🚨 Save the Wayback machine! 9to5mac.com/2026/04/13/i...
Sign: www.fightforthefuture.org/news/2026-03... @fightforthefuture.org #Waybackmachine #PublicRecords #Archiving #Openaccess #Opendata
Anyone looking for a breath of fresh air in the Atmosphere should try @anisota.net by @dame.is and @hypha.coop's spores.garden. It's a wonderful calming experience.
Everyone else, of course, too. 🌿
thegoodwork.blog/posts/planti...
April 29th! Come talk about atproto! Featuring demos by @bad-example.com, @hypha.coop, and more.
atmo.rsvp/p/atproto.to...
Written media, from this bluesky post to the front page of the Toronto Star, will soon be consumed primarily via AI mediators. We believe this can be a time of great opportunity for media publishers, so we put together this short survey of the evolving landscape to help you navigate the moment.
In case you missed our talk at #AtmosphereConf, it's now up on YouTube!
doing my favorite thing of composing stuff together, and toying around with x402 and Privacy Pass to see how they can play together to introduce a sort of a blinding layer to x402
an intro to both, how they're used and a possible way to integrate them!
kobi.leaflet.pub/3m3pyyctda22i
our @makeworld.space will be organizing some ATProto meetups in toronto soon! Follow @atproto.to for more
fascinating lightning talk by @makeworld.space and @udit.bsky.social !! def want to try out pollen #atmosphereconf
Pollen in one picture:
Find similar images across the atmosphere
@hypha.coop is at #AtmosphereConf and launching Nectar, a new API that allows searching for similar images across the Atmosphere. Use it to track reposts, make image annotations, watch social trends, and more.
Learn more here: nectar.hypha.coop/
Blacksky trending topics are live.
excited to announce our new platform for content moderation and community management at scale.
this replaces ozone for us.
built by @rishkebab.blacksky.team hand in hand with our T&S team.
I didn't train a new model. I didn't merge weights. I didn't run a single step of gradient descent. What I did was much weirder: I took an existing 72-billion parameter model, duplicated a particular block of seven of its middle layers, and stitched the result back together. No weight was modified in the process. The model simply got extra copies of the layers it used for thinking.
this is crazy — bro topped an LLM benchmark without changing weights at all
he sliced an LLM wide open, in the middle, and duplicated a block of ~7 layers
circuits ftw!
dnhkng.github.io/posts/rys/
Thrilled to be working with @bad-example.com, whom you know from the excellent @microcosm.blue tools, on making the atproto network ever more resilient
atproto.com/blog/introdu...
Can't wait to see y'all at @atmosphereconf.org next week! I'll be there representing @hypha.coop alongside @udit.bsky.social
We recently crossed $6K/mo in recurring subscriptions 🎉
A nice milestone
If I was solo on Blacksky that would be enough for me to make a living full-time on just all of your direct support
But this year is about growing and providing for the core team who took/will take risks to work alongside me
Paul's article prompted a long one from me: What is the purpose of protocols?
connectedplaces.online/the-purpose-...
The AI revolution could yield a more equitable society or entrench existing inequities. As always, indigenous communities find themselves at the forefront of this battle. We loved this paper's proposals for indigenous data governance. It's more than royalties: tech & expertise must be transferred.
resync completion rate line chart: "success" jumps to an oscillating pattern peaking over 400 and dropping under 300, dampening over a few hours to around 360.sec not_found and retry series stay at/near zero.
"describeRepos avoided/sec" stacked bar chart with "slice_complete" (green, about 3/4 of combined height) and "tiny_repo" (yellow, stacked above) doing a slightly oscillating pattern from over 125 to just under 100.
table: upstream crawl: in progress repos crawled: 565,392 (+2812/s) resync queue depth: 60 (+2) discovery queue depth: 2,325 (+821) resyncs completed: 442,971 (+494/s) resync-buffered events: 3,467 (+4)
db disk bytes, a line chart and two stacked series total (dotted) line chart rises in a tight sawtooth pattern. there's some disruption after the first growth stage where total gets away a bit, rising kind of high. then it goes back to growth and total stays tightly just above i a small short sawtooth cycle on top of the stacked series. the first stacked series ("default", yellow) stays very close to half the stack on top of default is "index" (blue) both rise steeply, then taper to a shallow rise (with some gaps), then rise again at a steeper, almost-constant rate. the y-axis maxes out a bit over 5GB
got lightrail backfilling over 400 repos/sec for bits, though it's settling closer to 300/s so looking like over 24h for full-network backfill
(it's not actually disk- or cpu-bound, this is all limited by scheduling resyncs from enough different PDS hosts concurrently)
Hello from WordPress! 👋
This post was published directly to my PDS! In the best case it show up on bsky and standard.site compatible sites!
matthiaspfefferle.blog/2026/03/19/hello-from-wo...
