Decentralized Web Nodes (DWNs) are self-hostable datastores containing personal cryptographic keys and any accompanying personal data.
It’s an open protocol that's being standardized by the Decentralized Identity Foundation (DIF).
Its purpose is to allow users to connect to a wide variety of apps via their DWN, so they can selectively allow apps to get data from their DWN or populate data to their DWN—data which they can then reuse across an ecosystem of compatible apps.
Verifiable Credentials and their presentations are envisioned to be used mainly as a means to exchange signed data between parties represented by their DWN—building online reputation (Web of Trust), without necessarily requiring externally issued identity Verifiable Credentials (passport, etc.).
DWNs are pushed by Jack Dorsey’s Block’s TBD team and the Web5 movement.
Advantages
- In the process of being standardized
- Low-level and flexible for a wide variety of apps, independently of the transport mode: reusable data, reusable identifier. Supports Verifiable Credentials, but it is not required. DWNs are represented by their DID, and every interaction is verifiable.
- Self-sovereign if self-hosted
- Avoid making users handle keys manually by default, hence providing a familiar user experience
Limitations
- Immature and lacks adoption for now, though we’re seeing promising developments
- Uncertain how DWNs would be interoperable between apps—use cases are still in exploratory mode
- Users, even many power users, generally don’t want to self-host their own server. So the vast majority of users would use a third-party service to host their DWN, essentially rendering it useless from a privacy perspective until the user self-hosts it.
- DWN would have been the perfect basis for AT Protocol’s Personal Data Store (PDS), but for some reason, to our knowledge, the two teams aren’t communicating for now.
Could it be useful for our requirements?
As DWNs are generally not self-hosted, DWNs are mainly meant to store non-privacy-critical Verifiable Credentials instead of identity Verifiable Credentials (e.g., digital passport), which should be stored in a mobile wallet. However, to achieve our sybil-resistant authentication system, we are mainly interested in these identity VCs.
We don’t have requirements about privacy as in encryption and confidentiality for data apart from credentials. We care more about anonymity, so the aspect of “owning your data” in DWNs seems unnecessary.
Like AT Protocol’s PDS, DWNs broadcast a unique pseudonymous identifier per user to the whole network. This can be an unwanted feature for anonymity's sake.
DWN can eventually help increase interoperability.
