Skip to main content
Sifa ID

Activity for ronentk.me

Active on:

Loading activity...

Posted: So excited for this!! The idea for Aster was sparked in conversations in Vancouver at @atmosphereconf.org and now after a few months we're …Bluesky network
PostedBluesky network

Expecting researcher taste and judgement to become increasingly valuable (this is from Anthropic's Opus 5.5 system card) anthropic.com/claude-opus-...

  • This is highly uncertain, but we expect that full automation of AI R&D will require large improvements in foresight, prediction, creating one’s own feedback loops, and generally other skills that might typically be referred to as researcher “judgement” or “taste”. ○ The evidence we have does not suggest that [Claude Opus 5.5] represents a large improvement over Fable 5.1 in these “judgement” skills.
PostedBluesky network

lol this is going viral over on the other site. kinda strange how much attention it's actually getting... x.com/tofuafficion...

PostedBluesky network

Beautiful essay. "Perhaps capitalism’s greatest strength–its ability to produce inconceivable material abundance while fundamentally altering our social chemistry–produces its weakness. Its waste products are the friction of emotion .. it cannot commoditize & the language to acknowledge unfreedom"

Feeling (un)freedom
www.planetcritical.com
Posted: This is fantastic! Imagine how cool it would be to have a version only for all the science conversations 🤯 #Academicsky #ATScience 🧪Bluesky network

Reviewed a book

Here Where We Live Is Our Country

Molly Crabapple

10/10
PopfeedBook4d ago
PostedBluesky network

New AT Proto architectural diagram just dropped 👀 arxiv.org/abs/2609.21608

  • Schematic diagram of the AT Protocol architecture, with modifiable components and data access infrastructure highlighted. This representation is a simplified adaptation of the architecture presented in the original paper [78]. We refer readers to that work for more comprehensive technical details.
  • Open Platform Field Experiments: Expanding the Design Space of Experimental Research on Social Media JORDI GUILLEM CONDOM-TIBAU∗, University of Pisa, Italy and Institute for Informatics and Telematics, National Research Council (IIT-CNR), Italy GIOVANNI PUCCETTI, Institute of Science and Technologies of Information “A. Faedo”, National Research Council (ISTI-CNR), Italy CLARA BACCIU, Institute for Informatics and Telematics, National Research Council (IIT-CNR), Italy MATTEO ABRATE, Institute for Informatics and Telematics, National Research Council (IIT-CNR), Italy STEFANO CRESCI, Institute for Informatics and Telematics, National Research Council (IIT-CNR), Italy Despite a growing demand for causal evidence about social media, independent researchers remain severely constrained in their ability to conduct experiments directly on online platforms. To cope, multiple method- ological workarounds have emerged—from controlled surveys and simulations to client-side overlays and platform partnerships—each requiring distinct trade-offs between desirable experimental properties. The recent emergence of open social media platforms offers a qualitatively different methodological opportunity. Here we propose a design space of social media experimentation and discuss Open Platform Field Experiments (OPFEs). OPFEs represent a distinct class of experimental approaches that enable independent researchers to directly intervene on functional platform components—such as clients, recommendation systems, and moderation services—within live social media environments. Through a comparative analysis of experimental archetypes, we show that OPFEs occupy a previously unexplored region of the design space. We then bridge theory and practice by characterizing the architectural and governance elements that enable OPFEs, mapping them onto Bluesky and the AT Protocol, and illustrating the end-to-end lifecycle of a complete OPFE design. Overall, this work establishes OPFEs as a practical methodol…
Posted: Interesting implications for the centralized vs decentralized AI debateBluesky network
PostedBluesky network

Super excited for this integration! Side note but I also like that the suggested similar cards come with provenance - instead of the typical opaque recommendations, you can see the actual people who added them to @semble.so Organic, know-your-farmer links 🌱

  • Semble knowledge card for "The Entangled Brain" neuroscience book added by Ronen Tamari, showing related research articles on neuroscience and brain structure from academic sources.
PostedBluesky network

"We are entering what one might call the latency period – that characteristic phase in which the belief which sustains a bubble is still in place but faces relentless assault, where everywhere there are indicators that the previous dynamic is unsustainable." 😬

Frédéric Lordon, Warning Signs — Sidecar
newleftreview.org
Posted: This piece is very importantBluesky network
PostedBluesky network

"banality of ignorance: the willful suspension of judgment that allows people to see suffering without confronting it and to hear lies without rejecting them. The monster does not merely demand obedience. It needs subjects who have learned not to think about what their obedience makes possible."

When Monsters Walk Among Us
www.counterpunch.org
Posted: especially social media built on open protocols #ignore_atproto_at_your_perilBluesky network
PostedBluesky network

Another fun thing to do with @semble.so connections - connect a quote post of a quote post of a post referencing X, to reference X. Given a reference, it's hard to find the helpful comments about it when some are nested 3 levels out. Now we can directly connect those comments to their referent! ↔️

  • Screenshot of a social knowledge platform showing a connection made by Ronen Tamari 1 minute ago. The connection links two resources: on the left, Colin's note disagreeing with a Doctorow post about ChatBot as a front-end to autonomous hacking tools, with a quoted text excerpt; on the right, a pluralistic.net article titled "Pluralistic: LLMs are real, AI is fake (12 Sep 2026)" discussing corporate AI culture and hyperscalers. The connection is marked "HELPFUL" with an arrow between them. Below, Will Stancil's reply begins "No. This Cory Doctoro..." Both resources show engagement metrics (1 interaction each).
Posted: Late stage enshittifcationBluesky network
PostedBluesky network

"A counterculture does something else entirely. It finds the ungoverned space, builds the parallel system, and reaches the audience first. The mainstream is then forced to respond to something it had neither authorised nor understood." Great theory of change for science (and a lot of other fields)

  • A counterculture does something else entirely. It finds the ungoverned space, builds the parallel system, and reaches the audience first. The mainstream is then forced to respond to something it had neither authorised nor understood. It’s less about heroism and it’s also not about victimhood. It is more of a niche-serving approach. You find the use case the dominant system cannot see, or cannot afford to serve, or has actively written off as worthless. You build for that. You make it work for the people who need it. And you let the alternative become obsolete not by confronting it, but by making it irrelevant. That is the crux of it.
Load more