![Unsubscribe Yesterday at 11:43PM
This message is from a mailing list. AN Andrew Newton (andy) [Atp] personnel assignments for ATP To: atp@ietf.org Hi all, Shuping Peng and Mallory Knodel have stepped forward to co-chair the ATP working group. Many thanks to them for taking on this role for the ATP wg and the ETF. With this assignment, the secretariat should be able to formally announce the creation of the ATP working group. Additionally, going forward I will be the responsible AD for this group. Many thanks to Orie for shepherding the creation of this group, and let us wish him well in his retirement from the IESG. -andy, ART AD Atp mailing list -- atp@ietf.org To unsubscribe send an email to atp-leave@ietf.org This message is from a mailing list. BN Bryan Newbold [Atp] Re: personnel assignments for ATP To: Andrew Newton (andy), Cc: atp@ietf.org This is so great to hear! Big thanks to Orie, our BoF chairs (Richard Barnes and Justin Richer), and everybody who has contributed in the BoF and chartering process. Hopefully we can nerd-snipe you all to stick around and participate in the WG. --bryan See More from Andrew Newton (andy) Atp mailing list -- atp@ietf.org To unsubscribe send an email to atp-leave@ietf.org
Unsubscribe
12:43 AM Details](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:6hbqm2oftpotwuw7gvvrui3i/bafkreib2fgrla6ro2hn3vkpwmaqyxknxs2zj4q5t437to35vzwvrj6qyqe)
Unsubscribe Yesterday at 11:43PM This message is from a mailing list. AN Andrew Newton (andy) [Atp] personnel assignments for ATP To: atp@ietf.org Hi all, Shuping Peng and Mallory Knodel have stepped forward to co-chair the ATP working group. Many thanks to them for taking on this role for the ATP wg and the ETF. With this assignment, the secretariat should be able to formally announce the creation of the ATP working group. Additionally, going forward I will be the responsible AD for this group. Many thanks to Orie for shepherding the creation of this group, and let us wish him well in his retirement from the IESG. -andy, ART AD Atp mailing list -- atp@ietf.org To unsubscribe send an email to atp-leave@ietf.org This message is from a mailing list. BN Bryan Newbold [Atp] Re: personnel assignments for ATP To: Andrew Newton (andy), Cc: atp@ietf.org This is so great to hear! Big thanks to Orie, our BoF chairs (Richard Barnes and Justin Richer), and everybody who has contributed in the BoF and chartering process. Hopefully we can nerd-snipe you all to stick around and participate in the WG. --bryan See More from Andrew Newton (andy) Atp mailing list -- atp@ietf.org To unsubscribe send an email to atp-leave@ietf.org Unsubscribe 12:43 AM Details
So exciting! Huge props to @bnewbold.net, @dholms.at, et al for making this happen. Definitely still a long journey ahead but this is a huge step! 👏
THIS IS GENIUS
Heeey, look, it's me!
I'm super hyped to announce that @bsky.app have given me a grant to work on the standards for the Federated Credential Management API (or FedCM) to make them really work for all decentralized web applications.
Thank you to Google @opensource.google for joining #AtmosphereConf as a major sponsor! We appreciate your support!
Next week! Hypha will be representing tech coops at the launch of the Commoning Labour and Democracy at Work book launch. There are still a few tickets left. Come hear some different takes on transforming the economy
We will be giving a lightning talk at ATmosphere 2026! Come along to hear @makeworld.space and @udit.bsky.social talk about how provenance tools can be used by communities to restore transparency in a world of AI-generated feeds.
today, we're announcing our €3,8M ($4.5M) seed financing round, led by byFounders with participation from Bain Capital Crypto, Antler, Thomas Dohmke (former CEO of GitHub), Avery Pennarun (CEO of Tailscale) among other incredible angels.
read more on what's next: blog.tangled.org/seed
If you want to put perceptual hashes on ATproto, this is a draft spec explaining how to do that. Let's chat!
Banner showing a square icon of a simple line-drawn flower, the name “v______”, the subtitle “ecotonal gardener,” and a horizontal row of small, differently colored abstract flower icons across the top.
You're invited to visit my garden and steal all my special spores.
spores.garden/charlebois.i...
Have you noticed how some people have bad takes? Horrid!
Couldn't we empower them to engage with unfamiliar topics in an informed way? How do we make context travel?
In our latest dripline post, @literaryarsonist.press uses the 1700s as a lens to explore the structure of online discourse.