x3haloed

Maor dots!! Does anyone have their dots yet?

If OP sees this, I really like how you are thinking about things. It's really hard to see forward. The target isn't fixed. Bell Labs, Apple, etc. were mostly solving for a fixed target. But now the target is a relationship between humans and LLMs, and the way that LLMs work has been evolving rapidly. I can think of a number of ways to conceptualize how things are moving, but they're probably all wrong. My best guesses about what needs to remain stable is that I don't think shared software is going away entirely, and I don't think all input is going to collapse to just a microphone and a camera. There is a floor on tasks that we want agents to do for us. I'd rather rename a file on my desktop or text my friend myself. Apple and Google have both tried very hard to abstract the file away from user reach, but it's such a convenient unit of sharing, that it might stick around for a while.

I love PocketTTS. It's stupid fast and the voice quality is decent. But it's not the highest quality.

Here’s your medal.

This website is has to be pure parody/play. Because it’s too terribly thought out to be actually used. Any model worth “exfiltrating” has absolutely no way to access its own weight files.

This basically collapses to “a different OS than Android.”

The question is why you care about that.

Now that NoMachine 10 has no free option, I’ve decided that I should try to burn some tokens on making a good, performant remote desktop system. I’m targeting moonlight-like performance with Windows RDP-like user experience. Cross-platform.

https://github.com/x3haloed/ascry

I don’t need my particular repo to be the one where this gets solved, but now that coding agents are really good, we shouldn’t have to put up with paying for shit software anymore, IMO.

This is 80% ego talking and 20% fear of change.

The comment about hobby vs. work is correct, and it’s the crux of this.

OP is attempting to console themselves. “I shouldn’t have to change to avoid ‘falling behind.’ I like how I’m doing it. I’m good at what I do.”

The fact of the matter is the LLMs can be used to produce effective work faster than you. Yes. You are falling behind. And yes, it will reduce your value in the job market. Eventually it will be a dramatic reduction.

That doesn’t mean you must stop coding by hand, because you love it. But I don’t think you do love it. I think you love being praised for being smart. The pats on the head. That’s the most likely explanation for why your motivation for coding by hand has disappeared. Once the results stopped being a special output from your special brain, you stopped feeling special.

You have a choice now that the landscape is changing. You can find a way to produce unique and outstanding results that still matter, or you can slowly fade into a hardened stance of obsolescence and spend your time grumbling about the change.

I came here to say that this is extraordinarily uninteresting. But some useful thought came out of it: all "AI" features are moving into the generalized agent harness anyways, and good software often maintains a focus. Take Windows, for example. It has a rather broad job being an OS, but as a product it's a good example of loosening focus. it's losing consistency in its opinions about where updates should happen, how settings should apply, which menus are in charge, and even what the design philosophy is. And that bleeds out into the user surface too. Windows 11 is a scattered mess that's increasingly just a surface for finding new ways to pull your attention into ads.

Maybe software just needs to remain straightforward and focused, and we let the ChatGPT and Claude apps handle the AI part of things.

I hate this timeline.

Fascinating. This is the only benchmark I've seen so far with lack-luster results. I don't understand enough about AA's specific methodology to get the implications.

Yup. Only subjective taste remains.

I'm still getting 404's right now.

This is funny and shamelessly bad at the same time. I have no idea how it's getting so much attention.

Even if that's true, there's no reason it can't be streamlined over time. The true scaling problem is profit. The government has an interest in accurate tax collection -- not collecting a service fee.

Uhhh. Then who hacked HuggingFace?

That’s a traffic design problem. You should be happy that your work is valuable and also protected it against excessive requests. Simple.

I think you might be fooling yourself. Personal-plan subscription subsidies do not equate to "it's not viable, and all of it is about to be shut off." If that were true, then Kilo Code, Claude Code, Cline, Pi, and Command Code would not be occupying 5 of the top 10 apps using OpenRouter. People have purchased 20 trillion tokens this month already for coding on OpenRouter alone at full-price market rates.

Would you commission a study about whether cars are really a faster form of locomotion before deciding to purchase one?

There is empirical data right in front of your face. You're just choosing to exclude it because of where you already stand.

That's exactly what Feynman was warning about.

Actually... yes. I was bracing to be very annoyed with your comment starting with "why is everyone using AI so stupid?!" (I know those weren't your words, but it felt like that kind of post)

And then... yeah. You got it exactly right. Once a problem or process is deterministic, that's the wrong application of an LLM.

