Uniqueness


#1

Any inspirational ideas for verifying only the uniqueness of an identity?


Proposal: Developer Tools Roadmap Q1+Q2 2019
#2

the blockstack user ID is unique.

If you’re talking about voting and don’t want duplicate accounts, it costs money to have a *.id username, so you could use that. otherwise (eventually) you could create your own domain/subdomain with its own pricing rules to ensure it to be hard/pricey enough…

if you wanted to get it outside of blockstack then you could invent your own crypto coin or use BCH colored coins.

or, finally, you could use real-world identification and tie it to an account but that breaks anonymity so yeah.

if it’s something else you’ll need to provide more details


#3

Thanks for your answer.

I’m interested in a minimalistic model of an universal basic income altcoin. It should be distributed as new money and financed purely by an inflation rate of under 2 percent.

For that I need to prevent individuals from recieving more than one income or at least make it sufficiently hard. Charging money per ID is an interesting idea but has some issues… I hope for a similarly ingenious solution to double reception as proof of work is for double spending…


#4

I’m working on a proof of concept for a system of decentralized reputation using decentralized IDs (like blockstack ID, which I just started to incorporate into the project). I envision that it will have many applications with one of them being what I refer to as a “crowdscreened” group (cGroup), where membership in the group is curated in decentralized fashion by group members. One of the most interesting features of cGroups will be a mechanism to establish singular identity (i.e. no human being is allowed to have more than one decentralized ID in the cGroup). Once it is (hopefully) working it could definitely be used to choose recipients for an airdropped universal basic income (UBI) coin, and I have given a lot of thought to making that happen for two reasons: first, it would provide a motivation for people to set up their singular identity; and second, it would provide a motivation for people to try to cheat the system by setting up multiple identities, so that we can see if it works and tweak it to make it better. It will never be completely foolproof but the hope is that it will be good enough. I’m betting that “good enough” will actually end up being better than centralized government issued ID systems that we all use today.

The basic idea is that users will have the ability to create attestations (digitally signed messages) about other users on a variety of topics. The topics will be user-defined, but for the purpose of establishing singular identity (and nothing more), I would envision two main types of attestations: 1) identity: Alice attests that Bob’s decentralized ID is owned by a human being whose existence she has verified in the real world, and that the name on the decentralized ID (and photo, and whatever other characteristics we want to include) is the correct one; and 2) singularity: Alice attests that to the best of her knowledge, Bob does not have any duplicate identities on the list in question. Actually a third type of attestation would be that Alice could flag Bob’s ID as being a duplicate or fraudulent in some other fashion. For the initial proof of concept, I would envision that Bob would be expected to attach his real name to his Blockstack ID. That way, it would be easy for Alice to look at the list of applicants and know with some degree of confidence that Bob hasn’t applied with multiple IDs (unless he’s doing so with a fake name, which would make it difficult for him to obtain identity attestations from honest users).

Here’s how I envision a cGroup to come into existence.

  1. Anyone can create a new cGroup. This would entail nothing more than creating a json object that defines and describes characteristics of the group, including criteria for membership. For the sake of UBI, criteria could simply be that you are a (living) human being. Alternate cGroups might be additionally defined, for instance, by citizenship in a given country of interest (eg the Singular Identity + Icelandic Citizenship cGroup, which would have been useful to the creators of Auroracoin back in the day).
  2. Individuals would apply for membership in the cGroup. Application could require a fee but would not have to. A signed message requesting membership could be all that is needed.
  3. The cGroup creator would suggest default attestation-types but users could propose new ones too.
  4. The group creator would choose an algorithm to determine membership in the group. For the sake of UBI, the rules could be as follows:
    a. The creator of the group is a member, referred to as the “anchor” or “seed” user for the group.
    b. Anyone who receives attestations by the seed user is a member.
    c. Anyone who receives attestations by N or more members is also a member, unless they have been flagged as duplicate or otherwise fraudulent by some threshold M number of members.
    d. A minimum fee could be required if spam membership requests becomes a problem (and also to fund development/management of the group).

One of the interesting things is that anyone could run the above algorithm (or even a modified algorithm) using themselves (or someone else) as the seed user, to see if their membership list matches the list generated using the group creator as the seed user. Before doing the UBI coin airdrop, these rules could be modified to address unforeseen attack vectors that make themselves apparent. Perhaps one criterion before doing an airdrop could be that the membership list must the identical no matter which member is used as the seed user.

I’m building a proof of concept for the reputation system in general (not limited to the cGroup concept) on my website. It is still in early alpha stage but anyone interested can follow along at OnTheBlockchain.com. Although I envision the reputation system being applicable to any social network, I’m using OpenBazaar as a test case since I love the idea of OpenBazaar, but it’s never going to take off without a good decentralized reputation system.

I’m in the process right now of getting blockstack ID sign-in working on my site. I’ll probably be posting elsewhere in this forum to get some bugs figured out.


#5

@davidstrayhorn.id.bl this is really cool idea. It incorporate human interactions into confirmation of identity. A kind of virtual village with some basis in the non-tech world.

It seems like it could be gamed tho, can an identity be challenged?


#6

Identities that are suspected to be fraudulent will be able to be challenged through an attestation that acts as a red flag. An identity that is flagged can be reviewed by other members of the community, and an identity with enough flags cannot be a member of the group (unless the flags are removed).

It will be impossible to know ahead of time exactly what threat models will be problematic, in part because threats will evolve with time. This is one reason why it is important for community members to have the ability to create new types of attestations.