Agreed, simple things like 'there identity' also make it hard for me to focus too much on it as well.
Everyone has a lot of different personas online, it's free to do and there's zero downside.
The first idea I had with this post was definitely: it would be funny to make an over the top long identity archipelago with unrelated identities. So I guess you're sort of right.
You're right. It is four only vaguely related topics that don't come together into a real thesis besides 'online identity feels a little noisy and precarious'. The fact that I couldn't come up with a better title than 'Online Identity' maybe should have been a clue that what I was writing was a little unfocused.
Those three points are related. Facebook/Twitter/Et al wanted to play gatekeeper and now are becomining festering corpses, which has made the disparate online identity problem worse.
Because these platforms are gamified, they are worth gaming and gave rise to lots of fake automated content. Go search "best tablet" or "best lawnmower" for several hundreds of thousands of examples. Or look at Twitter.
Instead of any of these achieving a plaza of conversations, people are farther apart than they were to begin with.
The thing about a single online Identity is that there should be no way for it to be revoked against the will of the person it identifies. In real life I am who I am, and unless I choose to change that, no-one can legally take my identity from me.
There's been too many horror stories of people being locked out of their Google or Facebook accounts by Google and Facebook, even for the most minor of infractions, and that person immediately also losing access to to all the other services they used 'Sign in with...'
Until this problem is solved, I will never switch to a single online Identity for access, and I certainly will never use my Google or Facebook account to register with third party services.
[US centric] It feels like we're almost there between usps.gov and login.gov, someone at USDS just needs to make the equity and digital access case to get the funding to build out the app. The current administration has already expressed its position on federal service delivery improvements [1], and frankly, USDS/18F has a track record of crushing it. This would go hand in hand with the USPS slowly making its way to being a meatspace trust anchor/identity proofing|credentialing provider [2]; problem with your email account? Get help in person at your local USPS.
Which government? US federal? State? County? City?
Part of the fun of the internet is that it is really international so I'd expect it to interoperate with Canada, France, Ukraine, etc. What if people from Barcelona want a Catalan e-mail address?
Ideally I'd like a global government to do it, although I don't like the current structure of the UN. I think what I see in the web3 space is possibly a yearning for a global registry, whether for identity of humans or businesses (DAOs), or even a global currency.
So for me, I'd like a global entity that more closely resembles the global nature of the internet, combined with a more globally representative structure.
I see where you're coming from, but it would also be an extreme liability. For example, imagine companies demanding that you sign in with only your uniquely-identifiable government email address so they can better track you. The necessary technical qualities to resist such attacks (eg allowing any user to generate a bunch of additional nyms) will never be considered in make-everything-legible government thinking. Social security numbers and drivers license numbers are already abused by surveillance companies, and so the last thing we need is another government-issued identifier without something like a US GDPR to prevent its abuse.
Also, the kind of long term stable online identity you're desiring can already be had by registering your own domain name.
Registering your own domain name still puts you at the mercy of public opinion in the form of registrars. Sure, becoming persona-non-grata to registrars doesn't happen every day but it does happen and it certainly isn't becoming less common.
I think government email services would be very risky, both in a legal sense and in a cybersecurity sense.
You can have government authentication without too many issues, though. The EU has a digital login system that's federated (EIDAS) and your government should never have to retract your ID unless you die. Some implementations use physical smart cards to sign in, others use usernames/passwords+apps/2fa. The system is currently intended only for official government business, but under the hood most systems use something close to OAuth2 so it should be possible to modify the system to allow generic logins.
The downside of such a system is that the using the same authentication for government grants and your social media makes it very easy to phish people. This increases the security requirements of the system significantly, which in turn make the system very annoying to use. The authentication mechanism also allows the government to track exactly what services you use and when.
Personally, I'd much rather see OpenID return. The ability to pick an identity provider and log in would allow for multiple accounts and the ability to control your own data.
Alternatively, wide-spread use of something like https://irma.app/ would be even better. This allows for decentralised identity that you can take with you, just like a regular ID. It also allows for exposing only scoped data (ie "is this person of drinking age" rather than "what's this person's birthdate").
