Matrix,  Net Number, EUDIW, 800-63-4, Chat Control, FOKS, Bitchat

1. Matrix - It's nice that Matrix is trying to address the well-known "flaws" (yes, I know, depends on the perspective you're viewing it from). MSC3414 (@ GitHub) Encrypted state events. MSC4014 (@ GitHub) - Pseudonymous Identities, yet it's currently being fiddled back and forth. It's interesting to see where it ends up and if it ever gets properly implemented. As currently proposed, this MSC doesn't go far enough. It fails to provide practical user pseudonymity because the mxid_mapping makes the user's true MXID known to all members of the room upon joining. This achieves metadata pseudonymization from the perspective of a homeserver, but it is not user or account pseudonymization from the perspective of other room participants. If I join three rooms pseudonymously, all of the users know it's still me. MSC4256 (@ GitHub) MLS mode Matrix - This is the most interesting one. I'm eagerly waiting for the results, but it might be a long long wait. I earlier labeled this a "nice to have hobby project from the Matrix team", something to tinker and burn countless hours on, and still be unsure if it ever works out.

2. Matrix-TTL - The automatic matrix content redaction and deletion script’s latest version is confirmed to work with the matrix room version 12 and its new id syntax: !Was3MnHgd-gTGfI-NIZ4Ly\_OCoD7wR0C7Uf0cL0hKhE without any new modifications to the code.

3. Net Number / 网号 (wǎng hào) & 网证 (wǎng zhèng) - China's pseudonymous online authentication system is an interesting concept. You register once with the government, get an alphanumeric token, and use that instead of your real identity on websites and services. I'm wondering if the EU EUDI Wallet will end up similar. Of course, it's pseudonymous only towards the websites and services, not the government - but that's still better than every platform holding your raw identity data.

4. EUDIW stands for EU Digital Identity Wallet (European Union Digital Identity Wallet). It's a digital identity solution designed to provide European citizens and businesses with a convenient and secure method to authenticate their identity for both public and private sector interactions. The wallet enables users to securely store and share their personal and financial information across EU borders, streamlining cross-border transactions and reducing paperwork. Key Features: Document Storage, Cross-Border Functionality, Secure Authentication. The system operates under the Architecture and Reference Framework (ARF). Privacy Context: As a digital identity system, EUDIW will handle highly sensitive personal data. Key considerations include data minimization principles, user control over information sharing, cryptographic security of stored credentials, and compliance with GDPR requirements.

5. NIST SP 800-63-4 (@ csrc.nist.gov) - Digital Identity Acceptance Statement (DIAS), IAL1 (pseudonymous, single / MFA), IAL2 (MFA, biometric (optional), device validation, Pairwise Pseudonymous Identifiers (PPIs)), IAL3 (non-exportable keys + phishing resistance, PPIs) - phishing resistance, verifier compromise protections, and federation integrity, privacy-by-design. Passwords, minimum 15 characters for single-factor use. Assertion provider requirements for FALs, IALs, and AALs.

6. The renewed push for "Chat Control" legislation has convinced me to renew everything related to dnskv.com (@ local archive, is shutdown), as its fundamental design is different. No useful metadata can be subpoenaed from the server, only the active data blobs. These blobs reveal neither source(s) nor destination(s), containing nothing more than a key, a value, and an expiration timestamp. This is a storage type which should be used with messaging services. Data is as stateless as it can be. A small reduction of metadata could be achieved by fixing or ceiling the TTL. Like to 24-hour window(s). Yet nothing prevents adding jitter to TTL when posting the blobs.

7. FOKS (@ foks.pub) and SOPS (@ getsops.io) - The first is a federated key management system and the second is a secrets management system with much lower overhead. As soon as Zoom bought Keybase I basically abandoned it. SOPS especially could be really handy in some cases. I appreciate the simple concept and love its execution. For small operations, FOKS is too complex. Yet for teams that need it daily, like any other tool, it can be highly automated and scripted to nicely mask the complexities. 

8. Bitchat (@ GitHub) - Ah, yet another mesh app. Which unfortunately are practically totally useless, except for the extremely minimal edge cases where the app is actually useful. Design is also extremely problematic (unreliable, inefficient), but that's nothing new in this context.

2026-06-28