Skip to main content
Unlisted page
This page is unlisted. Search engines will not index it, and only users having a direct link can access it.

2026-04-A-Selection: Selection

Context​

The previous cycle remained open for quite a long time, and expanded in scope to carry out maintenance on the project.

Community management focus​

During that time, Discord made a mistake with its age-verification policy, which contributed to a renewed interest in Matrix from Discord communities.

This highlighted a number of known weaknesses with the current community management onboarding UX for both Draupnir and Matrix, that we already have un-executed plans to address.

The UX of Spaces management on Matrix as a whole remains poor. And while ban synchronisation is one of the basic features of Draupnir and other moderation tools, this is taken for granted by Matricians. This is not a Matrix-native feature and there is significant work required to set Draupnir and other tools which makes them inaccessible and discouraging to use. We believe the aforementioned goals are perfectly placed to improve this.

Given that the full roll-out of age-verification within discord has been delayed, and various governments are developing laws that mandate age-verification thing for all platforms, it seems likely that there will be a similar wave of interest in alternative infrastructure in the near future. And that strongly signals that this should be a priority for the project.

Additionally it presents an opportunity to bring the project to a new audience without making fundamental changes.

Zero Touch Deployment​

Cat is currently working on improvements to deploying Draupnir for those that own their infrastructure, and this work is being done under the title of Zero Touch Deployment. Essentially minimising the number of manual steps required to get Draupnir running. And this work will align nicely with the community management focus as it would mean we are focussing on Draupnir setup across the stack.

Options​

Option A: Increment community management onboarding​

Given that there is a need to improve onboarding that tied to a window of opportunity around the rollout of age-verification, and that we have contributor energy and interest on improving the onboarding of Draupnir generally, this seems like the clear priority for the moment.

As for what gets delivered in this direction, we would most likely first further improve the UX of adding Draupnir to rooms, see Planning #117. And this includes the delegation of permissions to Draupnir and using Matrix Spaces as a source of protected rooms. Essentially minimising the steps required to protect a community with Draupnir.

Option Z: None of the above​

This is an expression of no confidence in the direction. If you have further comments, please see consultation.

Discounted options​

Increment explicit agreement & approval / disapproval​

In the last cycle selection we hadn't had time to rework the plan for approval/disapproval. This has now been done as part of a rework of the NLnet NGI Zero Core goals.

Given that there is a limited window of opportunity to bring Draupnir to more users, we believe it would be a mistake to make this the focus of the project at this time. However, this remains a valuable objective to pursue, both to enhance distributed moderation and hardening the security of Draupnir. And will be revisited.

Room upgrade support​

The room upgrade support in Matrix remains a diabolical experience. And the need to upgrade rooms is not going away, as there is significant work on the horizon in two more units in the Matrix foundation's project Hydra trilogy. And Element has secured some kind of investment into P2P work, which will also no doubt require room upgrades.

Given that the UX for managing groups of rooms in Draupnir isn't perfect, and would have to be revisited by the community management focus, we don't think this can come before. However it remains very important as no other project plans to provide the level of support that we do here. See also the NLnet NGI Zero Core goal

Community participant onboarding​

We believe that the onboarding flow is key to preventing abuse within matrix rooms, see a discussion of the topic on Gnuxie's blog. An emerging vector of spam in the advent of the availability of LLMs comes from users engaging in unrelated conversations in Matrix rooms before engaging with users privately through direct messages or sending spam, and so there is even greater need.

This line of work would consolidate two previously considered routes, policy server support and the user familiarity metric (previously also called user participation metric). And is generally where we want to head to supersede the role most protections play in Draupnir right now.

If this was to become the focus, what we would try deliver first would most likely be communicating community norms and a code of conduct to new users to start with. It would be a pretty significant improvement to the current situation. And this would have to be worked out with or without policy server support initially.

Again this is important work but it doesn't match up against the window of opportunity around refugees from age-verification rollout on other platforms.

Feedback and discussion​

Voting is still in progress or has not yet occurred.

Outcome​

Voting is still in progress or has not yet occurred.