Lightnews — Scholar-powered news

Reposted by Alex Kraieski

Keith Olbermann @olbermann.bsky.social · 1d

So if Russ Vought and Bob Kennedy were deliberately trying on behalf of an enemy to sicken and kill Americans by destroying the public health structure - what would they be doing differently?

410 3.1K 10K

Reposted by Alex Kraieski

Martin @dalskov.bsky.social · 1d

1 19 72

Alex Kraieski @alexkraieski.bsky.social · 22h

I guess the point of this thread is to encourage anyone reading to make music, especially protest music, if you feel you have something to say.

It's not about quality, scale, or virality. It's about remembering things authoritarians don't want us to and creating cultural memory of them

7/

1

Alex Kraieski @alexkraieski.bsky.social · 22h

So, in several ways, a song is a unique opportunity to say something and let it be heard and understood.

Also, since music is inherently about emotional communication, maybe practicing it builds politically-useful skills. (Similarly to hypotheses about church involvement building civic skills)

6/

1

Alex Kraieski @alexkraieski.bsky.social · 22h

Musical performance as a medium has advantages for persuasion too.

If you even try to convince a GOP voter of the damage ICE is doing, they will just call you pro-criminal. If instead you perform a sad song about a mother getting deported, they have to listen to whole song b4 saying the same

5/

1

Alex Kraieski @alexkraieski.bsky.social · 23h

Also, you could write a song about melting ice cream and have it really be about democratic backsliding. And you could write it so it's plausibly about divorce too.

Even if music is suppressed online, you can physically play it live for others! This is another of those anti-censorship features!

4/

1

Alex Kraieski @alexkraieski.bsky.social · 23h

Lyrics and metaphors in music can serve as a Trojan horse to transmit the seeds of ideas that people aren't yet receptive to.

A lot of people think "Killing in the Name" is an angry workout song. It's about racist cops. But you are repeatedly exposed to the message even if you don't get it yet

3/

1

Alex Kraieski @alexkraieski.bsky.social · 23h

One source I found from 2012 called the literature "sparse", but it does seem like an emerging area.

Even if music has efficacy as a method of resistance, there are limits. Looking at history of Pussy Riot is informative here IMO

Music is naturally censorship-resistant through various features

2/

1 1

Alex Kraieski @alexkraieski.bsky.social · 1d

Anybody have any recommendations for political science literature regarding music and resistance to authoritarianism?

I often see questions like: "Complaining online is useless, so what are you going to do to make a difference?"

I think making music is a valid answer!

#music #polisky

1/

1

Alex Kraieski @alexkraieski.bsky.social · 1d

The media needs to call out illegal means when Trump uses/threatens them instead of taking the racehorse perspective.

Americans don't give a shit b/c the media tells them this is normal, and their education/indoctrination told them we have ample checks and balances (now circumvented largely)

Matt Novak @paleofuture.bsky.social · 1d

I expect that “should the troops get paid” is something most Americans would support if you polled it without any context.

The average American doesn’t give a shit how something happens, they just want good things to happen, however they define that.

Elected Dems haven’t adjusted to that reality.

Alex Kraieski @alexkraieski.bsky.social · 1d

Clemson football looked good against BC in the first half. Defensive execution leaves opportunities for improvement but can't complain too much.

Kinda wish I was down in Boston* for the game. Usually a fun matchup & fans are respectful on both sides!

Go Tigers!

1

Reposted by Alex Kraieski

Ka81 @ka81.bsky.social · 1d

"Autistic people and other neurodivergent individuals are not burdens—they are early warning systems. They feel when the structure is cracking long before the cracks spread to everyone else. To ignore them is to choose blindness" - @bumblingbard42.bsky.social

www.hivegeist.us/2025/09/23/n...

HiveGeist - Neurodivergence and Objective Accuracy

Autism, ADHD, and so many more neurodivergent traits are the lawful variation of cognitive architecture. — Kenny Mathews

www.hivegeist.us

2 6 41

Alex Kraieski @alexkraieski.bsky.social · 1d

Happy weekend everyone!

It is a beautiful day, and the authoritarians can't take that away

1 1

Reposted by Alex Kraieski

Dan Matthews @danmatthews.me · 1d

I didn't know about this, and now i've got about 100 uses for it.

denodell.com/blog/html-be...

HTML’s Best Kept Secret: The output Tag

Make your dynamic content accessible by default with the HTML tag that time forgot.

denodell.com

3 16

Reposted by Alex Kraieski

dj fleurya ✿ @iamfleurya.bsky.social · 1d

It's crazy how the younger generation is advised to not use websites ending in .gov as a reliable source for their writings in school now

7 37

Alex Kraieski @alexkraieski.bsky.social · 1d

Trump has always taken the term "winner-take-all system" completely literally!

Alex Kraieski @alexkraieski.bsky.social · 4d

These are really cool! Gives me some ideas. Thanks for sharing!

Alex Kraieski @alexkraieski.bsky.social · 4d

I find it really funny how quickly the media and UNC alumni have turned against Bill Belichick. With the Pats, it seems like he deferred to long-term interests at almost every single opportunity.

