Post

There's a whole lot of conversation still to be had about proving identity, but just want to call out that this option is available right now (and super cool to see!) The only way Dylan could get a nytimes-dot-com handle like this is if nytimes set it up!

Dylan Freedman@dylanfreedman.nytimes.com

Update: I'm now @dylanfreedman.nytimes.com!

291 reposts2.3K likes

This would work if we could verify the domains themselves in the app. It's too easy to buy a look alike phishing domain and most are not going to tell the difference between nytimes.com and some punycode look alike.

The domain name should be the standard identifier in the AT protocol. Also- please remind people: This isn’t your fascist cousin’s Twitter.

It's possible if you create your account using did:web instead of did:plc But: 1. I don't know if it's possible via the current UI (only via API) 2. You'll never be able to change your domain

Then don't change it. Stay with the identity you already use if you are well known and trusted by that identity. This is how freelancers have been doing it forever on social media. You don't need to be tied to an organization if you already have a reputation independently.

The fact that so many people have a problem with DNS as proof of ownership really shows how many people's brains have been utterly cooked by corporate social media. If no one knows who you are and you don't have a website, respectfully, maybe you don't need to be "verified".

The problem is more that domain names in and of themselves are not a means of verification. If you have your-name.com, that means nothing on its own, as anyone else could register your-name.info and it'd look just as convincing. It still requires all users to check from other sources that it's you.

Serious question, if you don't have a recognizable website or employer you can be linked to then why does anyone need to know you're "verified"? Why does anyone need to know it's you if you don't actually produce anything or exist outside of skeets?

Plenty of prominent figures liable to impersonation won't always have obviously genuine website URLs or consistent employers—especially journalists (who change employers frequently or are freelance) and politicians (who don't necessarily stay politicians forever but remain important figures).

Verification exists because a child molester impersonated a supporting actor on a teen sitcom to lure abuse victims. It has been used to identify independent activists and journalists so people can't defraud the vulnerable. The people who need it are often not people you've heard of.

Lots of journalists don't *need* to have their own domain, because they work for a publisher. But If they e.g. say something pro-Palestine and get fired, their handle is lost. @bsky.app has dropped the ball a bit by not letting you keep your bsky.social handle when you switch to authing with DNS.

We can argue "Users should know to make a back-up account to squat on myname@bsky.social/have their own DNS just in case" but the whole reason this place is more popular than Mastodon is because non-tech people *don't* have to think about that kind of thing.

Can't agree more, it only takes looking at my colleagues complaining about the lack of verification to realise how much you have to lower the bar to entry bsky.app/profile/tk.g...

Hey @safety.bsky.app, can we have a "forwarding address" added to AT to assist in squatting name.bsky.app addresses when we want to use a domain name for verification. Particularly useful for journalists' adoption of news org domains as their old generic handle still exists should they leave.

When you change your username any mentions to the old username will still link to your account, even if someone else claims your previous username. Links and mentions from other websites to your old username would still be a problem, of course.

Domain names as verification is still something I think it super cool 😎 it really gives my collection of domains something to feel useful about.

Every word of this!! Publishing an account's DID in a place independently known to be controlled by the correct person/group establishes veracity much better than some central authority that can be gamed or weaponized.

One amazing integration would be to allow an API to display trusted third party badges to the user, about a handle they are viewing (hell this is probably a Jira ticket somewhere). This could also be another way to red-flag or green-flag an account for users.

What would stop me from making a fake badge that looks like nyt's. Or do you mean bsky should verify 3rd party orgs who then verify users.

It'd be more like a user subscribing to a service that would then show verifications of handles that applied for them. So imagine you, as a user, would subscribe to "verified-carpenters.me". When you are viewing posts from Jean Carpenter, they would have a "verified carpenter" badge appear.

So there's two parts. Part one is that the handle applies to be verified by a third party that does verifications. Part two is a user subscribes to this third party to gather info on handles, and that's when the badge is presented.

I'm just going to link this again. We could do journalist.ft.com tomorrow but it wouldn't have takeup because people change jobs and squatting your old name as a hack is something you have to teach bsky.app/profile/tk.g... (Might also fix your link issue if we could forward old domains)

Hey @safety.bsky.app, can we have a "forwarding address" added to AT to assist in squatting name.bsky.app addresses when we want to use a domain name for verification. Particularly useful for journalists' adoption of news org domains as their old generic handle still exists should they leave.

