves?” said Khan.

Khan said that tweets like the ones addressed to him send a message to these children that if they don’t look a certain way or subscribe to the same establishment beliefs, they will grow up thinking there’s no path for them in high-profile careers.

“We simply must do more to protect people online,” Khan said.

Khan urged companies like Facebook and Twitter to show “a stronger duty of care” so that “social-media platforms can live up to their promises to connect, unify, and democratize the sharing of information and be places where everyone feels welcomed and valued.” –Business Insider

Too bad for those offended by mass censorship in the name of creating an all-encompassing safe space. European officials, meanwhile continue to turn a blind eye as migrants to rape, pillage and plunder with near impunity due to “cultural differences” (not hate). 

Germany began enforcing strict “hate speech” laws on January 1, giving companies such as Facebook, Twitter and YouTube 24 hours after a complaint to remove postings containing hate speech. Failure to remove the offending posts in time so will expose the platforms to fines of up to 50 million euros ($60 million USD).

as

The new law was passed last June and went into effect in October – however social media companies were given until January 1 to prepare to maintain an “effective and transparent procedure for dealing with complaints” which users can submit freely. Upon receiving a complaint, social media companies have 24 hours to block or remove “obviously illegal content” – and up to a week in “complex cases.”

The new law isn’t just for the big three (Facebook, YouTube and Twitter) either: 

Social platform giants such as Facebook, YouTube and Twitter were couched as the initial targets for the law, but Spiegal Online suggests the government is looking to apply the law more widely — including to content on networks such as Reddit, Tumblr, Flickr, Vimeo, VK and Gab. –TechCrunch

German lawmaker Beatrix von Storch’s Twitter account was blocked in January after she lashed out at New Year’s greetings sent by Cologne police in Arabic.

as

“What the hell is happening in this country? Why does an official police site tweet in Arabic,” wrote the Alternative for Germany (AfD) conservative party member. 

As we wrote in January, Social media giants face a gigantic task – which some might say is impossible. It is estimated that around 38.1 million Germans alone will use social media in 2018.

as

Now, consider that every single flagged post from millions of users must be evaluated before the post is removed or blocked. This will require Artificial Intelligence (AI) or some other sort of algo. And as algos begin to identify and remove “hate speech,” people will adapt to the filters and begin to find ways around it – changing terms, spellings, and using coded language to communicate.

What if – bear with us – parents taught their children that sometimes mean people say mean things, and one simply needs to have the internal fortitude to overcome offensive statements instead of employing mass censorship?

as