Joe Bak-Coleman
@jbakcoleman.bsky.social
Research Scientist at the University of Washington based in Brooklyn. Also: SFI External Applied Fellow, Harvard BKC affiliate. Collective Behavior, Statistics, etc..
Lord. My aunt worked at a jazz club in New Orleans for forty years and made their pecan pie…
Hers comes to mind!
Hers comes to mind!
November 28, 2025 at 2:08 AM
Lord. My aunt worked at a jazz club in New Orleans for forty years and made their pecan pie…
Hers comes to mind!
Hers comes to mind!
But sweet potato > pumpkin every day.
November 28, 2025 at 1:30 AM
But sweet potato > pumpkin every day.
Someone did you wrong with pecan pie if it’s sitting in the middle.
November 28, 2025 at 1:29 AM
Someone did you wrong with pecan pie if it’s sitting in the middle.
If they’re really hoping to find an answer, not just launder known nulls through willing scientists, why not publish their results from 2019? They seemed to recognize they were of interest more broadly.
November 24, 2025 at 12:58 PM
If they’re really hoping to find an answer, not just launder known nulls through willing scientists, why not publish their results from 2019? They seemed to recognize they were of interest more broadly.
Industry-academic collaborations are usually justified under the argument that some data is better than none, but does this hold when the company knows and can guide what you find?
November 24, 2025 at 12:57 PM
Industry-academic collaborations are usually justified under the argument that some data is better than none, but does this hold when the company knows and can guide what you find?
By the time the dust settles on this suite of collaborations we’ll have a dozen or so null findings in the literature, after 6-8 years, at odds with internal research because scientists weren’t permitted to look where the problems were.
arxiv.org/abs/2510.19894
arxiv.org/abs/2510.19894
The Risks of Industry Influence in Tech Research
Emerging information technologies like social media, search engines, and AI can have a broad impact on public health, political institutions, social dynamics, and the natural world. It is critical to ...
arxiv.org
November 24, 2025 at 12:57 PM
By the time the dust settles on this suite of collaborations we’ll have a dozen or so null findings in the literature, after 6-8 years, at odds with internal research because scientists weren’t permitted to look where the problems were.
arxiv.org/abs/2510.19894
arxiv.org/abs/2510.19894
On teen mental health, the lawsuit is replete with indications that the content meta serves teens in the problem…
So when they engaged @cos.io for collaboration, they explicitly opted not to provide information about the type of content viewed. Chop off the mechanism and get a null.
So when they engaged @cos.io for collaboration, they explicitly opted not to provide information about the type of content viewed. Chop off the mechanism and get a null.
November 24, 2025 at 12:55 PM
On teen mental health, the lawsuit is replete with indications that the content meta serves teens in the problem…
So when they engaged @cos.io for collaboration, they explicitly opted not to provide information about the type of content viewed. Chop off the mechanism and get a null.
So when they engaged @cos.io for collaboration, they explicitly opted not to provide information about the type of content viewed. Chop off the mechanism and get a null.
As we discuss in this piece, withdrawal experiments will have downward bias in their effect sizes for things like polarization. If they got a small effect or a goose egg, jt may have made them feel a lot more confident about their collaborative experiments with academics.
arxiv.org/abs/2505.09254
arxiv.org/abs/2505.09254
November 24, 2025 at 12:46 PM
As we discuss in this piece, withdrawal experiments will have downward bias in their effect sizes for things like polarization. If they got a small effect or a goose egg, jt may have made them feel a lot more confident about their collaborative experiments with academics.
arxiv.org/abs/2505.09254
arxiv.org/abs/2505.09254
It's here, but my Pacer account is currently inactive. Otherwise it's pretty cheap...
www.courtlistener.com/docket/65407...
www.courtlistener.com/docket/65407...
IN RE: SOCIAL MEDIA ADOLESCENT ADDICTION/PERSONAL INJURY PRODUCTS LIABILITY LITIGATION, 4:22-md-03047 - CourtListener.com
Docket for IN RE: SOCIAL MEDIA ADOLESCENT ADDICTION/PERSONAL INJURY PRODUCTS LIABILITY LITIGATION, 4:22-md-03047 — Brought to you by Free Law Project, a non-profit dedicated to creating high quality o...
www.courtlistener.com
November 23, 2025 at 11:49 PM
It's here, but my Pacer account is currently inactive. Otherwise it's pretty cheap...
www.courtlistener.com/docket/65407...
www.courtlistener.com/docket/65407...
Asking for a me
November 23, 2025 at 11:12 PM
Asking for a me
The argument this time seems to be “what if their buried data indicating harm wasn’t very compelling?”
Notably the indication in the reporting here is they were running a very similar study in design to their election collaboration… which doesn’t get that scrutiny.
Notably the indication in the reporting here is they were running a very similar study in design to their election collaboration… which doesn’t get that scrutiny.
November 23, 2025 at 8:38 PM
The argument this time seems to be “what if their buried data indicating harm wasn’t very compelling?”
Notably the indication in the reporting here is they were running a very similar study in design to their election collaboration… which doesn’t get that scrutiny.
Notably the indication in the reporting here is they were running a very similar study in design to their election collaboration… which doesn’t get that scrutiny.
Selectively being skeptical of undisclosed internal research is a recipe for allowing companies to dictate our understanding of harms.
November 23, 2025 at 7:05 PM
Selectively being skeptical of undisclosed internal research is a recipe for allowing companies to dictate our understanding of harms.
We have to weigh I guess that against actively misleading externally facing research from selective funding and access granted to industry friendly academics!
arxiv.org/abs/2510.19894
arxiv.org/abs/2510.19894
The Risks of Industry Influence in Tech Research
Emerging information technologies like social media, search engines, and AI can have a broad impact on public health, political institutions, social dynamics, and the natural world. It is critical to ...
arxiv.org
November 23, 2025 at 6:37 PM
We have to weigh I guess that against actively misleading externally facing research from selective funding and access granted to industry friendly academics!
arxiv.org/abs/2510.19894
arxiv.org/abs/2510.19894
I don’t know that the average effect is really the estimand of interest when the concern is about (say) vulnerable teens being exposed to stuff like eating disorder content.
November 23, 2025 at 6:34 PM
I don’t know that the average effect is really the estimand of interest when the concern is about (say) vulnerable teens being exposed to stuff like eating disorder content.
They could, of course, publish rather than bury this work.
They seem quite keen on it when their confounded analyses suggest they’ve fixed hate speech.
They seem quite keen on it when their confounded analyses suggest they’ve fixed hate speech.
November 23, 2025 at 6:30 PM
They could, of course, publish rather than bury this work.
They seem quite keen on it when their confounded analyses suggest they’ve fixed hate speech.
They seem quite keen on it when their confounded analyses suggest they’ve fixed hate speech.
