J. Nathan Matias 🦣

Social & computer scientist who works alongside communities on science for a safer, fairer, more understanding Internet.

Founder, Citizens and Technology Lab · Prof at Cornell · Guatemalan-American · Co-founder of @transparenttech

Enjoys taking photos & listening to books/poetry on very long bike rides.

Public Key

npub163cvjt4awndd9szfzzpv8ed720yyx98ee2ua288vpcw3eu57n6rs5tcgde

NIP-05 Address

natematias@social-coop.mostr.pub

Profile Code

nprofile1qqsdguxf967hfkkjcpy3pqkrukl98jzrznuu4ww4rnkqu8gu720fapcpz4mhxue69uhhyetvv9ujuerfw36x7tnsw43qpke8us

Publishing to

relay.ditto.pub

Author Public Key

npub163cvjt4awndd9szfzzpv8ed720yyx98ee2ua288vpcw3eu57n6rs5tcgde

Show more details

Last Notes

2026-10-25 23:41:47 UTC

Here in Barcelona for #MozFest, I had a wonderful ride today out to the Abbey of Montserrat.
Riding the 10th century Montserrat Pilgrimage Path on the way back, I reflected on the countless millions of people who have taken this journey over the last millennium. Today, I too joined this long journey of human endeavor and reflection.
Inside the basilica, I sat for a moment in a chapel whose artwork invited prayer for those who flee their homes due to violence and persecution. Displacement has a long history, as does the deep tradition of advocacy and hospitality for people who face such things.
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/504/158/448/455/996/original/6860530bf8b09b65.jpg
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/504/158/856/108/100/original/7b12d59eaa32070b.jpeg
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/504/159/411/649/652/original/5a23312f75431041.jpeg
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/504/159/886/357/797/original/3907c7c41b995026.jpeg

2026-10-23 02:11:03 UTC

This is the work, folks.

2026-10-22 23:26:01 UTC

How are middle powers leading on tech policy? Grateful for the chance to catch up on Canadian tech policy with the global conversation at the Attention 2026 conference here at McGill University.
Grateful also for the ride up Mount Royal at sunrise 🌅
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/487/144/593/576/321/original/f3e51bec12bc60ec.jpeg
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/487/144/661/715/427/original/248795ffd81e9406.jpeg
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/487/144/740/383/413/original/75d6f8aec208df07.jpeg

2026-10-21 10:52:58 UTC

Who else is going to be at the "Attention: Democracy, Rebuilt" conference in Montreal? Looking forward to seeing folks- I arrive at the end of the day today.
Planning to do intervals on Mount Royal if anyone fancies a chilly bike ride tomorrow morning!
https://www.attentionconferences.com/conferences/2026-forum

2026-10-18 22:50:10 UTC

"The remaining farm walls we see today signify not the confidence of yeomen farmers investing their labor for generations to come but the last desperate measures of people trying to stave off Appalachian hardship and failure before heading out West."
https://www.theatlantic.com/ideas/2026/07/second-career-passion-life-meaning/687952/

2026-10-18 13:45:52 UTC

Movies about social change & organizing don't often have dramatic spreadsheet scenes where someone made a difference in the world through responsible financial projections, but if they did, my morning would have been epic.

2026-10-17 14:46:37 UTC

“The Tioga is going to go from dead to being a trout river, and not a little trout river,” he said. “The Tioga River is going to turn into a destination fishery.”
https://capitalandmain.com/once-a-bucolic-pennsylvania-river-the-tioga-is-now-all-but-dead

2026-10-15 18:30:10 UTC

It's now a cliche to say that the tech industry is more innovative than scientists. But often the tech industry does things that scientists pointed out were bad a decade ago

2026-10-13 21:16:27 UTC

TFW you post to friends to say you narrowly escaped a huge fire and Meta upsells you on reaching more of your friends with the update by paying extra.
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/435/674/522/099/600/original/7ddc3cb6b0114fc3.jpeg

2026-10-13 08:33:43 UTC

As more efficient, capable AI models are given more power and autonomy in the form of agents, how can we notice when they go wrong and prevent catastrophes in the future?
@nprofile…7l6p & I have written a new report on incident reporting for open models & agents.
https://citizensandtech.org/2026/10/incident-reporting-beyond-frontier-models/

2026-10-12 13:58:40 UTC

How is everyone else respecting indigenous people's day today?
I know many of you will use it to catch up with work & family away from the office, and that's wonderful.
As a Guatemalan-American of Mam heritage, here are indigenous projects I admire. Maybe take a minute today to learn & donate?

2026-10-11 22:56:50 UTC

Catching up on LudoNarracon, a virtual festival of story-rich games
https://www.ludonarracon.com/

2026-10-11 02:33:49 UTC

Looks like I'll be going to MozFest this year and am looking forward to meeting all sorts of new folks and old friends.
Also if anyone wants to ride bikes in the mountains around Barcelona - perhaps out to Montserrat - ping me!

2026-10-09 14:00:18 UTC

This spring, I wrote about the problems with the serial rapist model and why It's important to follow the evidence even when the data challenges our belief in a just world.
The post is circulating again 💔 so I thought I would share it here.
https://natematias.medium.com/when-science-isnt-gaslighting-you-after-all-0710a491fbcc

2026-10-08 22:20:26 UTC

Just ordered one of the last remaining Pace cycling caps in stock at a local bicycle shop. Grateful for their story and the quality of their work, even if the family business is no longer with us.
https://theradavist.com/pace-making-cycling-caps-since-late-70s

2026-10-06 19:17:42 UTC

Sigh, more emails from prospective students who use chatgpt tracking codes in their emails asking about PhD programs.
As a scientist who does participatory research, the idea at the heart of my work is the trusting human connection between people.
I disregard such outreach from decisions in whatever form it takes, but my heart sinks when I see people who, for whatever reason, feel pressured to treat one of the most significant intellectual relationships of their lives as just another unit in a decision-making system.

2026-10-06 12:57:19 UTC

Me, a scientist: I want to live in a world where precise estimates matter
Also me: that would be terrifying
https://www.youtube.com/watch?v=YEaicg1kYMk

2026-10-06 10:38:36 UTC

Deeply moved by the way that the Cornell community is coming together. I hope this catalyzes longer-term collective efforts for change.
https://www.nytimes.com/2026/10/05/nyregion/cornell-rape-case-protests-vandalism.html

2026-10-03 22:41:59 UTC

Persimmons, parks, ponds, and pizza 🍁
#outdoormoments
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/379/387/260/352/888/original/7baf8042bb710a14.jpeg
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/379/387/472/689/906/original/2f88efa86c32ef10.jpeg
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/379/387/581/059/863/original/cfc7858180c71760.jpeg
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/379/387/728/905/561/original/104abaf94a35a5c9.jpeg

2026-10-03 12:20:11 UTC

Huge news. David Robinson is one of the smartest, wisest, most principled, and most pragmatic people I know. The day he decides a problem is too endemic to solve from the inside, that is a day we need to notice and take action.
cc @nprofile…ugpy
https://www.theatlantic.com/technology/2026/10/openai-safety-team-resignation/688881/?gift=jE7sd3rFOqDdQaD53F1IhcqkhImekiWN9aB9R67GhAo&utm_source=copy-link&utm_medium=social&utm_campaign=share

2026-10-01 22:44:41 UTC

The 19th summarizes a New York law at the heart of the Cornell case. Many states hold that intoxication can prevent someone from consenting. Not NY:
"if someone chose to consume alcohol... and then was sexually assaulted, it can be harder to pursue criminal charges"
https://19thnews.org/2026/10/voluntary-intoxication-laws-assault-cornell-seven/

2026-09-30 20:24:49 UTC

"What we need to demand from universities far and wide is clear: Cease to be a place where sexual assault is often subject to less punishment than protesting and plagiarism."
Kate Manne with an important call to action for Cornell & higher ed in general
https://www.thecut.com/article/what-it-feels-like-at-cornell-right-now.html

