joel
LightNews LightNews
banner
jblechinger.ca
joel
@jblechinger.ca
140 followers 390 following 1.3K posts
he/him/his | views= my own | music takes | repair is the dream of the broken thing
Posts Replies Media Videos
jblechinger.ca
Good music again this year:
2025 Music Favs!!! Geese - Getting Killed Oneohtrix Point Never - Tranquilizer Nourished By Time - The Passionate Ones Oklou - choke enough Ryan Davis and The Roadhouse Band - New Threats From the Soul Friendship - Caveman Wakes Up Alex G - Headlights james K - Friend Cass McCombs - Interior Live Oak Wednesday - Bleeds Honourable Mention: Adrianne Lenker - Live at Revolution Hall Dijon - Baby Panda Bear - Sinister Grift Erika de Casier - Lifetime Destroyer - Dan’s Boogie U.e. - Hometown Girl Ribbon Skirt - Bite Down Nick León - A Tropical Entropy Addison Rae - Addison Brian Dunne - Clams Casino Emmylou Harris - Spyboy (reissue) Car Culture - Rest Here Brad Mehldau - Ride into the Sun Rafael Toral - Traveling Light Shinetiac - Infiltrating Roku City Mavis Staples - Sad and Beautiful World Bob Dylan - The Bootleg Series Vol. 18: Through the Open Window 1956–1963 Rosalía - LUX Haley Heynderickx & Max García Conover - What of Our Nature
November 24, 2025 at 6:40 PM
jblechinger.ca
I think winter really is here lol
November 23, 2025 at 5:57 PM
jblechinger.ca
Really good book so far!
The book "Why We Fear AI: On the Interpretation of Nightmares" by Hagen Blix & Ingeborg Glimmer
November 17, 2025 at 3:30 AM
jblechinger.ca
New season of "Shell Game" (www.shellgame.co) debuting next week!

Looks like he's gonna be tackling "agentic" AI within the start-up space.

Can't wait, to be honest.
Shell Game A podcast about things that are not what they seem, hosted by journalist Evan Ratliff. Season 2, which kicks off on November 12, tells the story of enterprise and entrepreneurship in the AI age. Or: how Evan tried to build a real startup, run by fake people. This season, Evan confronts head-on the assertion from tech entrepreneurs and VCs that we’re on the precipice of AI remaking the workforce. That we should expect not just AI colleagues working alongside us, but also — by Sam Altman’s telling, at least — the dawn of a billion-dollar company with only one human involved. Evan learns what it would take to create such a company … by doing it: Launching a real startup with five AI employees. Over the course of eight episodes, he explores what AI agents tell us about the work we do, the meaning we find in it, and the world that their makers say we’ll all be living in soon enough.
November 6, 2025 at 2:46 PM
jblechinger.ca
Seated for "Springsteen: Deliver Me from Nowhere" lol
October 31, 2025 at 1:33 AM
jblechinger.ca
🔥🔥🔥🔥🔥
October 30, 2025 at 3:07 AM
jblechinger.ca
It's kind of wild how much harder the first edition cover of George Will's baseball book goes than the update lol
George Will - "Men at Work" modern cover showing standard photo of baseball diamond George Will - "Men at Work" first edition cover where he is crouching in a suit against the backdrop of a baseball diamond looking incredibly swagged out
October 29, 2025 at 1:59 PM
jblechinger.ca
What if we all just collectively pretended that "Art Angels" was the last thing she was involved in? lol
"If Anyone Builds It, Everyone Dies" by Eliezer Yudkowsky & Nate Soares Dust jacket blurbs of the book from Tim Urban, Grimes, and Yishan Wong. Grimes: "Humans are lucky to have Soares and Yudkowsky in our corner, reminding us not to waste the brief window of time that we have to make decisions about our future."
October 22, 2025 at 9:49 PM
jblechinger.ca
Unsolicited/spammed (surely AI-generated?) BGPT "paper review" of a piece I wrote on AI - truly weird times, folks...
Email from BGPT: "Hi Joel, I prepared a short BGPT Paper Review: Reflections on Information Literacy in the ChatGPT Era (ranked for scientific quality, novelty, reproducibility, explanatory depth, and generality) — https://bgpt.pro/?q=Paper%20Review%3A%20Reflections%20on%20Information%20Literacy%20in%20the%20ChatGPT%20Era Novel hypothesis you might find immediately useful for classroom research: Does blending AI-generated lateral-reading prompts with human-curated reflective debriefs increase students' detection of LLM hallucinations and their acceptance of AI feedback versus human-only or AI-only instruction? (link to the encoded hypothesis and brief protocol) https://bgpt.pro/?q=Novel%20hypothesis%3A%20Does%20blending%20AI-generated%20lateral-reading%20prompts%20with%20human-curated%20reflective%20debriefs%20increase%20students%27%20detection%20of%20LLM%20hallucinations%20and%20acceptance%20of%20AI%20feedback%20vs%20human-only%20or%20AI-only%20instruction%3F Why this might be useful: it builds directly on your critique of SIFT/RADAR, connects to recent evidence that empathy/credibility cues improve student reactions to AI feedback, and yields a compact, testable intervention plus survey items (adaptable from recent AI-readiness instruments) you could run in a single semester. BGPT Team"
October 21, 2025 at 5:37 PM
jblechinger.ca
AI Chimp
Deepfakes Tracker Dashboard showing several reviewed items, one of which has the "Textual Rating" of "AI Chimp."
October 17, 2025 at 4:42 PM
jblechinger.ca
Uhhhh ... neither? lol
Some people believe AI could eventually wipe out humanity. Which statement are you more likely to agree with ... Stephen Hawking: "The development of full artificial intelligence could spell the end of the human race." Sundar Pichai: "AI is one of the most important things humanity is working on. It is more profound than ... electricity or fire."
October 16, 2025 at 10:02 PM
jblechinger.ca
Half-formed paper idea where one compares the proceduralist governance systems of liberalism to robots.txt and explores the breakdown of proceduralism both politically and in terms of norms of protocol.