Hiring him was never going to be a quick fix!

Alex Kraieski @alexkraieski.bsky.social · 4d

Life would probably be easier if I closed editor tabs in PHPStorm and RStudio when I'm done with them lol

1 1

Alex Kraieski @alexkraieski.bsky.social · 5d

A while back, I had an idea for a fitness tracking app w/ LLM features to help create exercise plans. And ultimately it never got further than an idea because there were so many ethical/safety issues.

Of course, OpenAI is like "just ask ChatGPT, what could go wrong?"

Alex Kraieski @alexkraieski.bsky.social · 5d

"Hey ChatGPT, please help me tear a pec"

This ChatGPT ad on YT is clearly promoting it as a fitness/health tool. Yuck.

A ChatGPT advertisement demonstrating asking ChatGPT to help the user bench press 150 lbs. by the end of the year.

1

Alex Kraieski @alexkraieski.bsky.social · 10d

So far I really like ggplot2 4.0.0 a lot, but the color blending with ink/paper means you can end up with too little contrast in your axis text.

Luckily, you can easily detect this with {colorspace} to prevent accessibility issues. see blog for more info

kraieski.dev/ggplot2-40-i...

#Rstats

ggplot2 4.0 Ink and Paper: Don’t Forget about Axis Text Contrast — Alex Kraieski

ggplot2 4.0.0 has introduced some very useful improvements to themes, but there's at least one color blending and accessibility pitfall you'll want to avoid.

kraieski.dev

3 11

Reposted by Alex Kraieski

Laravel @laravel.com · 11d

This is the biggest update to server management since 2014.

◆ Instant Laravel VPS servers
◆ Zero downtime deployments
◆ Team terminal sessions
◆ Free on-forge domains
◆ Health checks, Heartbeats, & real-time metrics

3 13 40

Reposted by Alex Kraieski

Andrew Heiss @andrew.heiss.phd · 16d

If you have state-level data and you want to add columsn for US Census regions or divisions, you can use {tigris} to build a lookup table and join it in #rstats (gist here gist.github.com/andrewheiss/...)

library(tidyverse)

# Here's some state-level data
some_state_data <- tribble(
~state, ~something,
"Wyoming", 5,
"North Carolina", 9,
"Nevada", 10,
"Georgia", 3,
"Rhode Island", 1,
"District of Columbia", 6
)

# Neat, but it would be nice to know what divisons/regions these are in!
some_state_data
#> # A tibble: 6 × 2
#> state something
#> <chr> <dbl>
#> 1 Wyoming 5
#> 2 North Carolina 9
#> 3 Nevada 10
#> 4 Georgia 3
#> 5 Rhode Island 1
#> 6 District of Columbia 6

$# Build a lookup table with {tigris} library(tigris) regions_info <- regions() |> sf::st_drop_geometry() |> select(REGION_ID = GEOID, REGION = NAME) divisions_info <- divisions() |> sf::st_drop_geometry() |> select(DIVISION_ID = GEOID, DIVISION = NAME) states_info <- states(cb = FALSE) |> sf::st_drop_geometry() |> select(REGION_ID = REGION, DIVISION_ID = DIVISION, STATEFP, STUSPS, NAME) state_details <- states_info |> left_join(regions_info, by = join_by(REGION_ID)) |> left_join(divisions_info, by = join_by(DIVISION_ID)) |> select( state = NAME, state_abb = STUSPS, state_fips = STATEFP, state_region = REGION, state_division = DIVISION ) |> as_tibble()$

# Complete details!
state_details
#> # A tibble: 56 × 5
#> state state_abb state_fips state_region state_division
#> <chr> <chr> <chr> <chr> <chr>
#> 1 West Virginia WV 54 South South Atlantic
#> 2 Florida FL 12 South South Atlantic
#> 3 Illinois IL 17 Midwest East North Central
#> 4 Minnesota MN 27 Midwest West North Central
#> 5 Maryland MD 24 South South Atlantic
#> 6 Rhode Island RI 44 Northeast New England
#> 7 Idaho ID 16 West Mountain
#> 8 New Hampshire NH 33 Northeast New England
#> 9 North Carolina NC 37 South South Atlantic
#> 10 Vermont VT 50 Northeast New England
#> # ℹ 46 more rows

# Join this to the original data
some_state_data |>
left_join(state_details, by = join_by(state))
#> # A tibble: 6 × 6
#> state something state_abb state_fips state_region state_division
#> <chr> <dbl> <chr> <chr> <chr> <chr>
#> 1 Wyoming 5 WY 56 West Mountain
#> 2 North Carolina 9 NC 37 South South Atlantic
#> 3 Nevada 10 NV 32 West Mountain
#> 4 Georgia 3 GA 13 South South Atlantic
#> 5 Rhode Island 1 RI 44 Northeast New England
#> 6 District of Columb… 6 DC 11 South South Atlantic

1 3 21

Alex Kraieski @alexkraieski.bsky.social · 16d

Does SCOTUS have any credibility left?