Hacker News new | past | comments | ask | show | jobs | submit login

yes but imagine +1% per week growth, even if 1% per month that is a serious chunk

twitter had the advantage of nothing else like it in the SMS days

and at least half of those 600 million are bots, I don't think anyone disagrees

I'm actually impressed so far how Bluesky works right out of the box, it's like oldschool twitter and I didn't even have to break out any adblock rules or stylus modifications

I think people and outlets just need to start cross-posting and see how the traction goes, there is nothing to lose unless Musk orders automated scans to ban people.




> and at least half of those 600 million are bots, I don't think anyone disagrees

Not that I disagree (or agree) but I would like to know (out of curiosity) how would you (or the source of this information) got the number > 300M bot accounts. Do this include inactive accounts? Does this include people with multiple accounts?

And how would you determine that without access to the database? Is there a tools that can help you derive this number based on some statistical analysis? I would be very interested in the details.

Hint: I am not trying to start a debate around X policies or Musk. I just find that it could be interesting the way we can get into this number if it is supported by any rigorous methodology.


there have been several serious studies over the years trying to determine the bot count but of course the api lockout by Musk now makes it impossible

* https://www.pewresearch.org/internet/2018/04/09/bots-in-the-...

* https://www.theguardian.com/technology/2023/sep/09/x-twitter...

* https://www.google.com/search?q=study+of+bots+on+twitter

but it is clear the signal-to-noise ratio on twitter is now unmanageable especially after tools like blocking are taken away


That's interesting although from a quick skimming it seems that these studies are focused on the activity of the bot accounts. This might give a hint on their relevance but due to the fact that their activity is probably above average it doesn't tell much about their percentage of accounts (in absolute numbers). So I am not sure if there is a studies that shows that something near 50% of accounts would be bots. The number is huge and that's why it is interesting for me to see if it is true and how could we determine that.

But I also don't think that the blocking change here would affect signal to noise ratio. You still cannot engage with accounts who blocked you but can see their public posts which you could do by logging out (before X required account to show tweets) and create alt account for that. Blocking access to the already public posts/tweets didn't make sense anyway. And I don't think this is even restricted to X or any particular website.




Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: