curl-up

> "I think it's really struck a chord with how people are feeling in the world right now - the world is in quite a stressy place," Rigden said. "With the game you are able to focus in on one thing, which kind of blocks all the other distractions out - it's a pure form of meditation."

People love providing such explanations to various observed trends, but never consider looking back to check if trends are new in the first place. FWIW, I don't see how these games are allowing players to "focus in on one thing" (and put things in order) any more than Solitare or Farmville.

Highly recommend a series on Lodoicea (aka Double coconut or Coco de mer) from the Weird Explorer yt channel: https://youtu.be/GqicsIDYmgU

It is one thing to not use AI because you enjoy the process, or because you believe that art is inseparable from the act of creation.

But she claims that the fact that AI exists decreases her drive to create. That is a much stronger and less obvious argument, and something that doesn't apply to you if you're still building those models.

If she simply "enjoys the effort", as you claim, she would do the same as you - continue enjoying it. But she is not.

I've spent some time looking into all these methods before, but all of them required substantial amounts of plastic in contact with plants/water and in full sun/heat. Are you worried about leachables?

To be clear, I'm not asking this in some new age way, and I'm sure it's better than the amount of pesti/herbicides used traditionally (and the whole movement behind hydro/aquaponics is fascinating to me), just wondering if this is something you ever tried minimising with such setups?

Prompt they use in `Figure 28.` is a complete mess, all the way from starting it with "Your are an expert" to the highly overlapping categories to the poorly specified JSON without clear direction on how to fill in those fields.

Similar mess with can be found in `Figure 34.`, with an added bonus of "DO NOT MAKE MISTAKES!" and "If you make a mistake you'll be fined $100".

Also, why are all of these research papers always using such weak LLMs to do anything? All of this makes their results very questionable, even if they mostly agree with "common intuition".

A very naive question: why are "dry and on the shelf" not worth buying, when so many of the food-related microorganisms obviously work fine through such distribution (baking yeasts, various yogurt starters, cheese molds, etc.)?

Oh that's nice! Any references on this shortcut? How do you activate that behavior? I was playing around with ES, but the only suggestion I found was to use `count` on filters before deciding (manually) which path to take.

I'm facing the problem you describe daily. It's especially bad because it's very difficult for me to predict if the set of filters will reduce the dataset by ~1% (in which case following the original vector index is fine) or by 99.99% (in which case you just want to brute force the remaining vectors).

Tried a million different things, but haven't heard of Turbopuffer yet. Any references on how they perform with such additional filters?

This thread should end up in the hall of fame, right next to the Dropbox one.

From a fellow LLM-powered app builder, I wish you best of luck!

Jobs called it "Google in paperback form" [1].

I wonder what obscure (probably online?) source of information of today we'll be comparing to the mainstream sources of tomorrow.

[1] https://www.theguardian.com/technology/2011/oct/09/steve-job...

Agreed, I failed to mention this detail, thank you for adding. But hopefully you agree that his description paints a completely ridiculous picture.

I tried explaining in the other comment here. In summary, it's beautiful, raw nature, that's different from anything else I've seen. It takes some preparation, but I haven't experience any of these "sun will kill you and ice will kill you too" vibes that the article suggests. It felt like one of the last places on Earth not affected by humans (at least directly through resource exploitation, as it definitely is affected by the warming - but that's another discussion).

> The first thing you notice when you land in Greenland is there are no trees or grass. There is snow and then there is exposed rock.

This is only true of the area around the airport. Even his pictures further into the article show how misleading this description is. I was actually very surprised how little snow/ice there was. Now when I think of Greenland, I think of something similar to [1].

Of course, in the winter, it's a completely different story (I was there in July). But he was there during the warm period as well (as is obvious from his photos).

> The city itself sits in a landscape so dramatically inhospitable it makes the surface of Mars look cozy.

If you look at a map, you will notice that Nuuk is at the same latitude as Reykjavik. There's a common meme about Iceland being green and Greenland being icy, and that's definitely true if you compare inland or northern Greenland with Iceland during summer (during winter, both are icy and dark), but hiking around Nuuk is a very "green" experience. Yes, there's a ton of mosquitoes, but nature itself is very inviting. I did not get any of the "inhospitable" vibes he mentions.

> But again even riding the bus around it is impossible to escape the feeling that this is a fundamentally hostile to human life place. The sun is bright and during the summer its pretty hot, with my skin feeling like it was starting the burn pretty much the second it was exposed to the light. It's hard to even dress for, with layers of sunscreen, bug spray and then something warm on top if you suddenly got cold.

This whole section is just overblown BS.

All in all, I enjoyed it a lot. Compared to Iceland, it's definitely a lot less "user friendly" and you need to prepare better, but I have never been to a place that is less affected by humans, and in our age, that is something worth experiencing.

[1] https://truewindhealingtravel.com/wp-content/uploads/2022/08...

Having been to Greenland recently, I found his description of it similarly flawed.

Code with logging is "self reporting". Adding logging statements is not reporting itself. Adding `logger.error(f"{job} failed")` is not reporting itself, and LLMs are perfectly capable of adding such statements in applicable places.

As to why, it's because I'm building an app with a growing userbase and need to accommodate to their requirements and build new features to stay ahead of the competition. Why you decided I'm describing a toy project is beyond me.

Exactly. I tend to like Hotz, but by his description, every developer is also "a compiler", so it's a useless argument.

My life quality (as a startup cofounder wearing many different hats across the whole stack) would drop significantly if Cursor-like tools [1] were taken away from me, because it takes me a lot of mental effort to push myself to do the boring task, which leads to procrastination, which leads to delays, which leads to frustration. Being able to offload such tasks to AI is incredibly valuable, and since I've been in this space from "day 1", I think I have a very good grasp on what type of task I can trust it to do correctly. Here are some examples:

- Add logging throughout some code

- Turn a set of function calls that have gotten too deep into a nice class with clean interfaces

- Build a Streamlit dashboard that shows some basic stats from some table in the database

- Rewrite this LLM prompt to fix any typos and inconsistencies - yeah, "compiling" English instructions into English code also works great!

- Write all the "create index" lines for this SQL table, so that <insert a bunch of search usecases> perform well.

[1] I'm actually currently back to Copilot Chat, but it doesn't really matter that much.

In the context compression approach, why aren't the agents labeled as subagents instead? The compressed context is basically a "subtask".

This is my main issue with all these agentic frameworks - they always conviniently forget that there is nothing "individual" about the thing they label "an agent" and draw a box around.

Such "on demand" agents, spawned directly from previos LLM output, are never in any way substantially different from dynamic context compression/filtering.

I think the only sensible framework is to think in terms of tools, with clear interfaces, and a single "agent" (single linear interaction chain) using those tools towards a goal. Such tools could be LLM-based or not. Forcing a distinction between a "function tool" and an "agent that does somethng" doesn't make sense.

The question was "why does it feel more real", and I answered that - because the best AI generated images today feel more real than the best 3D renders, even when they take all the compute in the world to finish. So I can imagine that trend going forward into real-time rendering as well.

I did not claim that AI-based rendering will overcome traditional methods, and have even explicitly said that this is a heavy assumption, but explained why I see it as exciting.

Not OP, but I have long thought of this type of approach (underlying "hard coded" object tracking + fuzzy AI rendering) to be the next step, so I'll respond.

The problem with using equations is that they seem to have plateaued. Hardware requirements for games today keep growing, and yet every character still has that awful "plastic skin", among all the other issues, and for a lot of people (me included) this creates heavy uncanny-valley effects that makes modern games unplayable.

On the other hand, images created by image models today look fully realistic. If we assume (and I fully agree that this is a strong and optimistic assumption) that it will soon be possible to run such models in real time, and that techniques for object permanence will improve (as they keep improving at an incredible phase right now), then this might finally bring us to the next level of realism.

Even if realism is not what you're aiming for, I think it's easy to imagine how this might change the game.

Can Chroma handle combined structured+vector search? E.g. filtering by `category=X and value>Y` then finding top N matches by vec similarity?

Note that (in the first test, the only one where output length is reported), Gemini Pro returned more than 3x the amount of text, at less than 2x the amount of time. From my experience with Gemini, that time was probably mainly spent on thinking, length of which is not reported here. So looking at pure TPS of output, Gemini is faster, but without clear info on the thinking time/length, it's impossible to judge.

Anyone who has recently worked on embedding model finetuning, any useful tools you'd recommend (both for dataset curation and actual finetuning)? Any models you'd recommend as especially good for finetuning?

I'm interested in both full model finetunes, and downstream matrix optimization as done in [1].

[1] https://github.com/openai/openai-cookbook/blob/main/examples...

I am in no way disputing that some people would feel that way because of AI, just as some performing classical musicians felt that way in the advent of the audio recorder.

What I am saying is that (1) I regard this as an unhealthy relationship to creativity (and I accept that this is subjective), and (2) that most people do not feel that way, as can be confirmed by the fact that chess, go, and live music performances are all still very much practiced.

Would you say that the chess players became "purposeless" with Deep Blue, or Go players with Alpha Go?

If the core idea that was intended to be broadcasted to the world was a "contribution", and LLM simply expanded on it, then I would view LLMs simply a component in that broadcasting operation (just as the internet infrastructure would be), and the author's contribution would still be intact, and so should his enjoyment.

But his argument does not align with that. His argument is that he enjoys the act of writing itself. If he views his act of writing (regardless of the idea being transmitted) as his "contribution to world's knowledge", then I have to say I disagree - I don't think his writing is particularly interesting in and of itself. His ideas might be interesting (even if I disagree), but he obviously doesn't find the formation of ideas enjoyable enough.

> The fun has been sucked out of the process of creation because nothing I make organically can compete with what AI already produces—or soon will.

So the fun, all along, was not in the process of creation itself, but in the fact that the creator could somehow feel superior to others not being able to create? I find this to be a very unhealthy relationship to creativity.

My mixer can mix dough better than I can, but I still enjoy kneading it by hand. The incredibly good artisanal bakery down the street did not reduce my enjoyment of baking, even though I cannot compete with them in quality by any measure. Modern slip casting can make superior pottery by many different quality measures, but potters enjoy throwing it on a wheel and producing unique pieces.

But if your idea of fun is tied to the "no one else can do this but me", then you've been doing it wrong before AI existed.

Can you explain your reasoning?

Over the last 7 years I've built 3 major products that needed charts. Every time I tested most of (at that time) popular solutions. All 3 times eCharts was by far the best option, across all criteria. Incredible library!

Relevant, I guess: https://www.astralcodexten.com/p/the-colors-of-her-coat

Cheap means small, small means low Q&A scores. I know that this isn't that important for the majority of applications, but I feel that over-reliance on RAG whenever Q&A performance is discussed is quite misleading.

Being able to clearly and correctly discuss science topics, to write about art, to understand nuances in (previously unseen) literature, etc. is impossible simply through powerful-reasoning + RAG, and so many advanced use cases would be enabled by this. Sonnet 3.5+ and GPT 4.5 are still unparalleled here, and it's not even close.