A Conversation About Technology and Democracy with Brian Eno and Evgeny Morozov #cybernetics www.cccb.org/en/multimedi...

Schismogenesis

Schismogenesis

the easiest way to control the lights in my apartment from one place is with a Flipper Zero. which is deeply, deeply weird. and i love it.

Have been using @sill.social for some time now. What a great way to catch up on popular links shared by folks i follow. Just turned on the daily digest feature today. Highly recommend joining sill and linking your account here for the full experience. Would gladly pay for this service.

#atproto lexicon is RDF for the kids

Hello user experience, content design, information architecture, UX research, and adjacent professionals 👋 Say hi if interested in the future of online social and/or #atproto. I plan to pilot a few UX office hours sessions for the AT Protocol community, and would like to invite new perspectives in

What a week.

i think it's funny that we've been trying to get people to write proper documentation for decades and the thing that finally does it is the prospect of getting the computer to write the code for you

Just started my Year 7 (lower secondary) biology curriculum. First lessons: Autopoiesis and biomass accumulation. Rather than a reductionist approach, which dives straight into the particulars, I begin with a sense of the whole organism. #SciTeachUK #iTeachBio

"the best way to fight back is to care. Care loudly. Tell others. Get going." Dan Sinker via @patak.dev

I have no Verizon cell service in Seattle. It is very eerie.

Soundsource by @rogueamoeba.com is probably my favorite mac app.

Sabine makes a good grain bowl.

#atproto social graph discourse, this one's for you

I like this framing @chenoehart.bsky.social shared today during our conversation about about computers, cognition, and cybernetics. "A problem with anthropomorphizing robots is that they inhabit physical spaces differently than we do."

The Unreasonable Effectiveness of Actually Caring and Giving a Damn

Apropos of our conversation about Beer's Viable System Model, @gordon.bsky.social

Platform for change---Stafford Beer

simulacra

the best taco plate in seattle is made by lovely people working out of a food truck in the Chevron station parking lot by 15th and 58th in Ballard

I'm going to say the word hypothesis a lot more in 2026.

Technosolutionism and fascism rhyme with each other.

i should not have had coffee at 6pm

thank goodness for aphanipoiesis

atproto is a causal thicket. bsky.app/profile/chri...

Submitted a couple panel discussion ideas for the Media & Civics track at Atmosphere Conf in Vancouver (in March). Check out the track and send in ideas!

is there a list of all atproto apps out there that’s sortable by public usage?

Just realized i can probably start talking about second-order science rather than harping so much on second-order cybernetics. Second-order science is the thing.

Live Now! stream.place/hantohoncomb...

If you find yourself wanting a say in what you work on or how you work, know that you do have the ability to define your job. It does take time to build confidence, bc confidence is informed by experience, self-awareness, system awareness, discipline in experimenting & consistency. It is a journey.

What're the good co-working spaces in Seattle? I would like to escape my home office at least a couple days a week

@storacha.network fyi i can’t sign up for a plan from my mobile browser (start storing buttons don’t do anything on tap) on iOS

AKA abductive process or multiple description: how one context informs and describes another context. Nature does this all the time, and it requires circular feedback, among other things. Good luck trying to get a computer or an algorithm to think or assemble anything by abductive inference.

All meaningful forms of resistance boil down to radical love of self and others, even when it’s inconvenient, and stubborn belief that change is possible, even when systems loom large and powerful. Don’t let the bastards convince you otherwise.

Tired: context engineering Expired: prompt engineering Inspired: software engineering

a twenty-something busker at the farmers market just said “ok now i’m gonna play an old classic.” and proceeded to belt out snow patrol’s chasing cars.

I have been reminded that I already wrote four years ago about how to design a control system for agentic systems. apenwarr.ca/log/20211201

A problem in education: ● Losing the term "cognition" to computer metaphors: encoding, storing, retrieving and classical cognitivist models. How can we get it back? ● Embrace enactive/embodied views of cogsci Cognition is the sense making all organisms carry out in their being and doing. #EduSky

i like this articulation

a problem with premature prototyping—or playing with high-fidelity visuals too early—is it’s easy to dress up an idea with low conceptual integrity and give it the appearance of a good idea

apps like imessage or gmail border on malpractice by suggesting messages automatically. let the user select “Suggest response if they’d like, sure. Showing a suggested response that 80% of the time is either inappropriate or obtuse is just a bad/distracting/harmful user experience.

User research?

One of my deep psycholinguistic concerns about AI chatbots is the use of self-referential language--chatbots using pronouns like "I"/"me"--which subliminally assert a sentient mind ("It says 'I think', therefore, it is"). @kashhill.bsky.social brilliantly unpacks: www.nytimes.com/2025/12/19/t...

if you go through the popeyes drive-thru with a fancy ai car there’s a non zero chance the speed limit will change from 25 mph to 110 mph

In which i offer to facilitate a research project to explore atproto ecosystem brand and user experience needs discourse.atprotocol.community/t/research-p...

like a buncha pals, i've eked out a living doing some form of information architecture labor for more than a handful of years. sometimes that work product has been branded by people i work with as AI readiness, data strategy, systems design, management consulting, or whatever else you wanna call it.

I don't know how many times this needs to be said, but apparently a lot: If you are not in a position to verify the accuracy of some synthetic text, the synthetic text is not useful/has a high potential to be misleading. This is one of those cases.