bottom999mottob

This analogy almost works, except about a fifth the hay making machines' output is needles.

This is literally one the most infantilizing and simultaneously insulting analogies I've ever come across on this site. Do you really think consumers of the latest AI tools have no ability to forecast? The parents in this analogy have every incentive to lie

I think you'd end up with more plastic from just breathing in air and eating seafood than Invisalign, but I haven't done a study on that.

One-size-fits-all never fits everyone :)

In theory, if a program is written in a high-level language and you have a correct implementation (interpreter, compiler, runtime) of that language on a new system, then the program should be able to run there.

In practice, this is not always so straightforward, especially as you move closer to machine-level details or consider compiled binaries.

Many compiled programs are built for a specific architecture (x86, ARM, etc.). They won’t run on a different architecture unless you provide either: A cross-compiler (to generate new native code for that architecture), or an emulator (which mimics the old architecture on the new one)

I'm curious how you've cured your schizoaffective disorder? Diet?

I think your LLM temperature analogy is interesting in this deprecated dichotomy between autism and schizophrenia.

One Youtuber Jreg used a breadth-first search (schizophrenia) vs depth-first search (autism) analogy when comparing the the two, but I think your temperature analogy is more apt. Higher temperature results in more disorganized thoughts like schizophrenia. And if you buy into the idea that the root of most schizophrenia is thought disorders, then this analogy implies that dialing up temperature corresponds to more signs of psychosis through speech

My experience with many friends on the autism spectrum is that their speech tends to be more scripted, but I certainly don't think autism and psychosis are mutually exclusive.

Sediments including lead tend to accumulate at the bottom of water heaters, so you'd be measuring the accumulated lead and not the locality's with this assumption.

What a dystopian world we live in where oligarchs controlling anti-trust companies deny medical coverage [0]. Am I surprised this happened...

[0] https://www.propublica.org/article/unitedhealth-healthcare-i...

When we ripped out the carpet to install luxury vinyl flooring it was off-gassing for weeks. Had to go over the vinyl with dozens of microfiber clothes before the outside coating stopped wiping off.

Imagine all of the people cutting into that inexpensive wood without air filtration is terrifying

Or it could be that people spend more time in their luxury vinyl planked house, vaping and having ingress air filled with formaldehyde from traffic

It'd make more sense and be less ethical to expose people to either formaldehyde or tire dust in a confined area, but that study is definitely not getting funded

If you read at least 3 paragraphs in, you'll see that this tool attempts to generate a query using an LLM. Dismissing a tool before attempting to understand it is astounding.

The US owes it to the Vietnamese people to clean up the arsenic, but of course they're never going to do that. I know several elderly Vietnamese who developed leukemia after serving in the war.

If you eat rice, please eat white not brown, parboil it with a ratio of 4:1 water to rice, and dump the water afterwards [0].

https://www.sciencedirect.com/science/article/pii/S004896972...

For long chats, I suggest exporting any artifacts, asking Claude to summarize the chat and put the artifacts and summarization in a project. There's no need to stuff Claude's context windows, especially if you tend to ask a lot of explanation-type questions like I do.

I've also read some people get around rate limits using the API through OpenRouter, and I'm sure you could hook a document store around that easily, but the Claude UI is low-friction

So I tried to search myself just to see how terrible/awesome this product is. On attempting to use the free plan:

Error: Failed to subscribe to the plan. You ran out of credits. Please upgrade your Plan

Dark pattern, lack of testing, or incompetence? Please do better if you're contributing to the Orwellian surveillance capitalist state

I'm curious how your experience is using Gameface for day-to-day tasks like coding. I assume you still use a keyboard for typing, but what about selecting blocks of text or general navigation?

This is really cool, but there's real-world instrument physics that might not be captured by simple Fourier transform templates, like a trumpet playing softly can have a significantly different harmonic spectrum than the same trumpet playing loudly, even at the same pitch

Trumpets produce a rich harmonic series with strong overtones, meaning their Fourier transform would show prominent peaks at integer multiples of the fundamental frequency. Instruments like flutes have more pure tones, but brass instruments typically have stronger higher harmonics, which would lead to more complex partial derivatives in the matrix equation shown in the article

So this script uses bandpass filtering and cross-correlation of attack/release envelopes to identify note timing. Given that brass instruments can exhibit non-linear behavior where the harmonic content changes significantly with playing intensity (think of the brightness difference between pp and ff passages), not sure how would this algorithm could handle intensity-dependent timbral variations. I'd consider adding intensity-dependent Fourier templates for each instrument to improve accuracy

That reminds me of Dr. Greenamyre in Pittsburgh who thinks he accidentally gave himself Parkinsons researching rotenone in pesticides [0]

[0] https://youtu.be/7mYEqUrx6dQ?si=x4P6pkUCd7EvCe1N&t=152

Assuming a lunch break is not included as part of 8 hours, do you really think most exmployees are productive for a full 8 hours?

Most office workers spend an exorbitant amount of time twiddling their thumbs, reading emails, going to the restroom. On average, I'd say knowledge workers are typically productive for about 2 to 3 hours per day. I'd estimate physical laborers range up to 4 or 5 hours.

