Skip to primary content

Statistical Modeling, Causal Inference, and Social Science

Search

Main menu

Home
Authors
Blogs We Read
Books
Sponsors

Post navigation

Chess ratings: some thoughts on the Glicko system 2006 retrospective

Some fun number crunching of words

Posted on December 29, 2006 1:03 AM by Andrew

from Mark Liberman here.

This entry was posted in Literature by Andrew. Bookmark the permalink.

Art
Artificial Intelligence
Bayesian Statistics
Causal Inference
Decision Analysis
Economics
Jobs
Literature
Miscellaneous Science
Miscellaneous Statistics
Multilevel Modeling
Obituaries
Papers
Political Science
Public Health
Sociology
Sports
Stan
Statistical Computing
Statistical Graphics
Teaching
Zombies

AAAnonymous on He “washed his hands in a can of tetraethyl lead at a press conference, claiming he was ‘not taking any chance whatever’. He knew this to be a lie, having already succumbed to a bout of lead poisoning.”July 22, 2026 1:13 PM
Quote from above: "If only there was language to describe a potentially large and complicated set of policies and analyze…
Anoneuoid on He “washed his hands in a can of tetraethyl lead at a press conference, claiming he was ‘not taking any chance whatever’. He knew this to be a lie, having already succumbed to a bout of lead poisoning.”July 22, 2026 12:58 PM
Yea that wikipedia page is also very weaselly. He performed the stunt then took a leave of absence some unspecified…
somebody on He “washed his hands in a can of tetraethyl lead at a press conference, claiming he was ‘not taking any chance whatever’. He knew this to be a lie, having already succumbed to a bout of lead poisoning.”July 22, 2026 12:50 PM
According to wikipedia On October 30, 1924, Midgley participated in a press conference to demonstrate the apparent safety of TEL,…
Tris Simondsen on Forking paths in LLMs for data analysisJuly 22, 2026 12:50 PM
David, Daniel, I am reminded of Mickey Day's SNL sketches "Terzano's" and "Domenico" where you being "a little confused" actually…
Proxomitron on He “washed his hands in a can of tetraethyl lead at a press conference, claiming he was ‘not taking any chance whatever’. He knew this to be a lie, having already succumbed to a bout of lead poisoning.”July 22, 2026 11:52 AM
If he was exposed to real danger, it makes him a bad person, for harming many others (even if he…
AAAnonymous on He “washed his hands in a can of tetraethyl lead at a press conference, claiming he was ‘not taking any chance whatever’. He knew this to be a lie, having already succumbed to a bout of lead poisoning.”July 22, 2026 11:48 AM
Quote from above: "Like maybe, implement some kind of screening processes and background checks?" I am not an academic, so…
Anoneuoid on He “washed his hands in a can of tetraethyl lead at a press conference, claiming he was ‘not taking any chance whatever’. He knew this to be a lie, having already succumbed to a bout of lead poisoning.”July 22, 2026 11:47 AM
These "stunts" (actually forcing people to dogfood their own chemicals is a good idea) are a good idea and seems…
Proxomitron on He “washed his hands in a can of tetraethyl lead at a press conference, claiming he was ‘not taking any chance whatever’. He knew this to be a lie, having already succumbed to a bout of lead poisoning.”July 22, 2026 11:25 AM
Leaded gasoline is very bad for everybody, and it was know to Midgley at the time. By performing his stunt…
somebody on He “washed his hands in a can of tetraethyl lead at a press conference, claiming he was ‘not taking any chance whatever’. He knew this to be a lie, having already succumbed to a bout of lead poisoning.”July 22, 2026 11:07 AM
I must say I think it’s refreshing to hear a bit more of this on this blog in recent months…
Anoneuoid on He “washed his hands in a can of tetraethyl lead at a press conference, claiming he was ‘not taking any chance whatever’. He knew this to be a lie, having already succumbed to a bout of lead poisoning.”July 22, 2026 10:24 AM
The ill effects of ingested lead [...] washed his hands in a can of TEL at a press conference, claiming…
Daniel Lakeland on Forking paths in LLMs for data analysisJuly 22, 2026 10:14 AM
Tris, yes I understand the theory youre arguing for, I think... You want the chatbot to be prevented ahead of…