2026-09-30 00:11:20 UTC

Reading Edward Taylor for no reason at all and realizing that the displacement theory of social media & mental health has deep roots in Puritanism:
The idea that when we follow our instincts, those uses of time displace things that are better for us. For Taylor, it's literally liquid in a mold.
If you're really want context, you can the long poem "God's Determination Touching His Elect."
The displacement theory of moral life appears deep in a debate between a Saint and the Soul on "Doubts from the Want of Grace Answered."
You can't say I don't cite sources
https://archive.org/details/poemsofedwardtay00tayl/page/314/mode/2up
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/357/087/028/958/233/original/d3e533d4c013183b.png

2026-09-29 15:38:58 UTC

Great question! We've seen a lot of folks ask about human-only social spaces. A lot depends on what you mean by "human-only"
For example, automation is pretty important for reducing the mental health toll of moderating very disturbing content online https://arxiv.org/html/2502.00244v1
Some Black online communities have experimented with requiring people to submit photos of their skin or otherwise apply to participate in Black online spaces. Our collaborator William Frey wrote about it here:
https://journals.sagepub.com/doi/full/10.1177/20563051261452021

2026-09-29 11:35:31 UTC

Wow, LinkedIn's algorithm really doesn't want people to read @nprofile…x40m's article about keeping the Internet human in the era of AI
Out of the 10,200 impressions the feed gave my posts last week, it only gave 65 to Sarah's *important* article in @nprofile…ay7g
As Sarah writes, "If AI content is allowed to drive out human content and prevent people from connecting with each other, we risk dual and intertwined crises of knowledge and community."
https://www.techpolicy.press/millions-of-volunteers-are-keeping-the-internet-human-for-now/

2026-09-26 21:58:28 UTC

Today a social media algorithm showed me a post stoking outrage about the cost of a local library's holdings.
One of the top comments was a thread started by a parent who whose child had donated those works to the library before passing away.
What started as an argument continued as a tribute. <3

2026-09-26 15:01:31 UTC

Thinking about that time ten years ago when I wondered if I was going too out on a limb to name, in my PhD dissertation, the problem of tech authoritaranism in what was basically a behavioral science document.

2026-09-25 21:16:23 UTC

Sigh. Slack has logged me out of about 15 communities, and I now have to click on 5 bridges and 3 motorcycles for every single channel. 🫠️

2026-09-25 19:47:53 UTC

When you think about it, why not express your admiration for surveillance technologies, efficiently introduced without the hassle of community consultation?
https://fnl.mit.edu/how-we-learned-to-stop-worrying-and-love-campus-surveillance/?trk=feed-detail_main-feed-card_feed-article-content

2026-09-22 18:06:00 UTC

Langdon Winner: "Characteristic of this way of thinking is a tendency to conflate the activities of freedom seeking individuals with the operations of enormous, profit seeking business firms."

2026-09-22 12:31:33 UTC

"We found more than 1,050 people who died within range of border surveillance towers between 2015 and early 2026. These deaths are not failures of a few towers.. We found deaths within the advertised range of nearly two-thirds of all the towers we analyzed."
Extraordinary investigation by James O'Donnell & @nprofile…u8e2
https://www.technologyreview.com/2026/09/21/1144166/border-towers-surveillance-investigation/

2026-09-11 11:31:53 UTC

Phrase I am very pleased to write in an email to collaborators today: "We can build the AI Thrutopia"
If the term "thrutopia" is new to you, here's some background:
https://www.cbc.ca/news/climate/what-on-earth-thrutopia-9.7206857

2026-09-07 11:32:47 UTC

"A world where there is no N.S.F. funding for social science, but where the A.I. industry funds research... will skew public debate. To navigate the shock waves of A.I., we must restore social science research as an enterprise that starts from the public interest."
Important article by @nprofile…n0xh, @nprofile…urqa, and James Evans
https://www.nytimes.com/2026/09/06/opinion/ai-social-sciences.html

