bvan

Very interesting. Nonzero correlation between two hazards.

Cool. I would have met a couple of these folks when they came over to demo the new system in Miami.

Exactly. If it’s happened once, it will happen again.

I want to believe… but..

Perfect for me. When you’re raised in a world of science your brain does not rest until you understand some level of ‘why’. Beginner music instruction and theory usually assumes you’re brain-dead and just take in, no questions asked.

This is analogous to what AI is turning into. Pretty soon, some folks won’t be able to tie their shoe laces without tech.

What has Hacker News come to… sigh..

Perhaps, however, this hacky language is more productive in the real world than most. What’s the point, otherwise?

Yes, it is worth looking at. My thinking was, how fast did that sequence revert back to the’mean’? It doesn’t take away from the author’s message though.

You’re welcome to go get the data and chart it the way you think it should be charted, and then we can have a conversation.

In the era of AI this is quaint. They ain’t seen nothing yet.

This is the kind of application which brings me back to shore as I feel I’m drowning in AI slop. Really well done and thanks for sharing.

Not to be too picky, but I think when the author refers to ‘number of samples’, he means ‘sample size’. One sample, one or more observations.

Reminds me of process I went through to create the charts for my humble grad thesis (1990-ish), by hand, with letter templates and one very expensive rotring pen. Then would transfer the figures to transparencies as backup for my defense. The real presentation was done using a slide projector (this was cutting edge back then, like over-the-top). Yes, took actual photos of my charts..

Gave back my green card the moment I left the US. No longer wanted the hassle and ties to an unpredictable regime. Haven’t looked back.

Why?

Along the same lines: https://iopscience.iop.org/article/10.1088/1361-6552/ad0542

Yes, in very specific cases where I fully understand the methodology(ies) that is (are) applicable, and am able to verify correct implementation. Also, as an enhanced ‘Google search’ to supplement what I have found. I am the skeptical type… yet, so far have been impressed. But, I wouldn’t trust using AI to blindly give me solutions to a problem I couldn’t solve myself, albeit much more slowly.

Fake it ‘till you make it. Seems like LLM’s have caught-on to that too.

That’s great. I find the Sultanate coins, for example, are much more refined in terms of style relative to those minted Western Europe around the same time. For example: https://www.numismall.com/collections/bengal-sultanate-1203-....

As someone interested in coinage, India is fascinating. Coins from the Sultanates and Mughal periods are just beautiful.

Totally agree. Especially for commercial or code that adds value. Writing code is just one element of developing quality, robust, software. In the rea world, commercial or production software must be maintained, supported, and must respond to changing user requirements. The human element is critical, unless you’re OK with relying on LLM’s, crossing your fingers, and have no care to support users.

WPF marked the end of well thought-out UI’s and standards. It’s been web-inspired slop ever since.

This a very insightful post. Thanks for taking the time to share your experience. AI is incredibly powerful, but it’s no free-lunch.

Absolutely nothing, as 99% of comments to the post. But it is the norm on HN it seems.

As much as you all dislike LinkedIn and the cringy posts, keep in mind that for certain parts of the market it is >the< main professional forum. It is where your investors live, and their capital providers live. So, play nice, yeah?

Nicely done. I have the same challenge with Bayesian stats and usually do not understand why there is such controversy. It isn’t a question of either/or, except in the minds of academics who rarely venture out into the real world, or have to balance intellectual purity with getting a job done.

In the very first example, a practitioner would consciously have to decide (i.e. make the assumption) whether the number of side on the die (n) is known and deterministic. Once that decision is made, the framework with which observations are evaluated and statistical reasoning applied will forever be conditional on that assumption.. unless it is revised. Practitioners are generally OK with that, whether it leads to ‘Bayesian’ or ‘frequentist’ analysis, and move on.

Very well said. Slow code for me. I fully understand what I build. The AI magic bus will come crashing down at some point.

Quattro pro was the bomb

Around the world in soundscapes.. great idea.