chrisb @boscolo.co · 28/09/2026Is it true that Anthropic had a wake for Opus 3 when they decommissioned it?static.klipy.comAwkward Monkey Looks AwayALT: Awkward Monkey Looks Away 000
chrisb @boscolo.co · 21/09/2026A flavor of AT Protocol where the PDS is addressed using a native P2P protocol instead of HTTP 4465
chrisb @boscolo.co · 21/09/2026Hearing @taylorlorenz.bsky.social say "𝕏 is better now than ever" is not an outlier. Seeing many return there. x.com/creatine_cyc...x.comatlas (@creatine_cycle) on X"i think X is better now than ever" - @TaylorLorenz "twitter was falling apart so i'm glad @elonmusk took over but i was like elon has never ran a social media product" "he banned me by the way. fi... 020
chrisb @boscolo.co · 20/09/2026TIL there are people that consider Mastodon is decentralized social media 😜 000
chrisb @boscolo.co · 18/09/2026Interesting take on the state of AI from someone that has earned my respect as both intelligent and holding to cypherpunk ideals. x.com/cass_on_mars...x.comCassie Heart (@cass_on_mars) on XBeyond the Horizon 010
chrisb @boscolo.co · 17/09/2026Good to see Signal decouple from phone numbers. Is a one time fee of $3 enough to prevent spammers from registering Signal accounts? 👀 122
chrisb @boscolo.co · 14/09/2026Assume any personal data you hand over about yourself will eventually be published to the world. Even the stuff your government requires banks to collect. 020
chrisb @boscolo.co · 14/09/2026Is it fair to say that this place 🦋 has stagnated? How do we bring more energy to decentralized experiments like atproto? 230
Reposted by chrisbLaurie Voss @seldo.com · 13/09/2026I have been writing this post about the economics of open source, on and off, for 13 years. I wasn't expecting to end up at "Microsoft should send corporations a huge bill" but nothing else we've tried has worked. seldo.com/posts/nobody-pays-for-ope…seldo.comNobody pays for open source. We can force them to. | Seldo.comThirty years of voluntary funding schemes haven't paid open source maintainers because asking doesn't scale. The real money already flows — to JFrog, Docker, Sonatype — at the registry layer. The fix: registries charge companies for supply (they already do), then route a slice automatically to the maintainers who make that supply worth having. 2016655
Reposted by chrisb3NUM @3num.co · 10/09/2026What's a phone number? The most widely used federated identity on Earth. Nearly 6 billion and counting. 121
chrisb @boscolo.co · 05/09/2026I'm curious how much existing laws can help in this new world of AI / robots 🤖 If I program a robot to break into your house and steal your fruit bowl, I assume I can be charged for breaking and entering. What if I prompted my autonomous robot to go pick me some fruit and it chose this tactic? 330
chrisb @boscolo.co · 04/09/2026off by one errors everywhere I look bsky38.combsky38.com38 Most Influential Bluesky Posters of 2026Vote for the 38 most influential Bluesky posters of 2026. 020
Reposted by chrisb3NUM @3num.co · 03/09/2026SMS and calls are HTTP Unencrypted. Logged. Interceptable. Plaintext on a carrier server like it’s 1998. We’re building the HTTPS: same number or use an atproto handle 🫨, same plan, E2EE private messenger, builtin spam prevention. The network carries it. Nobody opens it. Launching soon 👀 021
Reposted by chrisbAlex Wykoff @distributed.camp · 02/09/2026Hey #austin #atproto friends, Do we wanna do an informal hang this Friday? I’m thinking Cosmic Saltillo, breakfast tacos & coffee More than 3 likes and I’ll post up the rsvp. 343
chrisb @boscolo.co · 02/09/2026Identity verification is a cancer on society Instead of building better identity verification we should be building systems that don't require identity verification krebsonsecurity.com/2026/09/fbi-...krebsonsecurity.comFBI Probes Service Selling 153M+ Drivers Licenses – Krebs on Security 150
Reposted by chrisbekana stone @ekana.org · 14/08/20263 months later, but maybe this works lolblog.boscolo.coATSMS Group Chat: End-to-End Encryption Without the Lock-InPrivate group messaging that belongs to you, not to an app — built on the AT Protocol identity you already have. A proof of concept, published with its specifications, its code, and its open questions. 011
