3 ms·

> In this case the "normal" checkmark will lose any meaning No, it would mean 'this is a real person, and this is their real name', which is something that is

• Nextgrid 7y ago
  What would be the criteria for a "real person" checkmark then? There's still no universal, government-provided API that would allow you to check whether an ID is real, so bad actors can just cheat the system by providing fake IDs. And even if the ID problem is somehow solved, it wouldn't prevent bad actors from just paying people to create accounts under their own IDs and post their propaganda. The current blue checkmark is being sought after because of its exclusivity, if this disappears then the attention-seekers will cling onto something different to make themselves look special and supposedly better than you. Also, the reason the current checkmark is both exclusive and yet this fake politician was able to get one is because Twitter wants to maintain good relationships with powerful and/or popular people, and the exclusive checkmark is kind of a welcome gift. Removing the exclusivity would totally nullify its utility as a gift, which is why they keep the whole verification system in place despite being completely broken (as this incident proves yet again) and spew some BS about working on improving it just to buy themselves time.
  • ClumsyPilot 7y ago
    This is total lunacy. It has nothing to do with exclusivity and everything to do with verification. There are established KYC processes followed by every serious company, like a bank. Yes, there is no straightforward api, but that is irrelevant. Banning real people would actually have an effect, as you could only sell your ID once.that would massively drive up the cost of using fake account to post propaganda. People can sell out in real life for real, you can bribe real voters too ( and sometimes even officials!)
    • Nextgrid 7y ago
      > This is total lunacy. It has nothing to do with exclusivity and everything to do with verification. In this case they could admit the current system is broken and retire it completely (until they come up with a better solution). They don't, and just make up BS reasons to buy time because it turns out the system as it currently stands is good for them. The exclusivity (even more so now that the system is supposedly on hold, so you can't apply for verification as a layman even if you do meet the criteria like being impersonated, etc) is a feature that Twitter wants to preserve without admitting it. > There are established KYC processes followed by every serious company, like a bank. Banks' KYC processes operate within a way narrower set of constraints than a social network: * They are usually limited to the local country, so the main process only have to support the most common forms of ID available in the area (edge-cases such as foreigners can go through a fallback, more manual process). A social network will on the other hand need to support every single possible form of ID. * The bank can invest much more money into the verification process than a social network which would be lucky to make a few bucks a month per user. * Conventional banks use physical branches to open accounts, so it raises the risk for a potential attacker as they would now need to show up in person with their fake ID and the fake document has to match their looks. This significantly raises the cost of the attack and the potential risk the attacker incurs in carrying it out. Social networks can't do this. * Even for online banks where you can open an account online, you at least need to provide a mailing address to receive the card associated with the account so you can spend the money. So while you might indeed be able to trick an online bank into opening you an account with a fake ID, you wouldn't be able to use it until you somehow collect the card. This again significantly raises the cost of the attack and requires the attacker to be local (or have an accomplice) and take more risk than a fully online attack carried out from behind an anonymization service such as Tor. > as you could only sell your ID once There's a ton of people out there that have never even heard of the social network, couldn't care less about it, but would be happy to get some $$$ in exchange of something that is very unlikely to affect them directly (if they don't use the social network they most likely wouldn't even know or care about their identity's reputation on it). The potential stigma associated with loaning your account will become less of an issue if the practice becomes mainstream and everyone is doing it. In fact I'd argue it's already mainstream in the form of being an "influencer", and a lot of people, especially younger ones want to be one. In fact, people want to become one so much they apparently post fake sponsored content nowadays: https://www.theatlantic.com/technology/archive/2018/12/influencers-are-faking-brand-deals/578401/ https://www.theatlantic.com/technology/archive/2018/12/influ... > that would massively drive up the cost of using fake account to post propaganda In a third-world country you could convince people to do pretty much anything for what is pocket money for us. What's the solution then? You can't just ban these countries or make their verification have less weight than let's say a US verification without getting accusations of racism/xenophobia/etc and a huge PR disaster on your hands.
  • Beldin 7y ago
    > There's still no universal, government-provided API that would allow you to check whether an ID is real, Colleagues of mine have created just such a system: IRMA. Short for I Reveal My Attributes. The way it works: The government (or anyone else) can approve an attribute of yours, e.g. "over 18". You can then choose to share this verified attribute with whomever requests it. The requestor can then see who verified that you have the requested attribute. More info: https://privacybydesign.foundation/irma-explanation/ https://privacybydesign.foundation/irma-explanation/