Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I'm announcing a new and exciting "doe_eyes' bicycle rider" program. The requirements are simple: you need to have a bicycle and at least $200M in assets under management with our firm.

Knowing this, if you meet a person wearing a "doe_eyes' bicycle rider" hat, is your first thought "oh hey, that person owns a bicycle"?

If not, then I think it's a bit facetious to argue that the definition is technically true. If you combine a trivial criteria with a nearly-insurmountable one, the nearly-insurmountable one prevails.



I think the problem is that Twitter's definition of "verified" - and the one Twitter's users relied on - was only meaningful for "notable" persons to begin with, i.e. persons with some kind of public record.

The core problem they wanted to solve was "if a user encounters an account that calls itself 'JustinBieber' how can the user determine whether or not they got the real Justin Bieber, i.e. the singer, or just some other random account". So what the badge in effect was supposed to communicate was "The person behind this account really is who you think it is".

It's easy to see that this entire concept falls apart for anyone who isn't a global celebrity: What does "the real Justin Bieber" even mean? Would another, not famous person that happens to also be named "Justin Bieber" not be "real"? What about account names that don't even match a celebrity but where their fans "just know" who is meant? What does "real" mean for the account of a fictional character? Or for a random nondescript username? etc.

Twitter management effectively saw all those contradictions and said "we don't care", as long as their prime usecase of keeping celebrities on the platform was served.




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

Search: