ithkuil

Nice. I didn't build a skill but I noticed that for a few complicated features that I built recently, I asked the model to build an interactive "explainer" for it

Such an explainer would contain for example a simulation of the feature, where it shows how data flows between components and allows the reviewer to tweak parameters and see how it affects things. The explainer can visually show edge cases or introduce faults.

Basically my problem is that agents can write a lot. They can write a lot of code and also a lot of natural language design/spec which ultimately becomes hard to review (sloppificarion of design documents)

Sometimes you can tame this by focusing on very high level, human curated specs, but since often the devil is in the details, you need a way to see these details.

I found that interactive visual exploration tools are a very good way to both iterate on your problems and to share it with your colleagues.

Inspirations:

* https://ciechanow.ski/

* https://www.3blue1brown.com/

he's only half spinning in his grave

"please repeat, ELI5 without analogies (I'm not a child, just ADHD)"

works quite well

It's unbearable but nothing that couldn't be fixed with postprocess.

It's not just slowness, but what about case insensitive filesystems?

I'm surprised this article doesn't mention that classical latin pronunciation actually had V realized as W or U depending on context.

Latin didn't have two letters to distinguish V and U. You can still find inscriptions with VVA (modern spelling UVA)

Later that sound morphed to a voiced labiodental fricative which is what we think of V now

I think both your arguments are true. It all depends on the velocity of the capability growth and the fact that opportunity cost is expensive.

Once we get out of this hypergriwth phase the very same AI companies that now are giving you llms will provide a service that employed a rich mixture of optimized models that will reduce the operational costs to achieve the required results

yeah, my point is just that we agents can have their own docs and we can also have a nice curated experience for humans.

Perhaps we should have README.md and README.ai the latter containing a bunch of stuff that humans don't want to read but agents can use to answer questions humans ask

Hmm, I wonder if we can automate that

There are certainly consequences in participating in a war against Russia.

There are also consequences in letting Russia take Ukraine militarily.

Whatever you think about the motivations and the desired end goal, actions and lack of actions both have consequences

Plenty of old people now grew up in a phase in history where we were building ugly cheap things

Being good at matching patterns doesn't mean the patterns you found have the causal relationships you think they have

Superstition in pigeons: https://psycnet.apa.org/record/1948-04299-001

"Music Theory, Bells and Whistles"

I wonder if we're doing virtual filesystems wrong.

There is a good reason why traditionally filesystem access was mediated by the OS layer, but there are many use cases where you just want to give processes a different view of what they already can access and it could be done as a library in the same userspace process.

However, for that to work across all the processes in a session we'd need a standard way to install such a hook in all peocesses and that's achievable to some extent using LD preload but falls apart quite rapidly with statically built binaries or different libcs

When models start to effectively employ sarcasm were screwed

It's worse than that. No bank will have business with you because that would potentially affect _their_ other customer's ability to operate normally

This is a very big hammer and reveals how foreign entities are deeply dependent on the American financial system and thus exposed to the political whims of the US.

You can read about the Italian police obtained the encryption keys through the Aruba colocation provider in the Wikipedia page:

https://en.wikipedia.org/wiki/Autistici/Inventati

Italian Postal Police wiretapping (2004–2005)

And renaming things is hard, if not impossible

Sounding and being are not the same thing. I don't want to insult Mr. Pike and claim he is a dick just because how some of his quotes sound.

They haven't done it because their power would collapse if they try to do something and be ineffective at it.

There is a correlation between having strong opinions and sounding like a dick. But also between having strong opinions and having impact.

Luckily there are also people who manage to have impact while giving out a different vibe

Don't get me wrong, I also don't see it personally. I just think that the framing is useful: it highlights how our society has a bias towards business and a bias against morality-driven initiatives

If selling the weapons is "ok" because business must be allowed because the motive is purely selfish (I just want to make a living for myself and my family), and giving away weapons is "suspect" because "you clearly have an ulterior motive", if just means that we live in a world where selfishness is the norm being altruists is suspect.

I think this is the core divide of our society, much more useful than the obsolete "left" and "right" divide

That simile captures perfectly the logic here. I don't think it's technically sarcasm

EDIT: a better simile would be if guns were giving away for free. You can see how that could be construed as increasing the responsibility of the provider if a receiver then does something illegal with it

What does "naming" mean?

Assigning a symbol? But who said that the set of symbols must be countable?

Then the board can fire that CEO?

"paradox" is an overloaded term.

Jevons paradox is a veridical paradox, which, as you said, means that it's a true statement that merely looks wrong because it is counterintuitive.

I know that some people think that the word "paradox" should be only used to refer to antinomy paradoxes which are often called "true paradoxes" (such as "this sentence is false") which lead to a contradiction without requiring a flaw in reasoning.

You can get those in a democracy too!

The real difference is how can you get rid of those

big restaurants have many cooks that are trained and monitored by the chef, but the chef is certainly not necessarily prepairing your meal.

The chef is however responsible for the outcome so you're absolutely right to complain if the meal sucks and most of this AI written crap definitely sucks to read.

That said, it's not about the "principle" of having your meal cooked by the chef / the text written word by word by the human.

It's about the outcome. People should learn to tame their AI tools/harnesses in such a way to produce quality output (code and prose) and it's absolutely despicable that so many people are so eager to lower our standards in quality just because everybody else seems to be doing it.

Don't get me wrong, I loathe the AI style as the next person.

But I don't buy your argument: if a person used AI to produce a text it doesn't mean they haven't steered the I in the direction they wanted the conversation to go and (hopefully) review the output.

That's an added value

Yes, it would be nice if it was also pleasant to read