daniel 🫠

we may not actually need to do a working-group-forming bof if we can find consensus on the mailing list + IESG approval

if we end up doing a working-group-forming bof we would likely do an interim bof which is done remotely between meetings

oh not quite!

the charter is currently in IESG review: datatracker.ietf.org/doc/charter-...

it has a few blocking comments that we're still working through over here: github.com/ietf-artarea...

feeling cautiously optimistic though!

one of our competitors is now a major US military contractor with a global constellation of space lasers, whose leader chums around with heads of state, the international financial elite, and other members of a sprawling pedophilia conspiracy.

so cool! 🚀📈✨

"brew install goat, goat firehose" is an activation phrase for sleeper agents

What's Next at Bluesky - Bluesky As we head into 2026, we're entering a new phase for the Bluesky app. Last year was about scaling through rapid growth and getting the fundamentals in place. This year is about leaning into what's wor...

2026 is the year Bluesky and the Atmosphere really come alive

here's what's next

bsky.social/about/blog/0...

Tap: fix nil pointer err on prevData by dholms · Pull Request #1287 · bluesky-social/indigo Prev data can be nil in the case of legacy (non sync1.1) commits, causing a nil pointer error.

no you're right this can happen for events coming from old PDSs that don't do proper sync1.1 stuff

fixed it here: github.com/bluesky-soci...

atprotoans using tap, important update here!

if you build from source from indigo:
git pull && git checkout tap-v0.1.4
go build -o tap ./cmd/tap
<redeploy>

if you run with docker:
docker pull ghcr.io/bluesky-social/indigo/tap:latest
<redeploy> (ensure to replace container!)

I second this. If you are running a tap instance if you could please update to this. If you have heard me talking about @selfhosted.social's high CPU usage, it was this.

There was a bug where when a tap instance subscribed to a repo and the repo changed in any way, tap would download the whole repo

just published a fix for a bug in Tap that was causing over-fetching of repos when filtering based on collection

if you're using Tap, strongly encourage upgrading (to v0.1.4)!

you can upgrade to the latest by pulling ghcr.io/bluesky-social/indigo/tap:latest

Another perk of running your own appview, PDS and relay: When Bluesky goes down for others, it doesn't go down for you.

The killer app of atproto is interop

Tap: fix repo commit metadata tracking by dholms · Pull Request #1280 · bluesky-social/indigo Fixes a few issues in Tap. The most important is that we were not updating prev data &amp; rev when we didn&#39;t need to process any ops from a commit. This would cause commits to appear as disjun...

okay actually I think this was a bug 🫠
github.com/bluesky-soci...

about to be afk so i'm not sure i'll be able to get this out, but i'll try & land it soon

it shouldn't be doing that on every event unless the commit events themselves are invalid. just checking, are you running the Bluesky PDS distribution?

Tap: zombie websocket · Issue #1279 · bluesky-social/indigo Several reports of a zombie websocket connection to the relay. For instance: https://bsky.app/profile/ngerakines.me/post/3md4fhqz5i22f My expectation is we need some heartbeat ping/pong logic in he...

thanks for flagging

i opened an issue & will try to take a look soon: github.com/bluesky-soci...

YouTube video by number 0 40x less overhead! Rateless Invertible Bloom Filters - Part 3 of 3

also check out @b5.bsky.social's great video series on these: www.youtube.com/watch?v=B943...

i should write a leaflet on this

thanks for the issues 🙏

i'll try to take a look soon

remixable social

The Open Web means several things to different people, depending on context, but recently discussions have focused on the Web's Openness in terms of access to information -- how easy it is to publish and obtain information without barriers there. https://www.mnot.net/blog/2026/01/20/open_web

Let’s Build an Atmospheric Web The Atmospheric Web shows we don't have to accept the false choice of ownership vs. distribution

This is the most succinct version of a post I've been trying to write for two months. When I stop for just a sec to think about the possibilities of The Atmospheric web, I'm a little overwhelmed by how exciting it is. Come build it with us!

is ̶t̶hat ̶a̶n̶y̶t̶h̶i̶n̶g̶

nice

p̶r̶i̶v̶a̶t̶e̶ permissioned data

i'm actually pretty happy with this piece

i agree it's a little unintuitive at first but i like how it sorta embeds the requirements of the tree structure in the schema (even tho invalid nodes are still possible to create)

also arrays of tagged objects are kinda awkward in some languages (like Go)

How to invent MST from first principles:

1. Start with a content-addressed data store i.e. a k/v store where values are immutable data and keys are hashes of the values.

2. Build a mutable k/v store on top of that.

3. Make it have nice sync properties (CRDT w/ structural {unicity, sharing})

No there’s no way to do that yet. My recommendation is to set it up for a small set of dids in order to limit the amount of data that comes in

ah interesting I’ll have to take a look at the collectionsdir service

hey @pfrazee.com check out this fun word I heard

oh yeah starting to feel like the atmosyear