This could be gamified and or exploited in so many ways. Someone on the thread brought up the Bluesky user getting fired…if a TTL doesn’t exist already on the domain verification as a handle, it should be introduced. And another avenue but worth bringing up is domain compromise

What if there was a way with the UI to emphasize prominent/trusted TLDs like .gov and .edu? It could be a blue check like some people are asking for, or something completely different (i’m not imaginative) but this would not make Bluesky the sole authority since it’s still DNS

Domain verification can be subverted through. For example, someone in Will Stancil’s large coterie of trolls owns his name dot com. The troll has verified that domain on Bluesky and it’s almost impossible to know it’s not Will.

I think it’s essential to have IDs on platforms. Look at the difference between say LinkedIn and Twitter. On linked in you take care what you write, because you have one ID and it’s your reputation. On X, it’s the opposite. You can hind behind a fake ID. This breeds hate, lies and baseless comments

You can just make up a fake person on LinkedIn though. There isn't any actual independent verification. BSky's use of DNS is about as close as you can get to verification without involving government issued ID.

You can make one up of course, but if you want to build a professional reputation, or maintain a professional reputation there’s zero point doing that. There’s a big difference in the way people act when they have one profile, and build a much better environment for all of its users rather than hate

That would be true if you wanted to build a professional reputation on Twitter or BlueSky or anywhere else as well the common denominator is the desire of the user, not the verification method used. There just isn't any reason to fuck around on LinkedIn, but you certainly could if you want to.

Requiring “real IDs” is a non-starter, everyone that’s ever studied it says it subjects marginalized individuals to increased harm, pushed them away from the platform, and does little to nothing to promote “civility”

If anyone is interested I think my OK Cupid profile from DC is still up, but a decade out of date. I’ve mostly gotten doughier around the middle and lost a bit of hair.

Preventing people from impersonating public individuals or institutional actors is absolutely critical, that’s totally different than demanding all accounts use their own names.

Facebook and LinkedIn both require it. They have done pretty well. I also run social platforms for music artists, celebs, cult ATC show and alternative groups, and in those communities, everyone is not only happy, but proud to not have to hide behind fake names.

Legal name account requirements also prevent privacy and its layer of protection for pseudonymous communication, allowing outing, stalking, and worse. Vulnerable segments of the population *need* to be able to communicate without their legal names attached. Full stop.

The joys of building on existing mechanisms like domain verification…if only more people would resort to that sort of thing instead of creating new stuff for no reason! I recall when Amazon just went with HTTP requests over overly complicated SOAP for cloud scale…

The issue isn’t verifying a DID to a domain, it’s verifying a claimed human identity so you know a person is who they claim to be. Not everyone will have an obvious or relevant domain. Personally, I think a credit authorization would take care of this, since card issuers already verify identity.

A unique email address linked to a verified domain is how we authenticate into so many secure systems. So, it’s maybe 3/4s of the way there - using data enrichment similar to what Hubspot does (with some AI verification thrown into the mix) and we should be there…

what if it was a lesser known entity and we (most users) didn’t necessarily know what domain name was real?

can you just take the good parts from keybase? when using it I could prove that I owned accounts on GitHub, HN, Twitter, etc.

There are industries (like mine) who essentially prohibit employees from tying workplace to social media without strict firm supervision + compliance oversight. So…fuhgeddaboutit.

Paul... is there any procedure setup to "revoke" domain verification? A way for the domain holder to not only remove verification DNS but also force the account back to a ".bsky.social" handler or so?

Is the thinking that Bluesky will handle verification directly, or might it be left to third parties through the current domain system? Curious if there would be appetite for the bsky app throwing a badge onto accounts with handles vended from a "trusted" domain.

The likely answer is it would go unnoticed for years. When someone does finally figure it out they'll be afraid to remove it because something critical probably depends on it.

It is not the hardest thing in the world to remove a DNS record and can likely be automated while removing access to enterprise emails and services when an employee leaves.

No I know but you can't reasonably expect every one to have their own domains lying around, and additionally it requires a certain amount of technical knowledge to reach out to IT to get a DNS record added.

I think the solution is to let users subscribe to verification services like a labeler. Bsky adds their own as a default. Interface adds a badge with mouseover/click text describing identity/notability. Simple, effective, distributed, user controlled. Allow anyone for a small fee.

Maybe I should relinquish my hold on the flanneljetpack domain to give someone else a crack at being me, and see if they’re able to do a better job

I like that domain verification doesn't end up with the platform gatekeeping verification from smaller creators like Twitter's effectively did years ago.

What are the other ways one could prove their identity? There's this idea in the crypto-sphere about soul-bound-tokens that one could have multiple attestations from other reputable sources; that proves your identity.

Clearly, having a company's domain and being followed by known ppl are similar ideas in principle. I'm curious how this could be extended further to match this social-media context.