It seems like their proposed solution is decentralized blockchain identities which in theory should be impossible to revoke. I don't know what that would look like in practice, but if every web service needs to support it then I doubt it will ever happen.
Very well said. I’ll add that it creates a single point of failure for being hacked as well. Plus it drives me crazy that 1Password doesn’t support it.
It would be more like receiving mail from a service than sending mail to a service which isn't a problem. There isn't really a spam component.
I show up to bigsaas.com. Big SaaS asks me to identify myself. I submit that I am rob@robmccoll.com optionally with a proof of that claim.
If I have provided a proof (think attestation along the lines of "The holder of this certificate is rob@robmccoll.com requesting to use bigsaas.com at time T using signing key Kpub") it will be signed under a key chain whose root can be obtained from a specifically formatted DNS TXT record under robmccoll.com (similar to DNSSEC or DKIM - maybe it gets a custom record type). bigsaas.com can fetch said key and verify the signature.
If I haven't provided a proof, bigsaas.com looks up a DNS record under robmccoll.com that points to which server provides this authentication service for the domain and does something like an OAUTH or SAML type of exchange but without the need for point-to-point pre-configuration. It beings this exchange with a signed request "This is bigsaas.com attempting to authenticate rob@robmccoll.com at time T" which is signed under the key in its own DNS records (or at least a chain that can be verified under that key).
Either way, the service doesn't have to have some pre-seeded trust beyond trusting DNS (and this should probably insist on DNSSEC and TLS-only for whatever interactive authentication gets layered on top)
Edit: I guess this is kind of OpenID, but to be honest, OpenID looks way more complex than necessary.
> The problem it hopes to solve is we don't trust Google or Facebook. I have this problem as well. I trust them not to fuck me. Mostly because I trust them not to notice me at all. But I don't trust them to live forever. I'm some bacteria in the gut of a whale. What will I do when the plankton stop flowing?
My fear is actually the opposite here. I don't think Google will cease to exist overnight, or something without a ton of warning signs that others using them will have to adapt to. But being squashed like a fly because of some malfunctioning ML algorithm with no recourse? That makes it to HN a few times a month, so I have to assume happens more frequently once you count people without the following to independently draw attention to it.
Yeah, I guess it is sort of a personal ad to put all those account links in the middle of my post. I meant it as an illustration of how fractured our online identities are.
Also as a joke about how noisy the internet is: Almost every website has 2-6 such icons, usually twitter, github, rss and email. I don't know how most people use the internet but my eyes just glaze over them (Which is sort of sad actually, when I read a blog I like I really should subscribe to the author directly somehow). I think expanding it out to more than 20 is funny, but I may have an idiosyncratic sense of humor here.
I see trucks on the freeway where I can't really tell what company it is driving for except sometimes there's a large panel sticker that says LIKE US ON FACEBOOK and as far as I can tell it's a Facebook truck. I think some people are losing the concept that everyone's advertising for Facebook instead of themselves.
I just want a real federated identity protocol with public key distribution and access delegation through signed certificates. Do discoverability through DNS. Tie identities to providers via their hostnames (user@provider seems to work fine for email).
> I'm not really anonymous anywhere because I'm lazy and milquetoast
Worth having a few anon accounts so you can experiment with different concepts without having your real identity cancelled. I'm not talking about malicious hate speech or harassment, just toying with a new online avatar and toying with new concepts. I've tried tweeting under my real identity, and usually end up regretting the post a week later. With an anon account you can leave the tweet up because no-one knows it's you.
Online "privacy" is a bad meme. Our digital society would be 100x healthier if you could trust that the persona on the other side is real and can be held accountable for bad behavior.
Advertisers and crooks already have under-the-counter access to our private data (via Equifax, etc.), but we don't get to benefit from it.
Let's get over our misguided ideological fetishization of "privacy" that we adopted when we read a fantasy book by George Orwell in middle school and implement a proper identity scheme for the 21st century.
> Facebook's real name policy, Youtube's real name policy, Twitter's blue checks..
These are all opt-in verification mechanisms, likely with the main purpose of protecting the rights of influencers and other content creators so that the platforms keep making money off them.
