Hacker News new | ask | show | jobs
by elijaht 1315 days ago
I actually liked the ~idea~ behind this but not the execution, which I can't believe was so bungled

In my ideal world, Twitter Blue would act as "identity verification as a service". Pay $X/month and provide some proof of identity linked to your display name on Twitter. If the two correlate, you get the check. I think identity could be fairly flexible- could be "I am John Smith", but also could be "I am CorpA" or "I am <Internet Identity>"

Verified accounts could then get priority display of tweets or replies. This cuts down on spam severely as now the barrier to getting high visibility tweets is $$$ AND verification.

What about profile name changes? That's why it's a monthly payment. Allow Y changes per month still subject to the same verification process

What is the cost of doing so? I feel (maybe naively), fairly small. I think verifying I am John Smith is super easy for non-notable people (no one is trying to impersonate you)- send a driver's license or recent bill and you are probably okay. For more notable people/corps, you will need to provide higher documentation but at the same time, that's a much smaller # of accounts (and currently done today)

5 comments

> I actually liked the ~idea~ behind this but not the execution, which I can't believe was so bungled

How do you not bungle something you conceive, implement, and roll out to millions of users in 2 weeks (while in the middle of a massive wave of layoffs).

By not missing on the core feature? The MVP for this was not taking payments and putting a sign, it was having an actual verification, no matter how basic.

I also don't understand how it got to this bondoggle. I always imagined it was going to use the payment for verification: you can get a blue mark for John Smith only if you pay with a card belonging to John Smith. That can be automated in 2 weeks, no problem. And then add more complex verification procedures for companies, brands and non-name accounts.

I'm also reserving some probability mass for "we didn't get the whole story". Something like 25% that was an already verified account or something like that.

He's moving fast and breaking things.

Eventually I suspect things will get fixed and Twitter will probably benefit from the attention. Twitter hasn't been this interesting in years.

Is this the opinion of an active twitter user, or from outside? He's certainly making things interesting for spectators but the people who were already there using it don't seem that pleased from what I can tell.

Also idk about you but I've been in tech about a decade, seen a lot of the changes the startup era has wrought on our world and I can't hear "move fast and break things" as anything positive anymore. I wish we had gone a little slower, we broke some things I think we needed.

"May you live in interesting times" is a curse, not a blessing. People tune into disasters. Trolling aside, twitter has become hugely introspective over the last few weeks, which isn't particularly interesting.
At the rate it's going possibly he might just shut down Twitter for a few months and then relaunch it. More excitement and less of all this
> Twitter hasn't been this interesting in years.

Maybe for some. Black Twitter had some great sagas.

I thought it would use a service similar to this from Canada Post [1]. Obviously would only work for Canada, but I'm sure many other countries have a similar service. People will balk at the in-person side of things, but it would create a safer service if your real identify was verified. For users, not wanting to use their real name, perhaps they could opt into a low tier service with a different badge.

I know there are many online services providing a similar service, not sure how secure they are, but obviously another option.

1 - https://www.canadapost-postescanada.ca/cpc/en/business/posta...

Woah what is this thing?

I've never been prompted to use it though, have you? in which context?

I know (from experience) that HireRight use it for background checks.
I've used it once, but I can't remember why. Whatever company was using it, sent me a postcard with a barcode or something on it. It gave instructions for me to walk into any Canada Post and hand over my ID and the post card. The staff verified my ID and then scanned the barcode. I was then verified. It was quite a while ago, likely before most of the 'online ID verification' companies existed. I'm not sure how they compare in terms of cost, accuracy and compliance.
From tweets Elon made before becoming the CEO of twitter, it seemed this was the approach he was aiming for.

I can only speculate that he dropped the identity verification part to get this to market quicker as this is the part which would have required new investment/systems/vendors.

I speculate taking a payment would have been relatively low complexity for Twitter to implement and possibly could have leveraged existing internal systems.

I wouldn't be surprised if one day Elon suddenly announces everybody paying $8/mth now needs to submit identity docs to continue to use service.

I was imagining a way to filter content into three buckets:

1) All tweets

2) Tweets by paid accounts (low verification)

3) Tweets by official accounts (high verification)

Bonus points to filter content by 1st party client vs 3rd party tweet schedulers

the original tweet kinda made it sound like thats what the plan was. they already attach labels to politicians and such, having a label of 'us senator' and a blue check is redundant, by being labeled they must be verified. if they expanded the labeling to include a bunch more things(and make it more visible in the UI).. its a much better system, labels replace verification and blue just becomes premium(though to be fair id probably pick a different icon)