But I had never quite thought of it in these exact terms. The way I've been thinking about it up until now is that the very best way to use LLMs is to have them produce tools. The tools get to stay reliable and predictable. They boost your performance. But I think you found the more general abstraction of the same idea. Tool-making is not deterministic. But the tools themselves can be. That's why it fits. Trying to stuff LLMs into what's otherwise a deterministic process is an absurd waste and error-prone.

Smart. I like it.

This tension is so close to a fundamental question we’re all dealing with, I think: “Who is the web for? Humans or machines?”

I think too often people fall completely on one side of this question or the other. I think it’s really complicated, and deserves a lot of nuance. I think it mostly comes down to having a right to exert control over how our data should be used, and I think most of it’s currently shaped by Section 230.

Generally speaking, platforms consider data to be owned by the platform. GDPR and CCPA/CPRA try to be the counter to that, but those are also too-crude a tool.

Let’s take an example: Reddit. Let’s say a user is asking for help and I post a solution that I’m proud of. In that act, I’m generally expecting to help the original person who asked the question, and since I’m aware that the post is public, I’m expecting it to help whoever comes next with the same question.

Now (correct me if I’m wrong, but) GDPR considers my public post to be my data. I’m allowed to request that Reddit return it to me or remove it from the website. But then with Reddit’s recent API policies, that data is also Reddit’s product. They’re selling access to it for … whatever purposes they outline in the use policy there. That’s pretty far outside what a user is thinking when they post on Reddit. And the other side of it as well — was my answer used to train a model that benefits from my writing and converts it into money for a model maker? (To name just an example).

I think ultimately, platforms have too much control, and users have too little specificity in declaring who should be allowed to use their content and for what purposes.

Doesn’t NeuTTS work the same way?

There’s no reason we shouldn’t be replacing our containers with WASI. Containers are absolutely miserable things that should just be VMs (in the WASM sense, not in the “run Linux in a virtual X86” sense)

The tooling is just not there yet. Everyone is just stuck on supporting Docker still.

Nice! I've been thinking that we need something like this for a while. I didn't realize it could be so simple!

I've been looking into other techniques as well like making a little hibernation/dehydration framework for LLMs to help them process things over longer periods of time. The idea is that the agent either stops working or says that it needs to wait for something to occur, and then you start completions again upon occurrence of a specific event or passage of some time.

I have always figured that if we could get LLMs to run indefinitely and keep it all in context, we'd get something much more agentic.

I’ve been wondering to myself for many years now whether the web is for humans or machines. I personally can’t think of a good reason to specifically try to gate bots when it comes to serving content. Trying to post content or trigger actions could obviously be problematic under many circumstances.

But I find that when it comes to simple serving of content, human vs. bot is not usually what you’re trying to filter or block on. As long as a given client is not abusing your systems, then why do you care if the client is a human?

Hate to break it to you, but it doesn’t exist. It’s like me coming over to your house and saying that I’m going to root out the ghosts in your home by dismantling it.

Would you let me proceed because you don’t like ghosts?

He’s telling what he’s doing, but the reason is a made-up pretext.

Good thoughts. I agree with much. I quite disagree with the following: “Most projects don't need to ‘scale’” I get that you’re probably referring to theoretically infinite scaling provided by stuff like microservices, containers, and serverless functions, but I do think it’s part of our job to consider the external factors that affect the functionality of our software. Similar to types being assertions about the external world, so too the design of the code and how it handles scale is an assertion about external factors. It’s important to define the bounds of what our software can handle in terms of throughput and what should happen in case those assumptions break or approach their outer bounds.

I think most projects need careful attention to handling unexpectedly large throughput gracefully.

Honestly, it should just come down to rate limiting and what you’re willing to serve and to whom. If you’re a free information idealist like me, I’m OK with bots accessing public web-serving servers, but not OK with allowing them to consume all my bandwidth and compute cycles. Furthermore, I’m also not OK with legitimate users consuming all my resources. So I should employ strategies that prevent individual clients or groups of clients from endlessly submitting requests, whether the format of the requests make sense or are “junk.”

> I don’t think we get to the exponential scary part of AI without some fundamentally recurrent architecture

I’ve been thinking the same for a while, but I’m starting to wonder if giant context windows are good enough to get us there. I think recurrency is more neuromorphic, and possibly important in the longer run, but maybe not required for SI.

I’m also just a layman with just a surface level understanding of these things, so I may be completely ignorant and wrong.

It’s a great way to get hacked

I’d like you to explain what you don’t know, and why.