I'm talking about a single federal government-run identity provider that obsoletes all the above schemes. That's what we need and where we are headed, whether we like it or not. It is absurd that we still rely on SSNs and home address verification in 2022.
You mean like the IRS hiring a private shady biometric company to try and match you to a grainy drivers license photo before they canceled the whole thing due to the bureacratic nightmare hellscape it spawned?
That sounds awful. No, it would need to be an in-house thing, a proper government agency, complete with their own datacenters and software teams across the US.
Probably impossible with the current "print $$$ and outsource" culture in Washington.
The last people you want running an identity provider are the federal government. You don't want to give them the power to make you an un-person if you some day are not supporting the correct party or agenda.
Sometimes you need to know who a person is, most of the time you don't.
I don't need to know who you are to sell you a cup of coffee, or really anything you are paying for in full.
If I'm lending you money, I need to know to my satisfaction that you are who you say you are, in case I need to find you later to collect. That's my problem and your problem to work out. The Federal Government has no role here.
> You don't want to give them the power to make you an un-person if you some day are not supporting the correct party or agenda.
Too bad. They already have this power -- and more, they can un-life you -- and always have, and always will until they get replaced in the next revolution.
It is just SSNs but updated for the 21st century. Your dox would probably be a lot safer, because they would be held by 1 accountable public agency rather than 10 unaccountable private for-profit corporations.
With hatred and violence people are willing to post under their real name on Facebook, I sincerely doubt that enforcing a real name policy like the Chinese government is trying to enforce will make a difference.
People are more toxic when they're anonymous, but de-anonymizing people isn't a solution to toxicity.
I don't want to give up the ability for gay/trans/whatever people in oppressive societies to express themselves because someone was mean to me on Twitter once.
There are platforms out there that will enforce real name policies. You're free to leave communities that don't and enjoy a theoretical community where people are held up to standards if these policies work. You could even add your information to your account description if you don't care about your data and want to shape the community to your liking by setting an example for others. Just don't take down the rest of the internet with you because you don't care. I'm not in the Equifax leak or in any other American credit rating provider leak, so leave me out of your weird post-privacy world, thank you very much.
There are plenty of platforms that have real-name policies, such as facebook and nextdoor. Are these 100x healthier then platforms like HN, which allow pseudonyms? I don't know. At any rate, the important thing is that people get to choose.
If you want to interact with people trusting that they can be held accountable for bad behaviour, you can do that on these platforms.
If you want to interact with people free to express themselves without fear of getting "held accountable" for "bad behaviour" - knowing that there is some disagreement on what constitutes bad behaviour and not trusting everyone who's able to "hold people accountable" - to get them fired - to share your views on this, you can do that on platforms like this one.
Privacy and anonymity are essential for people who want to speak out against powerful or potentially violent groups or individuals. Anonymity is really the only way to protect freedom of speech imo. Privacy is also extremely important for any sensitive info you want to selectively share. Tools, services, and platforms need to take that into consideration. It's not just a meme.
>Advertisers and crooks already have under-the-counter access to our private data (via Equifax, etc.), but we don't get to benefit from it.
We shouldn't throw away the concept just because we haven't executed it well.
The actual problem is our obsession with free stuff and knowing who to trust. Trust should be with family, friends, and the people you know. Then the trust reduces the further someone is from you in the socal graph. That's why I'm big on Web of Trust. You can't (and shouldn't) make people trust those you think they ought to trust. People should choose for themselves.
Posting comments and content should have skin in the game, you're absolutely right about that. Being able to post things consequence free is what leads to more trolling and spam. It shouldn't be contingent on your identity though. Micropayments are a great for this. If you had to stake a tiny amount of money for each comment, I think it would reduce spam and trolling.
> Privacy and anonymity are essential for people who want to speak out against powerful or potentially violent groups or individuals.
This is the popular refrain, but really its about providing tools for HUMINT, signature reduction, and other covert operations of the US government and its proxies.
For instance, this is why the Tor Project is funded primarily by DARPA and the State/Defense Depts -- objectively speaking among the most violent organizations on the planet -- rather than actual dissident groups.
