sidpatil

I listen to The AI Daily Brief: https://podcasts.apple.com/us/podcast/the-ai-daily-brief-art...

Further, they may change the behavior of the existing model without providing clear notice. E.g. existing models start to behave worse shortly after new model releases, due to resource reallocation away from the older models toward the newly-released ones.

This phenomenon is referred to as the Sapir-Whorf hypothesis: https://en.wikipedia.org/wiki/Linguistic_relativity

I'm pretty sure Coursera or edX was using this same approach a while back — they'd make you type in a paragraph of text, then they'd use the timing information as a fingerprint or signature, to authenticate you as the actual student.

In software, this is known as Hyrum's Law: https://www.hyrumslaw.com/

https://rakhim.org/honestly-undefined/19/

There's also the weirdness of tunneling protocols like PPPoE which only exist to tunnel Ethernet through non-Ethernet systems.

Isn't PPPoE the opposite — carrying a non-Ethernet protocol over Ethernet?

Original title: 'Rubens Menin's "Very Very Old" Port wine'

Someone here mentioned a whole ago that the labs deliberately haven't tried to train these characteristics out of their models, because leaving them in makes it easier to identify, and therefore exclude, LLM-generated text from their training corpus.

Why?

Construction and home contracting follow building codes.

Only if it's decomposing aerobically.

Anaerobic decomposition, like what you find in a septic tank, doesn't produce any heat.

I don’t care if you make an honest mistake. Hell, I don’t even care if you make a careless mistake, as long as you fix yourself. Everyone messes up - it’s how you act afterwards that matters.

You're not the one in control of their employment status and workplace reputation.

Is nuclear energy considered renewable?

It looks like Slidev is designed for presentations about software development, judging from its feature set. Quarto is more general-purpose. (That's not to say Quarto can't support the same features, but currently it doesn't.)

I'm not affiliated with Slidev. I was just curious.

It's not, to be fair.

I use my own `.md` plan files rather than Claude Code’s built-in plan mode. The built-in plan mode sucks.

smaller scale stories are replaced with moving the big plot forward

This is pretty common in TV shows, from what I've noticed. It takes a few seasons for a show to find its footing.

Does this products directly compete with GitHub Models [1]?

[1] https://github.com/marketplace?type=models

You might be interested in Substrait (https://substrait.io/).

It may be that like clothes, there's only so much need for software.

Clothing demand has increased greatly in the past decade due to fast fashion. Much of this clothing is designed to cost a few bucks, last a few wears, then get thrown out. It's an ecological disaster.

Maybe we'll see something similar happen with software — as production costs fall, trends will shift toward few-use throwaway software. I highly suspect this is already happening.

I'm unfamiliar with all the changes coming soon. What would you need to refactor?

You would need to rerun the LLM, but you wouldn't necessarily need to rebuild the codebase from scratch.

You can provide the existing spec, the new spec, and the existing codebase all as context, then have the LLM modify the codebase according to the updates to the spec.

Perhaps not for much longer:

https://news.ycombinator.com/item?id=45785987

Japan Post is rolling out such a system: https://news.ycombinator.com/item?id=44117779

The Wal-Mart near me uses e-ink price tags for many (but not all) of their items.

What do you dislike about that style?

PRQL [1] is a pipeline-based query language that compiles to SQL.

[1] https://prql-lang.org/

Green beans are unroasted, yes. They can last for a long time in storage.

You can try asking ChatGPT to rewrite the original script to include the improvements.

What are some of the issues you've had with the Pkg system?