2026-08-30 22:58:14 UTC

Grateful for the chance to get back to sailing, which connects me to the deep time of human creativity, adventure, and wndless maintenance
#outdoormoments
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/186/933/120/685/230/original/1e9c69a72fa1ea3c.jpeg

2026-08-29 18:25:29 UTC

How can Christian communities and technology experts build shared understanding and action for the common good?
Over the last two decades, I've been privileged to see many of the ways that scientists engage with communities of faith.
In 2026, something feels different, as Christian communities have learned to ask critical questions about technology and face the urgency of choosing a more wholesome path than the one that tech firms try to scare us into adopting.

2026-08-28 11:51:53 UTC

“Canada’s willingness to support research that challenges commercial and political threats to democracy is deeply necessary at a moment where authoritarianism is on the rise around the globe”
https://www.ft.com/content/8cc488a5-54c2-41f9-8b87-327ff591be6a
- @nprofile…xxx3

2026-08-25 19:03:00 UTC

"This year, The 19th also launches a development fellowship — a critical and historically underrepresented career pathway in news. It’s designed to expand who participates in shaping how journalism is funded, while equipping the fellow with experience in fundraising strategy, donor engagement and the business of nonprofit journalism. "
https://19thnews.org/2026/08/introducing-2026-2027-frances-ellen-watkins-harper-fellows/

2026-08-23 19:46:46 UTC

- reply

@nprofile…73zx Thanks Brian. I was FASCINATED too!
I've experienced the push and pull of editors looking for a particular story and the negotiation that comes with that. But never something that extensive!

2026-08-23 19:43:33 UTC

- reply

In one astonishing tale, Wilson shows how the physicists Hans Bethe and Richard Garwin were “assertively hand-held” by Scientific American into becoming unexpected nuclear whistleblowers—after submitting a draft to support the weapon their article eventually opposed
https://thebulletin.org/2026/01/how-hans-bethe-and-richard-garwin-served-the-missile-defense-system-they-publicly-criticized/

2026-08-23 19:13:11 UTC

- reply

Wilson's book Strange Stability checked off a bingo card of issues that could have easily been in AI policy news today:
- warnings of existential risk from well funded insiders
- scientists who keep secret their industry ties
- definitions of science that uphold the politics of profit and war
- claims that only insiders can truly understand the scientific issues
- a popular movement against harmful tech valorizing its creators
- whistleblowers who profit from business interests they mildly critique in public

2026-08-23 19:12:18 UTC

What can tech policy folks learn from the nuclear age about the role of science in policy?
My read of the summer is Benjamin Wilson's new book "Strange Stability" about the backstage arrangements of nuclear physicists during the cold war.
The book checked off a bingo card of issues that could have easily been in the news today:
- warnings of existential risk from well funded insiders
- scientists who keep secret their industry ties
- definitions of science that uphold the politics of profit and war
- claims that only insiders can truly understand the scientific issues
- a popular movement against harmful tech valorizing its creators
- whistleblowers who profit from business interests they mildly critique in public
https://natematias.com/portfolio/2026-08-23-strange-stability/

2026-08-23 02:14:52 UTC

- reply

@nprofile…xyqv 🤩 🤩 🤩

2026-08-20 13:06:04 UTC

Hypothesis: even relatively weak causal arguments with timeseries can be persuasive because debates over causation (as people squint at the chart) generates debate that spreads a catchy plot further to more people who believe it.
https://www.tylervigen.com/spurious/correlation/1079_good-air-quality-in-grand-rapids-michigan_correlates-with_associate-professor-salaries-in-the-us
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/127/975/647/983/228/original/199c04436e443058.png

2026-08-20 12:16:30 UTC

- reply

The ACM and AOIR are just two of the organizations that have made public statements in defense of research in recent times.
DefendResearch.org is tracking censorship of knowledge and research in the US and has added their statements to an encouraging 15 page list of broader efforts #DefendResearch
https://docs.google.com/document/d/1Y5_fjGPNh4_1Hgc0Q68Hr2vJhVIOK1Qc-KynfIG6fZI/edit?tab=t.0

2026-08-19 02:17:34 UTC

- reply

@nprofile…sg42 ooooooh? Does that mean what I think it means?