This is a solved problem, but nobody realizes it yet.
We already have a decentralized system of unique identities, and we have since the '80s. It's called the domain name system. They're human-readable, they have strong guarantees about being able to own and control them (so long as you pay a nominal fee), and they have a rock-solid infrastructure behind them that backs everything from Google to the US government to your friend's blog. We even have a (somewhat less convincing) way of verifying that the server you're talking to really is the one that your DNS record points to.
What still needs work is getting from an identity system, which tells you which server to look at for a given name, to a system of authentication for specific tasks. Given that someone controls a given domain name, how can they use that to log in to a service or post messages that are verifiably theirs?
If you're willing to run a server for it, OpenID works. If you only want to send email, DKIM has you covered. The w3c's decentralized identity specs are really cool, and I think did:web [1] has the potential to bring us to a world where you can buy a domain, cname it to some host, and upload your public keys there so that you can sign anything and login anywhere. Making this easy for non-technical users will be important, but I think it can be done. The fact that ICANN policy requires companies to allow you to migrate your domain guarantees that you can sign up with some fancy startup that will manage everything for you and keep your identity if you want to move somewhere else.
There are many other related topics on indieweb.org and other related websites. What we need more of includes systems that make implementing such methods and protocols super easy.
This solves part of the problem, but not the whole problem. An important aspect of any general identity system is the ability to associate an ID with a specific human for cases where you need to hold the human accountable. DNS relies on banks and governments for this feature.
It's a mixed picture, but to me the fact that irangov.ir still resolves in spite of an all-out siege by OFAC tells me that it's decentralized enough for almost anyone.
The Oracle problem is defined as the inability for blockchain solutions to validate their data with the real world. In this case it’s that any specific ID is actually a singular human, and not a bot, sock puppet, etc. Note this means any one person cannot get two or more IDs.
How would you implement a system that issued such IDs?
It might be possible to have a multi-layer identity system, where at the top level we have face matching and social vouching to implement uniqueness, and then we add a layer of zero knowledge proofs to allow people to mint new identities for themselves which can share reputation or attestations.
58 comments
[ 3.2 ms ] story [ 80.6 ms ] thread* we have fractured identity online
* FB/Google might be going away
* Web3 may protect against that
* Large Language Models might cause the death of the internet.
I'm not really sure what to take away from the post. I found it confusing.
Thanks for reading though!
Because these platforms are gamified, they are worth gaming and gave rise to lots of fake automated content. Go search "best tablet" or "best lawnmower" for several hundreds of thousands of examples. Or look at Twitter.
Instead of any of these achieving a plaza of conversations, people are farther apart than they were to begin with.
There's been too many horror stories of people being locked out of their Google or Facebook accounts by Google and Facebook, even for the most minor of infractions, and that person immediately also losing access to to all the other services they used 'Sign in with...'
Until this problem is solved, I will never switch to a single online Identity for access, and I certainly will never use my Google or Facebook account to register with third party services.
120 year email + static hosting tied to library card.
Start with one branch or small system, scale from there.
Could reasonably be managed per branch with central support orgs. I guess most of them have a server room or colo or similar with a couple U free.
[1] https://www.whitehouse.gov/briefing-room/presidential-action...
[2] https://www.cfr.org/report/solving-identity-protection-post-...
Whatever email server/system the government created would be so bad no one one would use it.
Part of the fun of the internet is that it is really international so I'd expect it to interoperate with Canada, France, Ukraine, etc. What if people from Barcelona want a Catalan e-mail address?
So for me, I'd like a global entity that more closely resembles the global nature of the internet, combined with a more globally representative structure.
Also, the kind of long term stable online identity you're desiring can already be had by registering your own domain name.
Why in the world do you want them to host your email?
You can have government authentication without too many issues, though. The EU has a digital login system that's federated (EIDAS) and your government should never have to retract your ID unless you die. Some implementations use physical smart cards to sign in, others use usernames/passwords+apps/2fa. The system is currently intended only for official government business, but under the hood most systems use something close to OAuth2 so it should be possible to modify the system to allow generic logins.