Someone smarter than me please write this lol

From: ospolicyobservatory.uvic.ca/ai-bots-open...
The effectiveness of traditional protocols like robots.txt has also diminished. While this voluntary compliance system worked well with earlier generations of web crawlers, many AI training bots now ignore these requests entirely—representing a departure from established internet norms and etiquette: “The [robots.txt] protocol has not proven to be as effective in the context of bots building AI training datasets. Respondents reported that robots.txt is being ignored by many (although not necessarily all) AI scraping bots. This was widely viewed as breaking the norms of the internet, and not playing fair online” (Weinberg 2025).
October 16, 2025 at 4:30 PM
jblechinger.ca
Forget "PhD level intelligence"—my own personal measure of AGI will be ... actually competent ASCII art of Bob Dylan lol
Claude ASCII art of Bob Dylan with a harmonica Below the art it says, "The times they are a-changin'!"
October 6, 2025 at 4:06 PM
jblechinger.ca
Damn, how did wonderspencer know!?!?
wonderspencer comment on Genius page for Taylor Swift's "Wood": "thinking this one will be abt travis's wood (this is gonna be so funny when it actually comes out)
October 3, 2025 at 3:47 PM
jblechinger.ca
Checking in on what the indie/art house cinema near me is showi-
The Plaza Theatre showing Taylor Swift - "Release Part of a Showgirl" all weekend
October 3, 2025 at 1:54 PM
jblechinger.ca
Are AI books gonna look like this in 5-10 years? Discuss...
"Blogs, Wikipedia, Second Life, and Beyond" by Axel Bruns
October 2, 2025 at 4:31 PM
jblechinger.ca
This is the stretch on the Geese album for me 🔥
"Getting Killed" -> "Islands of Men" -> "100 Horses" -> "Half Real" -> "Au Pays du Cocaine"
October 1, 2025 at 7:41 PM
jblechinger.ca
BREAKING: me and my Dad love Alejandro Kirk
September 28, 2025 at 9:26 PM
jblechinger.ca
Good book day!
Mike Rothschild's "The Storm is Upon Us" and Patricia Lockwood's "Will There Ever Be Another You" on a bench
September 20, 2025 at 12:08 AM
jblechinger.ca
Oof, I felt that...
"Getting to know people and maintaining the relational aspect of librarianship is hard work. Then, once you have made some inroads, they are gone in a matter of months. I can’t think about belonging when I have to think in six-month intervals. I am just thrust into survival mode."
September 15, 2025 at 10:34 PM
jblechinger.ca
Thank you @travisview.bsky.social /"QAA" pod for the tip about this new @maargentino.com book on a recent ep!

Can't wait to dig into this one.
"QAnon: From Conspiracy Theory to New Religious Movement" by Marc-André Argentino
September 4, 2025 at 11:10 PM
jblechinger.ca
P4k's positive "Sunday Review" of "Fully Completely" making national headlines in Canada is funny lol
CBC headline: "A classic Tragically Hip Album receives retroactive recognition from Pitchfork"
September 4, 2025 at 3:47 PM
jblechinger.ca
I sure am glad we're crushing some compute so that Google AI Overview can unreliably tell me what the AQHI in Calgary is saving me ... one click over to the Environment Canada website lol
Google AI Overview feature providing an AI-generated answer of what the AQHI in Calgary is right now.
September 2, 2025 at 3:04 AM
jblechinger.ca
I think this style of YouTube content has some of the heaviest declensionist, end of empire vibes lol
YouTube thumbnail from the account Jubilee for ("1 Capitalist vs. 20 Anti-Capitalists (ft. Patrick Bet-David)"
September 1, 2025 at 9:27 PM
jblechinger.ca
I'm a neophyte to the field of "citation analysis."

Just kind of dipping my toes in, but it's funny to read stuff like this and compare with the stock reasons given for citing in library instruction (credit where credit is due, etc.)

(Excerpt from: asistdl.onlinelibrary.wiley.com/doi/10.1002/...)
"Of the 900 references studied, Brooks (1985, p. 228) found that about 70 percent were multiply motivated, concluding that “no longer can we naively assume that authors cite only noteworthy pieces in a positive manner. Authors are revealed to be advocates of their own points of view who utilize previous literature in a calculated attempt to self-justify."
September 1, 2025 at 9:01 PM