I'm pretty upset about it honestly. TikTok's algorithm has always done a fantastic job of providing interesting clips in a way that Facebook and Instagram has never been able to provide. I will say that upon a new account, it's mostly garbage, but it quickly learned what I was interested in and what I would tend to engage with. It also does this while showing me considerably fewer ads than the meta platforms.
That said, the algorithm got noticeably worse after 2021. Maybe because of the TikTok shop. I’ve categorized around 3,000 clips into different collections (with 600+ being in “educational”) but that fell off over the last few years. I would be a lot more upset about the ban if they had maintained quality, but now I’m like well, whatever.
Serious question: Why is the back end learning of so many human habits not creepy to you? It was weaponized once with how they created armies of teenagers who called their local representatives and made threats.
> It was weaponized once with how they created armies of teenagers who called their local representatives and made threats.
This is an extremely dramatic way of saying "they had a banner in the app which informed users of a policy that affected them and directed them to contact their representatives", something which plenty of other social media platforms, such as Reddit have done without controversy.
> Why is the back end learning of so many human habits not creepy to you?
In the context of banning TikTok for this particular reason, then X.com, Facebook, Instagram, and YouTube are all equally guilty in how they both boost and suppress different types of content based on interference from local government.
They weaponized the backend to drive that by injecting in recommendations for that inorganically, or that was something that emerged from having a recommender system backend?
With the billions[1] they already make on ad revenue I guess? I don't care. For me it's either ad blockers or I don't use the service.
YouTube is only still viable because the barriers to start a competing service are so high. Just like every other Alphabet-owned service, their single-minded focus on serving the bare minimum content merely as a vehicle for ads has just decimated the quality of their products.
I work at a Starbucks. The scanners are usually customer facing and all the barista needs to know is that you are going to pay with an app on your phone. As long as there are no complications I doubt the average barista will know or care exactly how you paid. As for the crowd sourced gold card, Starbucks' loyalty program gives you a special card after 30 transactions (Gold Card), once you hit that level, you get a free drink or food item. Every subsequent 12 transactions earns another free drink or food item. These would obviously be earned very quickly if multiple people are using the same account. However, I feel that if this is the way this app works, that it opens itself up to both Starbucks shutting it down quickly and user exploits. You could just generate your own image of this accounts barcode and use it whenever you'd like. The worst that would happen is that the barista would inform you that there wasn't enough money on the card, or that it was an invalid account, something that happens often. So I doubt this is the case with this app, I am interested in how they are running this though.
Edit: Up above it is mentioned that each customer gets their own account. So this isn't an issue.
reply