You got it right. Haven's built-in Blossom server also comes from Khatru, and there's no built-in client or admin interface for now. A built-in client likely won't happen, as we're trying to avoid doing too much within Haven itself... But nothing stops you from running your own clients ad well. I'm using a private, contenerised version of Nosotros in my own deployment. It was pretty easy to set up. I assume the Blossom client can also be deployed easily.
An admin interface might happen one day, but it's pretty far down the list of priorities, unless someone else jumps in and builds it. Contributions are always welcome.
## My Current Focus (in no particular order)
- Proper binary / PGP-signed releases
- Increasing the number of supported cloud services and properly documenting their settings
- Uploading and serving Blossom blobs from external storage like S3, CDNs, and maybe IPFS, as bandwidth has become quite a bottleneck for my own deployment
- Optional payment relays with `npub` whitelisting/blacklisting for NWC and Cashu payments
- Enhancing WoT support with a PGP-like ownertrust model (i.e., so I can follow someone without giving their followers write access to my inbox relay)
- Restoring event backups
- MLS support
- A CI pipeline
- Proper tests (or, well, any tests at all, really)
- Proper observability (e.g., OpenTelemetry support that can plug into your favourite dashboard/monitoring tools)
- Moving Haven to a SemVer model and likely releasing **Haven 2.0** (we need to introduce some non backwards compatible changes on Haven)
- Better, well-divided / linked / indexed documentation
- Postgres support
- For my own Docker Distro:
- Onion Service support + Podman Quadlet support so users have an "easy" way to deploy Haven, Nginx, Blossom, Tor, etc., with all the **Systemd goodness** (automatic restarts, logs, etc.)
- Long-shot experiment: Building a Haven-on-Linux-on-Android distro with DDNS + tunnelling support so users can take Haven on the go (I'm still not sure about this one, since Citrine exists and it might be easier to work on it and keep Haven simpler).
On top of my wishlist above there's also maintenance tasks that need some love as well.
- Separate the Blossom database
- Tackle concurrency and reliability issues, particularly around import and Blastr
- Deprecate and remove older or unused libraries and code paths
- Stay in sync with upstream Khatru and Blossom innovation
nostr:nprofile1qqsw9n8heusyq0el9f99tveg7r0rhcu9tznatuekxt764m78ymqu36cpr3mhxue69uhhyetvv9ujucnfw33k76twwpshy6ewvdhk6tcpzdmhxue69uhhwmm59e6hg7r09ehkuef0qy2hwumn8ghj7un9d3shjtn4w3ux7tn0dejj7ne6u4e has has his own agenda and ideas as well, so there’s plenty of great stuff on Haven's backlog.
Just from my own list above, given my current rate of contribution, that’s easily a couple of years' worth of development, assuming even half of it is feasible to begin with :)
