4 ms·
I accidentally stumbled on an interesting approach to sow discontent. There was a forum that was hotlinking a "words to images" script I had on one of my webs
by tyingq 8y ago
I accidentally stumbled on an interesting approach to sow discontent. There was a forum that was hotlinking a "words to images" script I had on one of my websites. They would post these images as their "text" instead of plain text.
It was hammering my server, so I changed up the script to return random inflammatory and/or inappropriate images if the referer was this forum site.
I was using the source IP as the rand() seed, so everyone on the forum had a different view of the images, but consistent in that it remained constant for any single user.
The resulting chaos and infighting was pretty funny until they figured it out.
- TeMPOraL 8y ago> everyone on the forum had a different view of the images, but consistent Reminds me of the endless fun with [you] tag on phpBB forums. That particular tag's function was to be replaced during rendering of a forum post with the username of the user viewing it. AFAIR it was retained in tag form in quotes, too. You'd post something like, "I think [you] is dumb in believing $something", and watch the hilarity ensue.
- 0xcde4c3db 8y agoSome years ago, indie game developer Arthur "Mr. Podunkian" Lee released a short game via a thread on an indie game forum, to similar effect [1]. [1] https://forums.tigsource.com/index.php?topic=9284.0 https://forums.tigsource.com/index.php?topic=9284.0
- w-ll 8y agoI'm confused to how this is to a similar effect? Are the images of the game, not the same as the actual game?
- tyingq 8y agoYou have to page through the discussion a bit. There's 11 pages. The game play is different depending on the machine it's being played on. Some folks get stuck on an unwinnable screen right away, some later, some play to the end, etc. Also, the screen/level names vary.
- CM30 8y agoThis is what I consider/fear the next 'step' for 'fake news' and weaponised propaganda. Giving different audiences different content, and using those differences to turn them against each other. Picture this: You've got a story which is seemingly very controversial, and will rile up one side of the political spectrum. However, this story is only shown for say, 10-20% of your readers/viewers, and the rest get an innoculous story, or one that says the complete opposite. Now the ones affected by it will post it on social media sites and forums, saying that this is proof of how 'evil' the other side is. But the others there can't see the same story, and will think said users are trolling them. After seeing them keep posting comments about how some minority group is evil with links to innocent stories about kittens and puppies, said users get banned. This causes said users to think they're being targeted by the other side, and then every story they see on the site is designed to build off that. Make the forum/social media/news site owners out to be illuminati pawns or what not and their banning policies a way to stifle dissent. Voila, the site has now basically weaponised A/B testing, and created a localised echo chamber to gaslight its readers into more and more extreme mindsets. That's basically the nightmare scenario. Imagine an enemy state like Russia or China using it against the alt-right/antifa/whatever. Or activists using it to recruit followers. Worse still, imagine it with compromised browser extensions as well. Imagine if a foreign government managed to do this in a way that could let them edit the text on fact checking sites like Snopes or major media outlets like the BBC or CNN or what not. Now that could have some worrying consequences. Still, it wouldn't be impossible to fix/detect. Set up sites like Reddit/Hacker News to automatically archive pages linked to, and send users to the archive rather than the original unless they choose to click through. Avoid using any browser extensions at all, with hosts file blocking for things like ads rather than Adblock/uBlock/whatever. Ask others to double check what every page you see says to make sure it hasn't been compromised/doesn't show different things. Do that and you can at least minimise this style of dissent sowing.
- dragonwriter 8y ago> This is what I consider/fear the next 'step' for 'fake news' and weaponised propaganda. Giving different audiences different content, and using those differences to turn them against each other. That's not the “next step”, it's an established practice. Notably, the recent Russian propaganda effort directed at destabilizing the US via false flag social media accounts and other channels was noted for doing exactly this. (They didn't invent the general concepts either, it's much older.) The actual mechanism in your example would be a bit novel, but it's also fairly easily accidentally penetrated, because real humans aren't 1:1 to online personalities.