HNHacker News
TopNewBestAskShowJobs

Jimmc414

16,226 karma · joined April 9, 2019

x.com/JimMcM4 github.com/jimmc414
submissionscomments
Jimmc414··on Alabama set to execute man who didn't kill anyone
Update: His sentence was commuted to life.

https://www.cbs42.com/alabama-news/alabama-gov-kay-ivey-make...

Jimmc414··on Homo Ignorans: Deliberately Choosing Not to Know
The framing of the whole paper makes the point implicitly. They open with Aristotle’s claim that all humans by nature desire to know, then argue the converse that choosing not to know is equally a part of human nature.

The examples they use span medical patients, Nobel laureates, lawyers, investors, and ordinary citizens across multiple countries. The paper treats it as a species level cognitive phenomenon.

Jimmc414··on Homo Ignorans: Deliberately Choosing Not to Know
The paper describes deliberate ignorance as a universal human tendency, not a partisan one. Reducing it to a conservative specific problem is ironically a good demonstration of the phenomenon.
Jimmc414··on Put the zip code first
This is accurate. For a scenario with a possibility of litigation you must ultimately geocode the address with google maps API or census geocoder, point in polygon against district boundaries (geopandas or shapely), then pass the result through a rules table keyed on jurisdiction + case type.
Jimmc414··on Homo Ignorans: Deliberately Choosing Not to Know
Follow up paper that extends the concept of identity protective cognition explaining that people avoid, neglect, and distort information to protect identity based belief systems.

The Good, Bad and Ugly of information (un)processing; Homo Economicus, Homo Heuristicus and Homo Ignorans

https://www.sciencedirect.com/science/article/abs/pii/S01674...

Jimmc414··on Homo Ignorans: Deliberately Choosing Not to Know
Link to paper:

http://www.spp1516.de/Publikationen/pdfs/Hertwig%20&%20Engel...

Jimmc414··on Lawmakers Want DoD Investigated for Biblical 'Armageddon' Claims
https://archive.is/OQurU
Jimmc414··on Millennium Challenge 2002: Persian Gulf War Game Exercise
related:

U.S. Joint Forces Command Millennium Challenge 2002: Experiment Report

https://www.esd.whs.mil/Portals/54/Documents/FOID/Reading%20...

The lost lesson of Millennium Challenge 2002, the Pentagon’s embarrassing post-9/11 war game

https://taskandpurpose.com/opinion/millenium-challenge-2002-...

War game was fixed to ensure American victory, claims general

https://www.theguardian.com/world/2002/aug/21/usa.julianborg...

Jimmc414··on Ask HN: What are your go to sources for relatively unbiased global news?
It's hard for me to reconcile with the NYT after their role in 2002-2003 of building public consensus for the Iraq war with front page stories that turned out to less than credible.

Right now, my main news sources are an amalgam of Reuters, BBC World Service, Al Jazeera English, Financial Times, NHK World / Channel News Asia.

No source is perfect and think it is a wise approach to even look at obviously biased sources with the understanding of what they are and see where the propaganda diverges and what the message is they want you to take from it.

Reading any one of these alone still gives you a skewed picture. The value is in the combination with an understanding of the motivations behind the voice you are hearing.

Jimmc414··on US President struggles to explain why he launched another Middle Eastern war
https://archive.is/lYbzQ
Jimmc414··on We Will Not Be Divided
What else can they do? Would you recommend they stay silent? It sounds like they are no longer the gatekeepers of this technology or they never were to begin with.
Jimmc414··on AIs can't stop recommending nuclear strikes in war game simulations
https://archive.is/2JAkM
Jimmc414··on The Pentagon Feuding with an AI Company Is a Bad Sign
https://archive.is/lvViA
Jimmc414··on No Skill. No Taste
I respect the feelings behind the post and I agree with a large part of it. I’m inclined to disagree on a few points made. The core problem is outsiders without taste are showing up in a space where there is a long history of dues paid by the current occupants. But how is taste developed? It’s not innate, unfortunately it’s a product of the long ugly process you are currently witnessing. Think back to the first program you were proud of and judge it with today’s eyes.
Jimmc414··on Testing Ads in ChatGPT
Not really. I’ve had YT premium for years and it removes the obtrusive ones, but I see ads in some form every time I use it.
Jimmc414··on Testing Ads in ChatGPT
Really don’t feel comfortable defending OpenAI in any way because there is a lot to complain about but ads for paid services that cost a lot more than $8 per month is not really an anomaly. Look at Amazon prime, prime video, Hulu, airline flights, any major newspaper subscription, YT premium, etc, etc. I get the annoyance but just cancel the service if you don’t want ads or pay for a tier that isn’t subsidized.
Jimmc414··on Testing Ads in ChatGPT
"The test will be for logged-in adult users on the Free and Go subscription tiers. Plus, Pro, Business, Enterprise, and Education tiers will not have ads."
Jimmc414··on Claude Opus 4.6
It might be wrong but that’s not really a hallucination.

