Archimedes

Karma: 1,029

Archimedes 2 Feb 2026 4:56 UTC
2 points
0
on: What would it mean for the Myers-Briggs personality test to be pseudoscientific?
Somewhat serendipitously, Spencer Greenberg just released a Clearer Thinking podcast episode interviewing Colin DeYoung on this topic just a few days ago. Worth a listen, IMO.

https://podcast.clearerthinking.org/episode/298/colin-deyoung-are-personality-types-a-statistical-mirage/

Archimedes 26 Jan 2026 0:25 UTC
2 points
0
in reply to: Jiro’s comment on: What’s a good methodology for “is Trump unusual about executive overreach / institution erosion?”

the easiest way for someone to be better on the measure “reactions to X” is not to do X

Yes? Just like the easiest way for someone to avoid criminal convictions is to not do crimes. Not a guarantee but it helps.

Ultimately, any metric does need to be contextualized and interpreted. Supposing metrics can be agreed upon, it’s another big step to cross the is-ought gap, even with a shared moral framework.

I agree with another commenter that the original post is a good candidate for adversarial collaboration.

Archimedes 25 Jan 2026 6:14 UTC
3 points
0
in reply to: Jiro’s comment on: What’s a good methodology for “is Trump unusual about executive overreach / institution erosion?”
Yeah, I can see that. Unfortunately, “how much should we be scared of him” doesn’t have direct metrics, so we have to come up with proxies of some kind.

Do you have any suggestions that don’t feel cherry-picked? What we likely need is input from some good historians. I don’t think I can escape my recency bias well enough.

Archimedes 25 Jan 2026 3:17 UTC
5 points
0
on: What’s a good methodology for “is Trump unusual about executive overreach / institution erosion?”
Potential metrics
1. Court-based metrics
  - Federal injunctions against executive actions per term
  - Executive actions struck down as unconstitutional
  - Instances of defying/ignoring court orders (distinct from appealing)
2. Structural/institutional metrics
  - Inspector General firings
  - Career civil servant resignation rates
  - Acting officials vs. Senate-confirmed appointees ratio
These metrics may not be available farther into the past.

Archimedes 25 Jan 2026 2:55 UTC
2 points
0
in reply to: Arjun Panickssery’s comment on: Strong, bipartisan leadership for resistance to Trump.
I’m unclear what institution is directly implicated in the Afghanistan withdrawal, so I don’t have an opinion on whether it qualifies as erosion or not.

Abolishing ICE would be taking down an institution, I suppose. I take (what I assume is your implicit) point that preserving an institution isn’t per se virtuous. Perhaps ICE is beyond reform and would need to be abolished and replaced instead, and perhaps the same is true of the institutions Trump is attacking?

Archimedes 24 Jan 2026 16:45 UTC
3 points
0
in reply to: Arjun Panickssery’s comment on: Strong, bipartisan leadership for resistance to Trump.

A few of these are, if somewhat unprecedented, not really institutional erosion, because they have a legitimate constitutional basis.

These are not mutually exclusive.

The qualification of appointments is just opinion and not institutional erosion.

I disagree. If you appoint a leader of an institution with no experience related to that institution (Rex Tillerson: Secretary of State, Dr. Mehmet Oz: CMS Administrator), or who believes that institution doesn’t exist (Rick Perry: Department of Energy, Mick Mulvaney: CFPB, Betsy DeVos: Department of Education), they are highly likely to erode that institution, whether intentionally or through incompetence.

Not all appointments require extensive experience within that institution, but some positions should have deep domain expertise, unless their purpose is to serve as a wrecking ball appointment.

Archimedes 13 Jan 2026 2:41 UTC
5 points
0
in reply to: lilkim2025’s comment on: Strong, bipartisan leadership for resistance to Trump.
Thanks for the perspective. I would like to note that pointing out earlier instances of bad behavior is important historical context, but doesn’t make more recent bad behavior any less bad. The USA has some really dark stuff in its past. We should remember and do better.