2026-08-18 22:09:27 UTC

- reply

Why would the US threaten researchers who study tech safety when our work is in high demand?
Writing in Science this January, Alondra Nelson explained that US promises of protection from regulation actually maximize discretion in how the gov punishes, rewards, & controls tech:
https://www.science.org/doi/10.1126/science.aee4900

2026-08-18 18:07:56 UTC

- reply

@nprofile…xsfl ooh is the defense now? Sending my best wishes!

2026-08-18 17:14:17 UTC

- reply

In the post I tell the story of a scholar who avoided risks at the border by declining to travel, despite being in a computing field that counts and rewards people for conference presentations.
Under those circumstances, when we review people's CVs, we become unwitting accomplices to censorship.
But when we understand how repression works, we can design responses:
- challenge unconstitutional intimidation in the courts, as The Coalition has done
- counteract peer effects by standing visibly for rights, as ACM and AOIR have done
- account for repression when making institutional decisions, which the lawsuit and statements from leading organizations help inform

2026-08-18 16:18:27 UTC

New article: How do governments rewire science and the tech industry by threatening scholars? And how can we make science and tech resilient to that intimidation?
Writing for the Union of Concerned Scientists, I share practical efforts by scholarly and professional societies to protect science and technology from:
- direct intimidation
- peer effects that deter people from important work
- selection effects that make our institutions unwitting accomplices
Many thanks to the @nprofile…curl and @nprofile…h0gd for making public statements in support of academic freedom, and to @nprofile…e774 and the @nprofile…cngd for working to defend researchers in the courts.
https://blog.ucs.org/guest-commentary/protecting-science-from-government-intimidation/

2026-08-18 14:31:05 UTC

"Without work permits, the Haitian residents are struggling to pay for rent and groceries. Some are hiding in their homes, too afraid of ICE to shop at local businesses or attend church..."
via @nprofile…c5kx
https://religionnews.com/2026/08/17/as-ice-closes-in-ohio-churches-line-up-families-to-care-for-haitian-kids-left-behind/

2026-08-17 22:01:40 UTC

- reply

@nprofile…3ge4 yes, I remember that too!
Our lab is doing a related project right now: https://citizensandtech.org/2025/12/online-communities-as-ecosystems/

2026-08-17 21:59:57 UTC

Hi folks, give a warm (re) welcome to Scott Klein, a longtime leader in technology + journalism.
Scott co-leads a number of projects that track and support open source software in the news.
https://www.kleinmatic.com/

2026-08-17 21:57:16 UTC

- reply

@nprofile…3ge4 <3
Mosstodon is a pun based hashtag where people share beautiful pictures of moss
Lichensubscribe is another pun based hashtag where people share beautiful pictures of lichen. This hashtag, I think, is a way for a group of folks who think about "rewilding" the internet to connect with and collectively enjoy something that grows on a much slower timescale than typical social media algorithm driven content.
https://www.noemamag.com/we-need-to-rewild-the-internet/

2026-08-17 21:20:30 UTC

- reply

@nprofile…3ge4 Welcome back! You have a lot of wonderful #Mosstodon and #lichenSubscrube pictures to catch up on, conversations about AI, and lots of mutual aid around pretty much any topic.

2026-08-17 19:47:22 UTC

- reply

@nprofile…n00n oh wonderful, thank you!

2026-08-17 19:37:33 UTC

- reply

@nprofile…n00n I see the link to audio and video is down- are they archived somewhere public?
https://library.harvard.edu/exhibits/war-words

2026-08-17 19:36:49 UTC

- reply

@nprofile…n00n oh no! This means I missed it! Thank you for curating!