Edit: to give you the benefit of doubt, it probably depends on whether the answer was a definitive “this does not exist” or “I couldn’t find it and it may not exist”

Jimmc414··on Clawdbot - open source personal AI assistant
I’ve installed and tested Clawdbot twice and uninstalled it. I see no reason to use this unless it’s with local models. I can do everything Clawdbot can do with Claude Code innately and with less tokens. I found Clawdbot to be rather token inefficient even with Claude max subscription. 14k tokens just to initialize and another 1000 per interaction round even with short questions like, “Hey”. Another concern is there are no guarantees that Anthropic isn’t going to lock down Oauth usage with your Max account like they did with OpenCode.
Jimmc414··on Claude Code's new hidden feature: Swarms
Their roles are defined by prompts. Only memory are shared files and the conversation history that’s looped back to stateless API calls to an LLM.
Jimmc414··on Show HN: Ocrbase – pdf → .md/.json document OCR and structured extraction API
I fought with Tesseract for quite a while. Its good if high accuracy doesn't matter. Transcribing a book from clean, consistent non-skewed data its fine and an LLM might even be able to clean it up. But for legal or accounting data from hand scanned documents, the error rate made it untenable. Even clean, scanned documents of the same category have all sorts of density and skew anomalies that get misinterpreted. You'll pull your hair out trying to account for edge cases and never get the results you need even with numerous adjustments and model retraining on errors.

Flash 2.5 or 3 with thinking gave the best results.

Jimmc414··on FUSE is All You Need – Giving agents access to anything via filesystems
I put together a spec for this where the entire LLM agent landscape adheres to the "Everything is a file" constraint. It uses the FUSE filesystem in the way described. I also created a possible limitation document to describe some areas where I thought it might be overengineered or locking in technical debt.

https://github.com/jimmc414/AgentOS

Jimmc414··on Anthropic cuts off xAI's access to its models for coding
https://sherwood.news/tech/report-anthropic-cuts-off-xais-ac...
Jimmc414··on Claude Code On-the-Go
Linode will provide a configured Linux box for $5/mo that works well with Claude Code and Termius. I had to jump through a surprising amount of hoops with Claude Code, Tmux and Termius to issue a shift+tab before Claude Code gained the ability to invoke plan mode conversationally
Jimmc414··on U.S. cybersecurity experts plead guilty for ransomware attacks
An indictment is a formal accusation of wrongdoing and only needs to allege facts sufficient to inform the defendant of the charges. Evidence is disclosed to the defense during discovery and presented to the court at trial.
Jimmc414··on OpenAI are quietly adopting skills, now available in ChatGPT and Codex CLI
Skills are lazy loaded prompt engineering. They are simple, but powerful. Claude sees a one line index entry per skill. You can create hundreds. The full instructions only load when invoked.

Those instructions can reference external scripts that Claude executes without loading the source. You can package them with hooks and agents in plugins. You pay tokens for the output, not the code that calls it.

Install five MCPs and you've burned a large chunk of tokens before typing a prompt. With skills, you only pay for what you use.

You can call deterministic code (pipelines, APIs, domain logic) with a non-deterministic model, triggered by plain language, without the context bloat.

Jimmc414··on Donating the Model Context Protocol and establishing the Agentic AI Foundation
Gatekeeping (in a good way) and security. I use Claude Code in the way you described but I also understand why you wouldn’t want Claude to have this level of access in production.
Jimmc414··on Ask HN: Should "I asked $AI, and it said" replies be forbidden in HN guidelines?
Banning, no. proper citations and disclosure, yes. Sometimes an AI response is noteworthy and it is the point of the post.
Jimmc414··on Mistral releases Devstral2 and Mistral Vibe CLI
We are getting to the point that its not unreasonable to think that "Generate an SVG of a pelican riding a bicycle" could be included in some training data. It would be a great way to ensure an initial thumbs up from a prominent reviewer. It's a good benchmark but it seems like it would be a good idea to include an additional random or unannounced similar test to catch any benchmaxxing.
Jimmc414··on FBI Paid $851K in Overtime for Epstein 'Transparency Project' Redactions
Incrementally and unevenly. The catholic church abuse scandal took decades of persistence before change. Tobacco litigation seemed pointless until it wasn't. Epstein was a known criminal for years before arrest. The gap is long and uncertain, but the alternative is what? You keep your own house in order regardless of whether the neighborhood is on fire. That is the only part you can control
← PreviousPage 5 of 33Next →