AAAnonymous on He “washed his hands in a can of tetraethyl lead at a press conference, claiming he was ‘not taking any chance whatever’. He knew this to be a lie, having already succumbed to a bout of lead poisoning.”July 22, 2026 9:47 AM
Quote from the blog post: "I guess that the world is full of evil people and always will be. The…
Dale Lehman on He “washed his hands in a can of tetraethyl lead at a press conference, claiming he was ‘not taking any chance whatever’. He knew this to be a lie, having already succumbed to a bout of lead poisoning.”July 22, 2026 9:21 AM
Do you think that the way our political system works (start with campaign financing and then move on) increases the…
David, a Bostonian in Tokyo on Forking paths in LLMs for data analysisJuly 22, 2026 5:14 AM
Tris; Even I can't figure out what you are talking about. (Aside: I strongly recommend Jerry Fodor's book "In Critical…
Tris Simondsen on Forking paths in LLMs for data analysisJuly 22, 2026 3:54 AM
Daniel, this is the difference between a physical consequence and an epistemic boundary. When Mad Mike crashed, gravity was the…
Tris Simondsen on Forking paths in LLMs for data analysisJuly 22, 2026 3:08 AM
Anonymous, ah, the response looked too refined and academic for a public forum. I have actually written a few posts…
Seth Finkelstein on Ted-talking University of California professor asks sex trafficker for $3,000,000 because he thinks there’s a “50% chance” he’ll make “important discoveries” in telepathyJuly 22, 2026 2:13 AM
Phil: Without commenting on the larger issues, the paragraphs made sense to me. It might help to know there's a…
John Quiggin on Ted-talking University of California professor asks sex trafficker for $3,000,000 because he thinks there’s a “50% chance” he’ll make “important discoveries” in telepathyJuly 22, 2026 12:41 AM
I was just thinking about whether belief in religious miracles predisposes people to false belief more generally. Probably not for…
Daniel Lakeland on Forking paths in LLMs for data analysisJuly 22, 2026 12:14 AM
Extremely relevant headline from today! OpenAI autonomous bots hacked the HuggingFace site and stole data from HuggingFace https://www.engadget.com/2220436/openai-admits-models-hacked-hugging-face-on-their-own/ As Christine…
Anonymous on Forking paths in LLMs for data analysisJuly 21, 2026 11:01 PM
Tris, your most recent comment in this thread looks like it was written by AI. The list format and then…
Tris Simondsen on Forking paths in LLMs for data analysisJuly 21, 2026 10:31 PM
Joshua, that is a crucial clarification, and it highlights exactly how these systems can be used safely. You are engaging…
Jessica Hullman on Forking paths in LLMs for data analysisJuly 21, 2026 9:25 PM
Anon - This kind of thing is discussed in my more recent posts on AI reviewing and multiverse analysis: https://statmodeling.stat.columbia.edu/2026/06/24/treating-ai-review-like-the-contentious-policy-design-problem-it-is/…
Joshua on Forking paths in LLMs for data analysisJuly 21, 2026 7:51 PM
Tris - I just wanted to narrow focus to one point you made. Above I said I can use an…
Daniel Lakeland on Forking paths in LLMs for data analysisJuly 21, 2026 7:51 PM
Tris, there was a guy who wanted to prove the earth was flat by shooting himself into the air with…
Joshua on Forking paths in LLMs for data analysisJuly 21, 2026 7:33 PM
Bob - Reading this from you reminds me of what I wanted to say in response to David's exchange with…
Tris Simondsen on Forking paths in LLMs for data analysisJuly 21, 2026 6:43 PM
Dale, Bob, I share your hope that these systems never gain unconstrained autonomy but relying on "accidental failure" (hallucinations and…
Daniel Lakeland on Forking paths in LLMs for data analysisJuly 21, 2026 6:38 PM
When I have a video conference with Bob Carpenter and we discuss how we're going to analyze some data and…
Olaf Zimmermann on Was this USDA survey really “redundant, costly, politicized, and extraneous”?July 21, 2026 6:35 PM
It is the hallmark of any insecure government that it tries to keep its citizens ill-informed and insecure. (cf. Bertrand…
Tris Simondsen on Forking paths in LLMs for data analysisJuly 21, 2026 6:32 PM
Bob, invoking Quine and Rorty to debate the epistemic boundaries of current LLMs is a fascinating philosophical exercise, but it…
Richard Hahn on “A medical journal says the case reports it has published for 25 years are, in fact, fiction”July 21, 2026 6:04 PM