2026-08-17 19:24:46 UTC

Yet again, science is saving me time and money. Today I learned that post-exercise massage, rollers, etc do reduce pain but have no measurable effect on performance and recovery.
That's one part of my life that I can now drop in order to focus more on the people and things that I care about.
https://bmjopensem.bmj.com/content/6/1/e000614
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/112/474/777/039/079/original/cbb94e25427812b1.png

2026-08-16 21:15:24 UTC

- reply

@nprofile…6ns8 @nprofile…uj42 that meme <3 😢

2026-08-14 22:39:16 UTC

- reply

@nprofile…577a Updated. Thanks for the suggestion.

2026-08-14 21:31:36 UTC

I hope folks realize that scholars of color, perhaps especially the most scrupulous with impeccable integrity, are wondering: given how effective intimidation was here, are they going to come for me now?
https://www.theguardian.com/uk-news/2026/aug/14/jason-arday-found-dead-after-resigning-from-cambridge-over-plagiarism-claims

2026-08-14 11:38:07 UTC

- reply

Reading the story of Athanasius gives me some hope for the endeavor of building understanding and consensus at times of deep disagreement where money and power are on the line and ideas are a risk because they matter.
The story also gives me empathy for all those who are forced to flee their homes for their ideas- Athanasius had to do so at least five times. Next time I speak the Nicene Creed, I will say a prayer for all those who face displacement for their ideas.
#outdoormoments

2026-08-14 11:37:49 UTC

Here in the Boston area, I have finally found a hill I can train on, in Arlington MA just a half hour ride from home.
At the base of the hill is this beautiful Orthodox church with a stunning mosaic depicting Saint Athanasius.
It's been years since I read his notes on the incarnation, so I looked up his story and had no idea how involved he was in the politics of his time- the cause of his first exile (of many) was an accusation by the followers of his theological opponent Arius who claimed that Athanasius had restricted African trade to the Roman Empire.
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/093/658/093/087/489/original/652ca75c96fc72e7.jpeg
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/093/658/112/739/233/original/e04e32c2915da9d1.jpeg

2026-08-13 13:44:51 UTC

- reply

@nprofile…c5kx working on Penelope's Bones by Emily Hauser, and continuing to read @nprofile…dncl's "For Such a Time as This
https://www.eerdmans.com/9781467470575/for-such-a-time-as-this/

2026-08-13 11:39:37 UTC

- reply

@nprofile…97y5 thank you for your work on this, both the work that your book tells us about, and all of the work to tell your story in a way that others can learn from

2026-08-12 13:23:51 UTC

- reply

@nprofile…lhdz you're welcome! I haven't used the tool and am not sure it's even still available, but I hope it ends up being a helpful wayfinder!

2026-08-12 13:17:20 UTC

- reply

@nprofile…lhdz The PreprintMatch project does this:
https://www.scicrunch.com/preprintmatch
Here's the paper they wrote about global inequality in publication rates as a result:
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0281659

2026-08-12 13:14:52 UTC

- reply

@nprofile…lhdz cc @nprofile…ev4a ^

2026-08-12 12:50:31 UTC

- reply

@nprofile…hrrk my guess is that one reason for this success is a program that has expanded the number of places where people can take the training. Until last year, people had to study in Oklahoma City.
A surge in admissions is a smart idea, because it helps with budgeting for the colleges that have stepped in to grow their programs, in time for the new academic year.
https://www.nytimes.com/2025/10/31/us/politics/air-traffic-control-training-shutdown.html

2026-08-12 01:31:34 UTC

- reply

@nprofile…8txc O_O

2026-08-11 14:23:06 UTC

- reply

@nprofile…d4hg hurrah for consequential bureaucratic victories even if they're not the kind of thing that is very visible

2026-08-10 00:45:19 UTC

- reply

@nprofile…99z9 oh, thanks for this recommendation; it looks like a beautiful film!

