Want to identify large warehouses that are vulnerable to becoming ICE detention sites? Here's a quick, proactive guide anyone can use. No special skills or access needed. 👇🧵
Oh, this is already a fascinating dataset. Behavior patterns from early adopters -- especially when they don't just set up accounts + bounce away -- are great early indicators of what might be sticky down the road
Kudos for giving us a sneak peek!
PS: Speaking of cohorts, are individual mushrooms of various vintages -- new ones get added over time as older ones fill up -- roughly comparable in scale + activity to this one?
If so, then maybe we have useful data for some natural experiments
When AI makes generation of all kinds of content free...
... the review burden on folks with domain expertise increases immeasurably.
You're absolutely right—that was not a command post, that was an elementary school. Those were not military leaders—they were schoolchildren. I said the opposite, and that's on me.
ah, anthropic is not beating the Iranian Murder Bot allegations
This particular demographic skews towards early adopters, so you can't necessarily generalize very far
But given that we're seeing evidence of how far people are getting through the funnel to explore each offering, it'd be interesting to see a similar analysis for broader mushroom cohorts
sh.tangled.actor.profile,1390 sh.tangled.repo,826 sh.tangled.publicKey,755 sh.tangled.feed.star,650 sh.tangled.graph.follow,617 site.standard.document,1042 site.standard.publication,425 computer.aesthetic.mood,982 computer.aesthetic.painting,584 net.anisota.beta.game.session,717 net.anisota.beta.game.log,717 net.anisota.beta.game.progress,321 net.anisota.harvest.minigame,248 place.stream.chat.profile,412 place.stream.chat.message,252
space.roomy.space.personal,622 so.sprk.actor.profile,369 blue.flashes.actor.profile,285 net.wafrn.feed.bite,281 social.pinksky.app.preference,254
Now zero in on services which appear more than once on that list:
tangled = 5
site standard = 2
aesthetic computer = 2
anisota = 4
stream place = 2
Ranking collection names *within* each service gives you a quick peek at which features might be getting any initial traction
Excellent 🧵 explaining the pragmatic reasons why the leverage of a great leader rests on three pillars:
- choosing the right objectives,
- communicating them clearly, +
- trusting good folks to execute
That trio is an unmistakable cultural marker of how effective an organization will be
I said what I said.
Are war and tariffs the best ways to accomplish any goals in 2026? That's the approach the waning superpowers are taking. How's that working out for Russia and the USA? How's that war fighting and tough talk working out for us? Are we winning yet?
China is *removing* tariffs
🙂🙃
Oh they must *hate* Mamdani! The man started a fire.
This is how you win elections in New York now.👍🏿
Related: Lindsay Boylan took her power back, but is not content just to get back to even. No.
She wants more, so that she can make sure that what happened to her doesn't happen to others.
Respect
👨🏻💼Sir, we recommend a daytime strike. 9:40 am to be precise.
👴🏻I thought we strike at night? Aren't all the high priority targets in the bunker?
👨🏻💼High priority ones, yes sir. But this target moves around. We know where he will be at 9:40 am.
👴🏻How do we know?
👨🏻💼He... drops his daughter off at school
Online communities are always reduced to the same structure: Stewards as Landlords
Part 3 of Composable Trust wields ATProto to architect a solution: decomposing communities into 3 sovereigns—Members, Rosters, and Venues—each with full jurisdiction over their own domain, and none over each other’s.
This is a radically different idea than the advocacy I’ve seen for “data portability” and ”data sovereignty”. It’s a *restructuring* of power rather than small adjustments to existing structures.
atproto and local-first, both in their own ways, demonstrate what such a restructuring can look like.
Exactly. It's easy to underestimate how radical it is escape app-centric paradigms where *they* control access to your identity + your data
When your data + your identity exist independently, power shifts away from any/all apps *you* allow to access them
Check out my new blog post, 'ATProtocol Patterns: Record Elicitation'. In ATProtocol, only the user's client can write to their repo. But what if the AppView has information the client doesn't?
Insofar as OAuth effectively delegates PDS access, it should almost certainly be the latter
On that point, I suspect you're both violently agreeing 😉
If I understand @fry69.dev correctly, the question is about mental models. What domain should people associate their "bsky password" with:
- an application (.app) or
- an identity/network (.social)?
That password/domain binding encapsulates a ton of trust, so it's worth taking time to decide:
- which domain earns that trust,
- how clear that OAuth flow is, +
- what level of access is granted
... preferably in as few words as possible
( See the linked 🧵s for more on what that may require )
I've been saying for nearly a year that getting the UX right for bsky's OAuth deployment is a major inflection point for reinforcing users mental models
If we phrase things right, people will learn that this password UX controls access to my identity + slices of my atmosphere data (via my PDS)
Not sure I understand your ActivityPub analogy, though
The "store + forward" model of AP inboxes is a very different storage paradigm from the separation of concerns (PDS vs. indexes) enabled by the AT sync model
App designers used to the former often miss the radical potential of the latter 😉
Fair enough. Migrating to take advantage of atproto's decentralized storage model makes the most sense when it enables core features your users value
If you're better off delivering those from your centralized web infra, then AT-powered features will remain a bolt-on for a subset of your users
Naive question:
Is there a world in which it'd make sense to "enable" (not force) them to use atproto accounts you host on their behalf?
Would that create awkward UX flows for existing users who already have other atproto accounts they'd rather use for this?
Or do you prefer to not be that host?
Software Design is weird. It is undoubtedly the most impactful medium shaping the world today, yet even those of us working in it know very little of its history. We have no broadly-read books, no docu-series, no video essays. Most see the works of the past as obsolete rather than a rich heritage.
Sure, but feel free to use favicons if you find 'em (+ are capable of caching them)
As you point out, this is an easy, obvious convention to start supporting for anyone who wants to encourage branding support for PDS operators
Whether you or @pds.ls go first matters less
The friendly header is great, but for similar reasons, it's probably worth including the full PDS hostname for mushrooms:
Inkcap (US East)
Bluesky-hosted PDS
This account's data is stored on a Personal Data Server (PDS): inkcap.us-east.host.bsky.network. A PDS is ...
Yep. Suspension is a great example of the difference between:
- pruning from an appview's indexes (reflects app moderation policy) vs.
- deleting from my PDS (expresses my choice of who to follow)
Moderating @furryli.st always made me feel that something was missing from atproto. If we disappeared, so too would the community we built
This feeling’s finally crystallized into this proposal: Composable Trust
We created identities that survive platform failure. Lets do the same for communities!
Forest bathing, winter edition
Stop asking for permission. Build the thing you want to build.