Archimedes 13 Jan 2026 2:30 UTC
6 points
6
in reply to: No77e’s comment on: Strong, bipartisan leadership for resistance to Trump.
One major category of concern is institutional erosion. Here are a few salient examples:
1. Mass IG firing (Jan 2025): Fired 17 inspectors general in one night without the legally required 30-day notice. Federal court ruled this “obviously” illegal. Several were investigating Musk’s companies. (IGs are the internal watchdogs that catch waste, fraud, and abuse.)
2. USAID dismantling: Shut down the agency that administered most U.S. foreign aid, firing 83% of staff. This destroyed implementation capacity built over decades. A Lancet study estimates 14 million excess deaths by 2030 from the cuts.
3. Schedule F: Reclassifying ~50,000 career civil servants to strip job protections, enabling political firings. This converts expertise-based positions into loyalty-based ones.
4. Firing independent agency heads: Removed FTC commissioners, NLRB members, MSPB chair. DOJ sent a letter to Congress stating it would no longer defend the constitutionality of for-cause removal protections—explicitly signaling intent to overturn Humphrey’s Executor, the 1935 precedent that lets Congress create agencies insulated from presidential control.
5. Unqualified appointments:
  - Robert F. Kennedy Jr. (HHS): Longtime anti-vaccine activist with no public health administration experience. His department’s official MAHA report cited nonexistent studies, then introduced additional errors when revised.
  - Pete Hegseth (Defense): Fox News weekend host with no Pentagon or senior military experience. Participated in the Signalgate breach, sharing military operation timing on an unsecured group chat that accidentally included a journalist.
  - Kash Patel (FBI): No law enforcement leadership experience. Authored a book, Government Gangsters, containing a list of government officials he considers enemies of Trump, and has stated publicly he intends to pursue them.
  - Tulsi Gabbard (DNI): Former congresswoman with no intelligence community experience. Sen. Dick Durbin said she “would not be qualified for an entry-level position” at an intelligence agency.
  - Betsy DeVos (Education, 2017): Billionaire Republican donor with no education policy experience. First Cabinet nominee in history to require the Vice President to break a tie vote for confirmation.
This is a small sampling; a comprehensive account would be far longer.

Archimedes 13 Jan 2026 1:42 UTC
7 points
5
in reply to: Richard_Ngo’s comment on: Strong, bipartisan leadership for resistance to Trump.

For example, what kind of coalition would be able to actually update that Trump is less bad than it thinks if your current fears don’t come to pass?

The things that have already come to pass are already deeply damaging to our institutions. We are not in 2016 when we could only extrapolate how he would behave as president. We have a decade of evidence since then. He may not play out every single thing people fear, but, like, if he doesn’t end up trying to acquire Greenland by force, that’s not much of an update. Not everything worth worrying about comes to pass. It’s absurd that such a thing is plausible for him, even if it’s not probable.

Archimedes 11 Jan 2026 6:46 UTC
4 points
0
in reply to: Hastings’s comment on: Hastings’s Shortform
My first thought was Amazon leveraging this for drone delivery.

Archimedes 10 Jan 2026 7:26 UTC
5 points
0
on: What do people mean by “recursive self-improvement”?
It’s more of a continuum than that. For example, some people argue that Claude Code writing most of the code that goes into improving Claude Code is an early form of partial RSI. Clearly, the loop is incomplete, but Claude is accelerating his own advancement in a real way. How far out of the loop do humans need to be for it to fully count?

Archimedes 10 Jan 2026 6:53 UTC
8 points
1
in reply to: Karl Krueger’s comment on: What do we mean by “impossible”?
How about these?
1. A dense ball of electrons with no confinement or a blob of molten tungsten not exchanging energy in a bath of liquid helium.
2. A spacecraft indefinitely “parked” at an unstable equilibrium (L1/L2 Lagrange points) without any course corrections.
3. Revive Julius Caesar exactly as he was, including his memories.
4. Land a rocket upright on a 1-meter target on Mars with zero controls/sensors/guidance. Everything depends on perfect initial calculations.