Assuming 8 hours of full productivity is a strange number to focus on

This sounds like the video game Orwell where you investigate social graphs to find dissidents of the state.

Of course the NSA has been doing this for years, but their UI probably isn't as pretty as Palantir's :)

Well back in my day we had to walk to school in the snow uphill, downhill, work on the farm, raise a family, survive in a capitalist dystopia and stay sane

You youngins' just don't get it

I'd argue that saline should be the panacea here. I doubt very many people do at-home saline rinses with filtered, sterilized water and a simple mixture of salt and baking soda.

Do people really want to spray PFAS water directly into their mucus lining?

I bought an Arm and Hammer Saline spray out of curiosity. It smelled awful, and the BPS lined can had an awful smell despite the ingredients being: water, salt, baking soda, and no suspicious preservatives.

One can also absorb chemicals through the skin (it is the largest exposed organ infact), so I've been thinking of getting a shower or whole-house filter. Definitely upgrade from the Brita though. Brita's aren't very effective at filtering PFAS last I read.

But to answer your question, I'd speculate that the filtered water from the RO filter would be superior to anything coming out of your tap, even if it had phthalates in the filter. You also have to account for PFAS, heavy metals, medicines people excrete back into the water supply, and many many other things...

This is one of the worst over-generalizations.

Cultures are not monolithic, static entities. How do we go from "different cultures" to "negative outcomes?" That's a complete non-sequiter.

Imagine if all of Germany moved to India. What would happen? What if part of Britain moved to UK? At some point the politicians will have to look at the issue...

SmolLM v2 1.7GB Q8 has been running pretty fast. Only needs about 3.5GB of memory. I think it's fine for chatting, but it's only really good for Python code compared to larger models.

Deepseek-coder is pretty slow, even with the quantized models.

https://huggingface.co/blog/smollm

The point isn't about whether he should let the options expire - of course he shouldn't. The point is about accuracy in discussing Oracle's corporate governance.

You specifically claimed "Larry has a Blind Trust that sells his stock as they see best fit without his influence." The SEC filings directly contradict this:

1. This is a Revocable Trust, and while technically a blind trust can be revocable, the filings show this trust is actively managed with Ellison's involvement:

- The transactions were specifically timed around options expiration

- The sales were carefully structured in declining price blocks

- The trust uses Ellison's authority as Officer/10% Owner

- Forms are signed by his Attorney

2. These weren't discretionary sales by trustees "as they see best fit" - they were planned options exercises with a clear purpose and strategy.

I don't have an axe to grind. I'm just correcting a mischaracterization about how Larry manages his Oracle holdings. Accuracy matters when discussing corporate governance and SEC filings, especially for a company as significant as Oracle.

If this were a true blind trust they would've exercised the options at $40.47, immediately sell at market (~$143), and net ~$100 profit per share. Instead they did price stepping, and used Larry's Officer status under the pretense of independence.

Making Larry lose money while you sleep, truly a reverse Uno :)

Actually I think you're mischaracterizing and spreading misinformation about how Larry's stock sales work. Let's look at the ACTUAL SEC Filings on their website [0]. These filings show this is the "LAWRENCE J. ELLISON REVOCABLE TRUST", but being revocable doesn't automatically mean it's not a blind trust. However, several details from the filings indicate this was NOT operating as a blind trust:

1. These were planned exercises of stock options that were expiring on July 24, 2024 [1]

2. Two transactions of 1,125,000 options each were exercised at $40.47 [1]

3. The shares were then sold in declining price blocks to minimize market impact

4. All forms are signed by either Paul T. Marinelli (trustee) [2][3] or Aimee Weast (Attorney in fact for Ellison) [1]

5. Ellison is explicitly listed as "Officer" and "10% Owner" [1]

Go to page 4 to see these documents: [0] https://investor.oracle.com/sec-filings/default.aspx

July 17, 2024 - Form4 - Statement of Changes in Beneficial Ownership

[1] https://d18rn0p25nwr6d.cloudfront.net/CIK-0001341439/1c6945f...

July 16, 2024 - Form144 - Report of proposed sale of securities

[2] https://d18rn0p25nwr6d.cloudfront.net/CIK-0001341439/3bc4831...

July 15, 2024 - Form144 - Report of proposed sale of securities

[3] https://d18rn0p25nwr6d.cloudfront.net/CIK-0001341439/0b76c30...

While you're correct that Ellison isn't hurting for cash, the evidence suggests these were scheduled options exercises ahead of expiration, not discretionary blind trust trading.

I'd encourage you to review the actual SEC filings before making claims about how Larry handles his investments...

Yeah I refuse to give Oracle a dime, so I'll keep that in mind.

I only run the tiny models for fun (the CPU inference is enough) and any complex coding tasks go through GPT or Claude. Mainly just using the one instance as another backup server, run scrapers, host a bunch of containers, and experiment with networking services. I don't see a reason to split up my one instance, but maybe in the future.

Did you link a card with Pay-as-you-go? It seems like OCI puts free tier users in a low-priority queue, so that people have resorted to using auto-clicker scripts for weeks to even provision an instance.

That repo looks cool, if I really wanted to learn Kubernetes. My usage right now is for running inference on CPU RAMs because I refuse to upgrade my 980-ti, so I'm not sure it would work for me