Comment by suddenlybananas
2 days ago
It's hardly unique to Twitter but it's so frustrating how much "content" online requires an account. Whenever my friends send me something on Instagram or TikTok or a restaurant puts their menu on Instagram, I'm barraged with a thousand dark patterns forcing me to log in. Even Reddit is doing it now. I do wonder how much of this is to boost engagement versus another "wonderful" side effect from LLMs since they need training data.
The other day someone on here gave me a useful framing. Non-publicly accessible sites are the dark web. If a project told you they were moving all their docs and support to the dark web they would sound crazy, but if they say they're moving it to discord (the same thing) then it's fine.
>the dark web
I prefer the older term "deep web" [1]. It's mentioned in Wikipedia that the terms are used as synonyms now, but I think it's important to differentiate between sites behind login/authentication/etc, and sites that are still openly accessible, albeit over Tor, i2p, etc.
[1] https://en.wikipedia.org/wiki/Deep_web
> Whenever my friends send me something on Instagram or TikTok
Reply with a screenshot showing that you can’t view the content.
> a restaurant puts their menu on Instagram
Don’t go there. Plenty of other restaurants to choose from. They just lost a customer due to their hostility.
> Don’t go there. Plenty of other restaurants to choose from. They just lost a customer due to their hostility.
This is a wild take to me. Around here pretty much nobody posts their menu online on any website, the technical literacy is just quite poor. If you limited yourself to places that did, you'd only get big fast-food chains.
That customer is the exception because most are signed in all the time
I think we will see the internet closed off more and more as AI usage grows.
Twitter was doing stupid growth hacks, like requiring login to view replies, even before AI scrapers were a concern. (And I suspect many of the scrapers are just using logged-in accounts.)
> before AI scrapers were a concern
didn't the scrapers scrape everything well before concerns were raised. were the concerns raised after the LLMs were released and then people realized what happened? so, forcing logins just closed the barn door after the horses got out.
libredirect is very good for these kinda of things. But the platforms keep trying to make it stop working so is a constant battle of the frontents to build workarounds
It's tracking and data harvesting, always is