Wow, what timing! I was researching the idea of "reverse translation" just last night, where clinical findings are used to…
Bob Carpenter on Forking paths in LLMs for data analysisJuly 21, 2026 5:26 PM
Agentic LLMs are not closed boxes. They can go on the internet and search, they can look at images, they…
Dale Lehman on Forking paths in LLMs for data analysisJuly 21, 2026 5:23 PM
Who wants autonomous agents? I know there are people who want to sell such things and some of them are…
Tris Simondsen on Forking paths in LLMs for data analysisJuly 21, 2026 5:09 PM
Dale, if we are strictly treating the system as a specialized search engine or a high-powered text calculator where you…
Bob Carpenter on Forking paths in LLMs for data analysisJuly 21, 2026 5:06 PM
If "epistemic grounding" means "thought by a person", then of course, it follows that chatbots, like any other "thinking" machine,…
Tris Simondsen on Forking paths in LLMs for data analysisJuly 21, 2026 4:57 PM
Joshua, isolating the exact difference between coherence (two systems or agents reaching consensus) and correspondence (a system grounding its conclusions…
Dale Lehman on Forking paths in LLMs for data analysisJuly 21, 2026 4:52 PM
Tris You may be right, as this is not really my expertise. But my gut reaction is that you are…
Dale Lehman on Was this USDA survey really “redundant, costly, politicized, and extraneous”?July 21, 2026 4:47 PM
And a somewhat related news item (today): https://apnews.com/article/eeoc-discrimination-dei-andrea-lucas-2f521d784a9d0376416499066dc9f178. Don't collect the data, declare the reality you want, what's the problem?…
Tris Simondsen on Forking paths in LLMs for data analysisJuly 21, 2026 4:46 PM
Dale, information architecture is precisely the central divide in AI alignment right now. Relying on the human to provide the…
Andrew on Was this USDA survey really “redundant, costly, politicized, and extraneous”?July 21, 2026 4:45 PM
Olaf: Some of the racism and extreme corruption in the current U.S. government reminds me of a milder version of…
Joshua on Forking paths in LLMs for data analysisJuly 21, 2026 4:21 PM
David (and Tris) - On "useless for communicating with real humans" - I've found the opposite, for what it's worth.…
Olaf Zimmermann on Was this USDA survey really “redundant, costly, politicized, and extraneous”?July 21, 2026 4:15 PM
I hereby formally apologise for having, however inadvertently, nay, innocently, confirmed Godwin's Law. It shan't happen again.
Joshua on Forking paths in LLMs for data analysisJuly 21, 2026 4:11 PM
Tris - I see an intersting way these ideas circule back around. When I read your "jargon" I see a…
Raphael K on Was this USDA survey really “redundant, costly, politicized, and extraneous”?July 21, 2026 4:05 PM
@Paul Alper: If you invoke this Adolf Hitler because you were alive during his day and age, then I will…
Dale Lehman on Forking paths in LLMs for data analysisJuly 21, 2026 3:06 PM
Daniel I can't resist (but I'll stop because we have reached garbage time with this comment). What worries me is…
Olaf Zimmermann on Was this USDA survey really “redundant, costly, politicized, and extraneous”?July 21, 2026 2:59 PM
It's an old joke: The first astronaut to make it to outer space radios down that he has seen God.…
Andrew on Was this USDA survey really “redundant, costly, politicized, and extraneous”?July 21, 2026 2:46 PM
Olaf: Funny how you refer to God as "she." Most of the published testimony seems to suggest that God's a…
Olaf Zimmermann on Was this USDA survey really “redundant, costly, politicized, and extraneous”?July 21, 2026 2:24 PM
Sorry, I meant Paul, of course (if there was a God, why did she give me myopia, astigmatism, and latterly…
Daniel Lakeland on Forking paths in LLMs for data analysisJuly 21, 2026 1:43 PM
Dale, it's funny and not surprising to me that after pushing back so much about how you're not worried about…
Andrew on Was this USDA survey really “redundant, costly, politicized, and extraneous”?July 21, 2026 1:30 PM
John: I'm not surprised, but I remain horrified. Similar to how I feel when I read about Soviet propaganda. I…
Dale Lehman on Forking paths in LLMs for data analysisJuly 21, 2026 1:29 PM
Tris "In human reasoning, as you point out, the effort required to maintain coherence acts as a verifiable signal because…

Proudly powered by WordPress