I think there should be a verification method for physicians, attorneys, scientists, etc. Could be a good way to help reduce misinformation, especially when discussing content relating to their profession.

sorry to add more here, but don't see these principles: * verification is _for the readers_ * _verification_ is 3rd party: not poaster, not reader; a trusted someone else * self-verification of any kind can be abused I suspect we just gotta pay somebody to check (public figures/orgs)

This. Readers should at a glance be sure an account is genuine w/o additional knowledge or having to check external sources. A checkmark achieves that, as it means the platform asserts authenticity. Domain usernames require users recognise domains as genuine, or check externally (which they won't).

IMHO this is the ideal answer, especially if someone launches a platform for known institutions to endorse people without necessarily having to setup direct subdomain records. I’ve seen some freelance journalists complaining that they don’t have a single institution to go to, so groups seem natural

Either doing, say, `paul.reality.app` directly and publishing endorsements on via a static site served on that subdomain, or `paul.nytimes.reality.app` for non-techy institutions. Or even `paul.idpals.reality.app` for arbitrary intermediary orgs! Could see that being popular with govt

Above all I’m dubious of the value of putting the onus on Bluesky, at least without some sort of free delegation as referenced above. It’s just too big of a job. Reddit has issues (moral and otherwise), but IMO the bones of the moderator system are something to be appreciated and emulated. Also wiki

Yeah there's a huge amount of space in what #atproto allows to have even multiple third parties doing different verifications. This is ripe for market solutions, plural. The missing middle is a way to show a "badge" that people easily trust in-app or on the web experience.

people with proven identities ruined all other social media with marketing, advertising, general capitalist interest. it's inevitable that the liberal/conservative monopoly will do it here too but it's sad to see any celebrating its advance. anonymity should be the primary concern

I don’t get why people seem to be saying, “most people don’t already have their own domain, therefore it’s unreasonable to expect people to obtain domains for Bluesky verification”. Bluesky is a new world. Maybe you had no need for a personal domain before, but they have a new use now.

just wondering: what if people want to show that they’re affiliated with an org and also keep their personal domain? have there been any discussions about this

You can verify multiple handles & then they'll all point to you if used in a post, iirc. The identifier for verification is unique to you, not unique to the domain, so just reuse it 😊

Paul domain verification is great, the media people whining about it mostly resent it being available to everyone unlike a blue tick and you guys are being too deferential. The detach post/detach blocking is a bigger and tricky issue to figure out because it enables grifts. bsky.app/profile/womb...

wombot 👀@wombot.quest

20 minutes to block @climateball.net and the design of BlueSky means his reply is invisible to everyone in that conversation now.

An open protocol (good thing!) lets you instablock and censor anyone with legit criticism and questions thanks to block/quote post detach. You can sell snake oil here and no one can point out it’s snake oil. bsky.app/profile/womb...

wombot 👀@wombot.quest

One of the worst boomer-targeting grifters in Australia is thriving on this site because all the blocking features have left no way to call them out as lying grubs. What’s the solution exactly? All your detach post and detach block features mean none of their followers will see them be called out.

Other organizations like unions or trade orgs could also serve that role Custom labels is also an option if you don't want to use another organization's domain in your handle (although labels are only visible to labeler subscribers, so the trustworthy ones would need to be advertised to people)

She is mad there’s no blue tick to indicate you are a cut above the average pleb poster, the lack of a class system pisses off people who thought BlueSky would status match like an airline frequent flyer program.

I don't think being verified means you're trustworthy at all. It at most verifies you are who you say you are (and that you're probably somewhat famous). Also when looking for breaking news with legitimate info, I would look for newspapers, etc, not individuals.

If you do not distrust this information, I do not know what you would trust. These are cryptographic signed records, which the Bluesky network needs to trust to work. Bluesky (the company) could remove the records at the moment, but never fake them (if the private keys are on a self-hosted PDS).

Yeah, the ability toll for other people to register misleading domains just degrades it to how good your SEO is to see which site is the real you in google results. I’ve seen a couple of adversarial but not malicious domain registrations already and a few malicious impersonation registrations.

Molly is right. Also this isn't realistic or quick for IT departments at large orgs. I have heard this has been an issue on the Hill where members of Congress also would need a personal page separate from the domain anyway. This is just divorced from how public figures and big orgs currently operate

It is, custom names are done at the domain level, but the account is still Bluesky hosted. It’s a simple DNS tweak where the domain tells Bluesky “that’s the username” If the domain stops telling Bluesky that, it reverts because usernames are cosmetic. That’s not what your identity is attached to

It would go a long way to have a visual indicator to confirm someone's identity. I don't know if that's a similar check mark system, but something besides a handle.