The downside of such a system is that the using the same authentication for government grants and your social media makes it very easy to phish people. This increases the security requirements of the system significantly, which in turn make the system very annoying to use. The authentication mechanism also allows the government to track exactly what services you use and when.
Personally, I'd much rather see OpenID return. The ability to pick an identity provider and log in would allow for multiple accounts and the ability to control your own data.
Alternatively, wide-spread use of something like https://irma.app/ would be even better. This allows for decentralised identity that you can take with you, just like a regular ID. It also allows for exposing only scoped data (ie "is this person of drinking age" rather than "what's this person's birthdate").
Anyone know if Bitwarden does?
I show up to bigsaas.com. Big SaaS asks me to identify myself. I submit that I am rob@robmccoll.com optionally with a proof of that claim.
If I have provided a proof (think attestation along the lines of "The holder of this certificate is rob@robmccoll.com requesting to use bigsaas.com at time T using signing key Kpub") it will be signed under a key chain whose root can be obtained from a specifically formatted DNS TXT record under robmccoll.com (similar to DNSSEC or DKIM - maybe it gets a custom record type). bigsaas.com can fetch said key and verify the signature.
If I haven't provided a proof, bigsaas.com looks up a DNS record under robmccoll.com that points to which server provides this authentication service for the domain and does something like an OAUTH or SAML type of exchange but without the need for point-to-point pre-configuration. It beings this exchange with a signed request "This is bigsaas.com attempting to authenticate rob@robmccoll.com at time T" which is signed under the key in its own DNS records (or at least a chain that can be verified under that key).
Either way, the service doesn't have to have some pre-seeded trust beyond trusting DNS (and this should probably insist on DNSSEC and TLS-only for whatever interactive authentication gets layered on top)
Edit: I guess this is kind of OpenID, but to be honest, OpenID looks way more complex than necessary.
My fear is actually the opposite here. I don't think Google will cease to exist overnight, or something without a ton of warning signs that others using them will have to adapt to. But being squashed like a fly because of some malfunctioning ML algorithm with no recourse? That makes it to HN a few times a month, so I have to assume happens more frequently once you count people without the following to independently draw attention to it.
Also as a joke about how noisy the internet is: Almost every website has 2-6 such icons, usually twitter, github, rss and email. I don't know how most people use the internet but my eyes just glaze over them (Which is sort of sad actually, when I read a blog I like I really should subscribe to the author directly somehow). I think expanding it out to more than 20 is funny, but I may have an idiosyncratic sense of humor here.
Worth having a few anon accounts so you can experiment with different concepts without having your real identity cancelled. I'm not talking about malicious hate speech or harassment, just toying with a new online avatar and toying with new concepts. I've tried tweeting under my real identity, and usually end up regretting the post a week later. With an anon account you can leave the tweet up because no-one knows it's you.
Advertisers and crooks already have under-the-counter access to our private data (via Equifax, etc.), but we don't get to benefit from it.
Let's get over our misguided ideological fetishization of "privacy" that we adopted when we read a fantasy book by George Orwell in middle school and implement a proper identity scheme for the 21st century.
This hasn't worked. Not once. Full stop.
IRC continues to be the best place to have real conversations. Discord is pretty great too and even has built in alt support.
These are all opt-in verification mechanisms, likely with the main purpose of protecting the rights of influencers and other content creators so that the platforms keep making money off them.
I'm talking about a single federal government-run identity provider that obsoletes all the above schemes. That's what we need and where we are headed, whether we like it or not. It is absurd that we still rely on SSNs and home address verification in 2022.
Probably impossible with the current "print $$$ and outsource" culture in Washington.
Sometimes you need to know who a person is, most of the time you don't.
I don't need to know who you are to sell you a cup of coffee, or really anything you are paying for in full.
If I'm lending you money, I need to know to my satisfaction that you are who you say you are, in case I need to find you later to collect. That's my problem and your problem to work out. The Federal Government has no role here.
Too bad. They already have this power -- and more, they can un-life you -- and always have, and always will until they get replaced in the next revolution.
People are more toxic when they're anonymous, but de-anonymizing people isn't a solution to toxicity.