2026-08-10 00:19:35 UTC

- reply

@nprofile…3mlt and I learned much about cultivating ideas over time.
My gift honors that: a maple box, syrup from the local forest, & collective nouns. This fall, I will harvest maple seeds to add.
> Seize the minute, build a new world, sing an old song
> Keep up the fire, maple syrup time
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/068/343/213/423/568/original/a7119d7b7fdf914f.jpg

2026-08-10 00:18:08 UTC

- reply

What goes into a good graduation gift? You may have heard a phrase from "Paterson" by William Carlos Williams, "there are no ideas but in things." (I visited Paterson this March)
Sherry Turkle quotes Williams in her beautiful collection "Evocative Objects"
https://mitpress.mit.edu/9780262516778/evocative-objects/
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/068/336/880/419/531/original/da1f6a04aaab54dd.png

2026-08-09 23:37:18 UTC

- reply

If MIT was where I was flung into onstage drama, Cornell fostered deliberate, care-filled growth.
So when @nprofile…3mlt defended, I thought of Pete Seeger:
"In life or evolution
Rarely is there a quick solution
Anything worthwhile takes a little time"
https://www.youtube.com/watch?v=aFeiFLv1HdE

2026-08-09 23:36:18 UTC

For the last nine years, after @nprofile…xxx3 pulled out a plastic sword & attacked me on stage, many have asked whether I plan to make my own students' defenses real in that way.
The sword, signed by my PhD committee, is one of the most special items I own. But I wanted to do something else.
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/068/172/960/675/275/original/7f997d34252939ea.png

2026-08-08 14:34:36 UTC

- reply

@nprofile…uya3 @nprofile…p388 the article by @nprofile…u8e2, is SO deeply and well researched!

2026-08-08 11:49:36 UTC

- reply

@nprofile…6ns8 🤩

2026-08-08 11:33:34 UTC

- reply

@nprofile…s45g @nprofile…d2v0 @nprofile…6r9q Nice! Are those the Continental Contact Urban tires?
They have pretty good rolling resistance! https://www.bicyclerollingresistance.com/tour-reviews

2026-08-08 02:35:42 UTC

- reply

@nprofile…d2v0 @nprofile…s45g @nprofile…6r9q
Congrats on getting a folding bike! My top tip is to get tires with a very good rolling resistance and puncture resistance, and always bring spare tubes since you can't count on shops to have them (even if you run tubeless).
A substantial part of people's belief that folding bikes are slow comes from stock tires that have high rolling resistance.
I see people spend so much money on aftermarket lightweight and aero parts for their folding bike, and then use very slow tires.
#bikenite

2026-08-07 20:12:59 UTC

- reply

@nprofile…zejv Yep- just looked at the time and clicked "schedule send"

2026-08-05 21:14:08 UTC

Reading the law in Washington state, it looks like employers are required to take measures to control exposure above AQI of 100, but there doesn't seem to be a maximum threshold.
For reference AQI of 200 is "Very Unhealthy"
> EPA's risk assessment (Whitfield et al., 1996) indicates that at this level for healthy individuals (adults and children) at moderate exertion: 1) approximately 50% are estimated to experience moderate or greater lung function impairment, 2) approximately 20% are estimated to experience large or greater lung function impairments, and 3) approximately 10-15% are estimated to experience moderate to severe respiratory symptoms (i.e., chest pain with deep inspiration and aggravated cough)
https://app.leg.wa.gov/wac/default.aspx?cite=296-820&full=true

2026-08-05 00:32:53 UTC

- reply

@nprofile…4hqr ooof. I am currently stuck on a platform acquired by them and wondering how to build the collective action momentum to support a workable alternative. I would very much welcome suggestions.

2026-08-03 23:50:33 UTC

- reply

@nprofile…8atu ooh thanks! Flagging to read.

2026-08-03 14:12:40 UTC

Ohai MIT Press- happy to see our book out in the wild!
#jsm2026
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/031/984/125/285/618/original/6d470df79a094019.jpeg

2026-08-02 19:45:38 UTC

- reply

@nprofile…kv5n I would reach out to scholars in Communication who have built expertise monitoring ads- which many of them building infrastructures to study political ads.
There's less money/energy in consumer protection than in studying political campaigns. But the archiving practices are similar.