chrisb @boscolo.co · 18/08/2026designing a decentralized notification service that isn't also a surveillance tracking vector is the most elusive problem in tech thought popped in my head while pondering how mu.social and bsky.app were delivering a shared notification experience (clearing state in one app clears it in the other)mu.socialmu 050
Reposted by chrisbdman-os @dman-os.bsky.social · 17/08/2026I like how flexible the whole thing is. I love the size bucketing thing, def using that idea elsewhere. One-shots! Yes! 121
chrisb @boscolo.co · 17/08/2026blog.boscolo.coYour Private Messages Should Belong to You - Now They CanATSMS brings end-to-end encrypted messaging and calls to the AT Protocol identity you already have. 1113
chrisb @boscolo.co · 15/08/2026this post as taken over the part of my brain that used to be dedicated to thinking about the Roman Empire 041
chrisb @boscolo.co · 14/08/2026What if we just accepted that a top-level "decentralized namespace" doesn't follow the same rules as our centralized names? Is this too inconceivable for institutions like ICANN? 100
chrisb @boscolo.co · 12/08/2026Someone asked, why call it AT SMS, that sounds like SMS? blog.boscolo.co/3msvnllic5c2...blog.boscolo.coATSMS Group Chat: End-to-End Encryption Without the Lock-In - blog.boscolo.coPrivate group messaging that belongs to you, not to an app — built on the AT Protocol identity you already have. A proof of concept, published with its specifications, its code, and its open questions... 151
chrisb @boscolo.co · 12/08/2026blog.boscolo.coATSMS Group Chat: End-to-End Encryption Without the Lock-InPrivate group messaging that belongs to you, not to an app — built on the AT Protocol identity you already have. A proof of concept, published with its specifications, its code, and its open questions. 1237
Reposted by chrisbvrypan @vrypan.bsky.social · 11/08/2026Iron is cool. But how many decentralized projects using it don't rely on n0 relays? Honest question, and I'm interested to learn how you do it. 111
chrisb @boscolo.co · 06/08/2026seven steps into a massive rewrite and claude is asking to git stash and git reset... 😨🫣 020
chrisb @boscolo.co · 05/08/2026haven't felt this energized in a while 🔋 We're building a proof-of-concept E2EE library & demo on AT Protocol, using BeeKEM for DCGKA. There's a long road between a PoC and a foundation we can all depend on, but every journey has a first step🐾 bsky.app/profile/bosc... bsky.app/profile/expe... 1174
chrisb @boscolo.co · 05/08/2026Every time my AI uses perl to accomplish a task, I'm reminded that we are standing on the shoulders of giants. 💪 020
chrisb @boscolo.co · 05/08/2026i'm going to start talking like my AI One thing worth flagging: it will be obvious to anyone that spends their afternoons with Claude Code. 040
chrisb @boscolo.co · 04/08/2026i've noticed that Claude (fable/opus) has started referencing time a lot more, mentioned that such and such was stuff we did "this week" or "yesterday"... 000
chrisb @boscolo.co · 30/07/2026curious how the Cambrian explosion of “decentralized control” & “edge-user empowerment” happened completely independently of the IETF 🤔 010
chrisb @boscolo.co · 27/07/2026buzz.xyz is a cool idea. Nostr's simplicity and light-weight public key based IDs makes it fertile ground for experimentation like thisbuzz.xyzBuzz — Your people, your agents, your project — all in one place.Come test the early stages with us. 000
chrisb @boscolo.co · 23/07/2026chat, is it safe to assume when I see the term "load-bearing" the text was cooked by AI? 110
chrisb @boscolo.co · 22/07/2026"BeeKEM performs competitively with OpenMLS when connectivity is good. It also handles partitions that would take MLS offline, at a cost in efficiency." 000
chrisb @boscolo.co · 22/07/2026This is the most exciting thing I've seen on 🦋 since joining! No more centralized servers needed for encrypted group chat! 040
chrisb @boscolo.co · 22/07/2026my AI is getting a little snarky Starts off with: "This is a design question I can answer from the specs" Translation: rtfm 010