Archimedes 6 Jan 2026 5:15 UTC
2 points
0
in reply to: anaguma’s comment on: Continual Learning Achieved?

Seems plausible, but I would put this under the category of not having fully solved continual learning yet. A sufficiently capable continual learning agent should be able to do its own maintenance, short of a hardware failure.

Depending on how it’s achieved, it might not be a matter of maintenance/hardware failure so much as compute capacity. Imagine if continual learning takes similar resources to standard pretraining of a large model. Then they could continually train their own set of models, but it wouldn’t be feasible for everyone to get their own version that continually learns what they want it to.

Archimedes 6 Jan 2026 4:54 UTC
3 points
4
in reply to: Gordon Seidoh Worley’s comment on: Claude Wrote Me a 400-Commit RSS Reader App
I consider Claude’s “taste” to be pretty good, usually, but not P90 of humans with domain experience. I’d characterize his deficiencies more along the lines of a lack of ability to do long-term “steering” at a human level. This is likely related to a lack of long-term memory and hence the ability to do continual learning.

Archimedes 6 Jan 2026 4:36 UTC
1 point
0
in reply to: Hastings’s comment on: Claude Wrote Me a 400-Commit RSS Reader App
Is this sufficient? I don’t really know the best place to put a disclosure.

https://en.wikipedia.org/wiki/User_talk:Alexis0Olson/Multilayer_perceptron#LLM_Disclosure

Archimedes 5 Jan 2026 4:22 UTC
5 points
0
on: Claude Wrote Me a 400-Commit RSS Reader App
Claude Code is excellent these days and meets my bar for “AGI”. It’s capable of doing serious amounts of cognitive labor and doesn’t get tired (though I did repeatedly hit my limits on the $20 plan and have to wait through the 5-hour cooldown).

I spent a good chunk of this weekend seeing if I could get Claude to write a good Wikipedia article if I told it to use the site’s rules and guidelines and then let it iteratively critique and revise against those guidelines until the article fully met the standards. I wrote zero of the text myself, though I did paste some Q&A back and forth to NotebookLM to help with citations and had ChatGPT generate an additional flowchart visual to include.

After getting some second opinions from Gemini and ChatGPT, I will have Claude do a final round of revisions and then actually try to get it on Wikipedia. I will share the link here if it gets accepted—I don’t really know how that works, but I bet Claude can help me figure it out.

Archimedes 1 Jan 2026 20:11 UTC
1 point
0
in reply to: Dave Orr’s comment on: 2025 Year in Review
Would the default valence be the valence of the “thing”?

Archimedes 24 Dec 2025 5:19 UTC
2 points
0
in reply to: lilkim2025’s comment on: Dancing in a World of Horseradish
My hypothesis for the airline industry boils down to “commodification”. Airline companies follow incentives, and competition on price is fierce. Customers have little brand loyalty and chase the cheapest tickets, except occasionally avoiding the truly minimalist airlines. The companies see the customers voting with their wallets and optimize accordingly, leading to a race to the bottom.

In my experience, non-US carriers aren’t that different. Maybe just a bit further behind and a bit more resistant to the slippery slope toward enshitification.

Archimedes 5 Dec 2025 3:31 UTC
1 point
0
on: Archimedes’s Shortform
Anthropic is currently running an automated interview “to better understand how people envision AI’s role in their lives and work”. I’d encourage Claude users to participate if you want Anthropic to hear your perspective.

Access it directly here (unless you’ve just recently signed up): https://claude.ai/interviewer

See here for Anthropic’s post about it here: https://www.anthropic.com/research/anthropic-interviewer

Archimedes 22 Nov 2025 5:25 UTC
3 points
4
on: Infinitesimally False
supremum: the least value which is greater than all the values in the set
Should be “greater than or equal to all the values in the set” or a closed interval like [0,1] has no supremum.