I don't want to give up the ability for gay/trans/whatever people in oppressive societies to express themselves because someone was mean to me on Twitter once.
There are platforms out there that will enforce real name policies. You're free to leave communities that don't and enjoy a theoretical community where people are held up to standards if these policies work. You could even add your information to your account description if you don't care about your data and want to shape the community to your liking by setting an example for others. Just don't take down the rest of the internet with you because you don't care. I'm not in the Equifax leak or in any other American credit rating provider leak, so leave me out of your weird post-privacy world, thank you very much.
If you want to interact with people trusting that they can be held accountable for bad behaviour, you can do that on these platforms.
If you want to interact with people free to express themselves without fear of getting "held accountable" for "bad behaviour" - knowing that there is some disagreement on what constitutes bad behaviour and not trusting everyone who's able to "hold people accountable" - to get them fired - to share your views on this, you can do that on platforms like this one.
>Advertisers and crooks already have under-the-counter access to our private data (via Equifax, etc.), but we don't get to benefit from it.
We shouldn't throw away the concept just because we haven't executed it well.
The actual problem is our obsession with free stuff and knowing who to trust. Trust should be with family, friends, and the people you know. Then the trust reduces the further someone is from you in the socal graph. That's why I'm big on Web of Trust. You can't (and shouldn't) make people trust those you think they ought to trust. People should choose for themselves.
Posting comments and content should have skin in the game, you're absolutely right about that. Being able to post things consequence free is what leads to more trolling and spam. It shouldn't be contingent on your identity though. Micropayments are a great for this. If you had to stake a tiny amount of money for each comment, I think it would reduce spam and trolling.
This is the popular refrain, but really its about providing tools for HUMINT, signature reduction, and other covert operations of the US government and its proxies.
For instance, this is why the Tor Project is funded primarily by DARPA and the State/Defense Depts -- objectively speaking among the most violent organizations on the planet -- rather than actual dissident groups.
I disagree with this unpopular idea so I'm going to mark it as spam...
We already have a decentralized system of unique identities, and we have since the '80s. It's called the domain name system. They're human-readable, they have strong guarantees about being able to own and control them (so long as you pay a nominal fee), and they have a rock-solid infrastructure behind them that backs everything from Google to the US government to your friend's blog. We even have a (somewhat less convincing) way of verifying that the server you're talking to really is the one that your DNS record points to.
What still needs work is getting from an identity system, which tells you which server to look at for a given name, to a system of authentication for specific tasks. Given that someone controls a given domain name, how can they use that to log in to a service or post messages that are verifiably theirs?
If you're willing to run a server for it, OpenID works. If you only want to send email, DKIM has you covered. The w3c's decentralized identity specs are really cool, and I think did:web [1] has the potential to bring us to a world where you can buy a domain, cname it to some host, and upload your public keys there so that you can sign anything and login anywhere. Making this easy for non-technical users will be important, but I think it can be done. The fact that ICANN policy requires companies to allow you to migrate your domain guarantees that you can sign up with some fancy startup that will manage everything for you and keep your identity if you want to move somewhere else.
[1] https://w3c-ccg.github.io/did-method-web/
* For identity by way of DNS, such as via domain name, see: https://indieweb.org/personal-domain
* For logging into systems by authentication based on controlled domain name, see: https://indieweb.org/Web_Authentication
There are many other related topics on indieweb.org and other related websites. What we need more of includes systems that make implementing such methods and protocols super easy.
I see three requirements:
1. Uniqueness: Exactly one ID assigned to each human
2. Irrevocability: nobody may invalidate the ID except it’s owner in the case of replacement
3. Anonymity: The owner of the ID can use it to prove their singular humanity without revealing anything else about themselves.
#3 is relatively easy to achieve provided #1 is solved. But #1 requires an Oracle [0] and cannot be decentralized to ensure #2 is upheld.
The only practical solution we’ve found is identity assignment by nation states, eg. social security numbers.
0: https://encyclopedia.pub/entry/2959
Can you explain what the problem is exactly? The infinitely small possibility that the same ID is generated for two different people?
How would you implement a system that issued such IDs?