As a preliminary to how it ties in to the world of misinformation with limited speculation on what’s going on, I’ve managed to break YouTube. The speculative part. I’m assuming it’s just for myself. It’s probably related to my constant FAFO attitude on the mechanics of how YouTube operates. It seems like it’s a bug instead of deliberate foul play (i.e hacking or virus). Only time will tell. Which is important for Part 2 and maybe a Part 3 depending on my decision to keep the posts to a reasonably digestible length.
Good science requires creation of a hypothesis that can be experimentally tested and checking the results to see if there’s any merit to your theory. (Theory meaning idea not established science Theory of Gravity/Evolution type theory). In all fairness I admit to bypassing having any clue to what I’m doing, and just half assed plunge in and look what happens. However, there is a method to my madness. I engage in a lot of counterintuitive activity. On purpose. And by a lot I mean a gazillion metric tons whole lot. I’m purposefully looking ahead to “What happens if a random not very bright person does something strange or a purposefully malicious person tries to take advantage of the system?”
There are lots of parts to effective computer coding. The small taste you get in a high school level class doesn’t tell you about the parts fitting together as a whole. You learn [Perform A; Perform B; Check if C; Output X if C>0; Other Output D; Return to Start unless Counter=E] or something similar. Like a machine, the more moving parts there are, the greater the chance something goes wrong. If you code [block climate change denial] and it messes with the code for [verify legit advertising] it’s back to square one. Social media is a business and therefore trying to make money (duh). By US Constitutional standards they are allowed to operate without direct interference by the government. What I object to is the often lack of human intervention and reliance on algorithms to moderate while things are going wrong.
Years ago and for a fictitious example, you would watch one video where somebody put out content related to how they knew the winner of the big lottery prize. Then Alex Jones and company would spam the Internet insisting that the person did not really exist. Less than brilliant people would share the untrue negative story and just by math suck people into the rabbit hole of crisis actors, evil empires and hoaxes. YouTube did finally, after years of complaints, stop boosting the hoax videos instead of algorithmically just assuming that because you watched one lottery winner video you were interested in more.
But things are always going wrong and nothing is perfect. So here’s the meat and potatoes of what I’ve found. Screenshot 1: One of the top trending videos of the day. A music video, 1.6 million views. #3 trending for music. 8k comments and….. only 111 likes?!? WTF?
Screenshot 2: The same video less than a minute later. But on the browser version of YouTube, not the app. The stats on the number of likes is “fixed”. 130k likes for a top trending is much more believable.
Screenshot 3. Semi random music video from someone that doesn’t get a lot of traffic. 111 likes with only 5 views? A gazillion metric tons of WTF!?!
I’m not disclosing my exact method that is likely the cause of this. I’m fairly certain for now I can duplicate this kind of result. Here’s some other info. Every single video I checked had exactly 111 likes. It didn’t matter what channel or type of content was there. The only change was if I liked a video it would tick up the count to 112. I was able to almost duplicate this the following day where the like total for every video was 110 instead of 111.
Without spoiling the upcoming part(s) too much, imagine this happening to your average conspiracy minded drone with reality problems based in delusional thinking.
No comments:
Post a Comment