2026-08-02 17:55:16 UTC

- reply

@nprofile…ylve thanks! It's about 30 years old now, and a ship of theseus that I keep upgrading as parts wear out !

2026-08-02 10:27:53 UTC

- reply

@nprofile…fd3p Wow. And I think academia has an intense schedule.

2026-08-02 00:43:24 UTC

- reply

@nprofile…zejv @nprofile…s7k3 this is coming at just the right time, thank you!

2026-08-01 21:42:58 UTC

- reply

@nprofile…xrhq thanks for the link; Klein, Elter and collaborators have done great work to investigate the challenge of replication, and put it into its historical context!

2026-07-31 01:37:41 UTC

Sometimes words just aren't enough to describe the grace of living in a world of such beauty.
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/012/012/535/931/231/original/cd70b92f4fcbc0ac.jpeg
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/012/012/434/595/739/original/1205d6806bbeae91.jpeg
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/012/012/537/373/446/original/cd58cbe1916e0427.jpeg

2026-07-30 13:59:17 UTC

"I’m not sure what these students should be learning... Increasingly, I find myself asking not simply whether students can do the work, but whether they should want to do the work."
"So much of it is surveillance & extraction"
Carl DiSalvo, Georgia Tech
https://underthecircumstances.substack.com/p/fall-teaching-part-1-before

2026-07-29 20:04:58 UTC

- reply

@nprofile…x0ca Yes! The Nature Reviews Psychology covers the issue of open source in greater depth, both in respect to code and model weights for folks who work with LLMs.
I'm glad this is helpful!

2026-07-29 16:39:14 UTC

"Sociocracy is a governance system, just like democracy or corporate governance methods. It’s best suited for organizations that want to self-govern based on the values of equality."

2026-07-29 11:11:29 UTC

- reply

In "Proprietary LLMs impede scientific transparency and reproducibility,"
Killian McLaughlin, Alexis Palmer, Molly Crockett, and I explain how the rush for LLMs in social science is destabilizing scientific reliability.
Proprietary models lack the transparency for repeatable science, and even worse - their behavior drifts over time. Alexis has shown that research findings can completely change between two runs of the "same" model over time.
Our article offers suggestions, and Killian is working on a stealth project in this space.
https://www.nature.com/articles/s44159-026-00593-8

2026-07-29 11:03:58 UTC

- reply

In "Commercial Determinants of Replication Infeasibility," Cassidy Waldrip, David Lazer and I ask if fears of corporate control of computational social science have become true.
The answer? Few papers that rely on tech firms can be replicated. In contrast, studies that rely on Wikipedia are much more replicable, and the replication barriers are often due to third-party corporate data that they included in the study.
https://journalqd.org/article/view/9661
https://social-coop-media.ams3.cdn.digitaloceanspaces.com/media_attachments/files/117/002/928/851/750/496/original/19009fb4288102d4.png

2026-07-29 10:50:39 UTC

When people seek guidance from science, they expect it to work more than once.
I have TWO new articles out commercial and proprietary forces that make science less reliable, with suggestions.
Many thanks to co-authors Killian McLaughlin, Cassidy Waldrip, Alexis Palmer, Molly Crockett and @nprofile…03w5
Blog post here: https://citizensandtech.org/2026/07/commercial-determinants-science/

2026-07-29 00:43:56 UTC

Looking for something witty or insightful to add, but all I have is a weary sigh and a commitment to keep working as long as it takes

2026-07-27 15:49:01 UTC

- reply

> "Some find me a sword; some
> The flange and the rail; flame,
> Fang, or flood" goes Death on drum,
> And storms bugle his fame.
> But wé dréam we are rooted in earth—Dust!
https://www.poetryfoundation.org/poems/44403/the-wreck-of-the-deutschland

2026-07-27 15:45:32 UTC

Reflecting today on "The Wreck of the Deutschland" by G.M. Hopkins, with its tumultuous convergence of culture war, tech failure, deportation over thought crime, global media, and the drive for mastery over understanding.
It is also impossible to explain on social media