"For over two decades, Professor Toshitaka Suzuki dedicated his life to studying the Japanese tit — a small songbird native to Japan’s forests. Through years of careful observation and experiments, he discovered something incredible: these birds use grammar-like rules and combine sounds to form meaning, much like how humans use language."
We'd just be overloading "lessons" as well, and even more so because it takes more work to ground the concept, given its larger semantic distance from what we're describing.
I've been making digital agents since the GPT-3 API came out. Optionally fully local, fully voiced, animated, all of that. Even co-ran a VC funded company making agents, before a hostile takeover screwed it all up. The writing has been on the wall for years about where this was headed.
I have been using and evolving my own personal agent for years but the difference is that models in the last year have suddenly become way more viable. Both frontier and local models. I had been holding back releasing my agents because the appetite has just not been there, and I was worried about large companies like X ripping off my work, while I was still focused on getting things like security and privacy right before releasing my agent kit.
It's been great seeing claws out in the wild delighting people, makes me think the time is finally right to release my agent kit and let people see what a real personal digital agent looks like in terms of presentation, utility and security. Claws are still thinking too small.
Very neat. I built an agentic web DAW last year. Models were pretty crap at producing anything good, but that's changing rapidly.
I forked anthropic's MCP at the time to use it in the browser, but it was just too much trouble and I wanted to wait for something like WebMCP to appear before fiddling with it more.
Planning on dusting off the DAW and releasing it very soon.
After playing around all day over the past few days, I've come to the conclusion that the system prompt goes a huge way! We've incorporated music theory and many other instructions in the system prompt for it to be able to come up with something like this.
With that, definitely looking forward to models producing good music the analytical way and not pattern finding as we see in specialized audio/music-gen models.
Yep, I totally agree that context engineering is everything here, but the jump in model quality just in the last 4 months has just been insane. They are just way better at this now.
In the case of my DAW, I went even fundamental and created a node-based visual UI and gave the agent the ability to program new modules using the Web Audio API, and to choose from selection of stock instruments and effects. Modules are editable after instantiation, and automatically create UI for each module based on the parameters, input and output. It could spawn and wire things up, do sound design, that sort of thing.
I also have recently tried out Gemini 3.1 Pro out on audio, and you should give it a spin if you haven't yet. It actually is the first model I've seen really able to talk about music in terms of frequency and time with great accuracy. It can break down songs by instrumentation, composition, sound design, arrangement, etc.
Its philosophical take on the music itself isn't always great, but it does have precision and at a high level you can see where things are headed. Some of its advice was definitely valid and actionable. I want to plug it into my DAW or Ableton MCP and see what happens. It might actually be able to do real sound design. What I want to do is not just ask for a melody, but be able to say things like, "let's throw a Reese base in there" or "sidechain everything under the kick" and for the model to know what I'm talking about. So not just music theory, etc. but sound design as well.
I'd love to chat about this more somewhere and cross-pollinate ideas if you're up for it, email's in my bio.
I do it and never had an issue. I get odd emails every now and then with an unused address, for services/people I never contacted though. But I'm talking about perhaps 2-3 per year.
> A handful of enterprising activists could blind all the flock cameras in a region in a day or two, and without destroying them, which makes it less of an overtly criminal act
No, that would likely end in a RICO or terrorism case if it continued. Just because the cameras aren't destroyed doesn't mean CorpGov won't want to teach a lesson.
Is every author who wishes to convey certain messages to their audience through narrative also responsible for every single thing his characters say? Character-driven narrative would seem to be at odds with such a view.
I was wondering about that too. But what I mean by "responsibility" is that the ideas presented have a definite form and don't get to evade criticism by being mercurial and shape-shifting. Not sure about art, like fiction. I'm not seeking to prevent authors from being ambiguously provocative, but it's a crappy way to reason.
Yes, that's why modern literature and media dealing with diverse opinions are terrible now.
You are expected to caricature and refute people saying "bad" opinions in the work itself since otherwise the reader could believe in those opinions. Leaving something open to interpretation is tantamount to endorsement.
When you put it that way; I guess after some reflection, I realize my algorithm is optimized for efficiency and I immediately try to hone in on strong agreements or disagreements in taste/politics/etc. so that I don't waste my time getting to know a shitty person, or miss out on a potential best friend.
These means engaging in a level of provocative speech/behavior that sometimes makes people uncomfortable (not my problem of course; I have little interest in euphemism or politeness, my energy goes toward transparency and kindness)
Progressive self-disclosure can have its uses but if I can't break the ice in two minutes with a stranger, it's not a good sign for our compatibility.
Now, I did grow up in an environment where I was never really allowed to exist. I am an atheist raised by an hyper-abusive, hyper-religious, ex-boxer Catholic deacon in an extremely conservative part of the United States. The police were at my house every couple of weeks. So this may have influenced my comfort with radical transparency; I had to learn at a young age to literally fight constantly for my right to think my own way, and I'm ready to do that at any time.
But I have definitely been in some neighborhoods where the most interaction you should have with a stranger is a nod of the head, anything more is asking for trouble no matter who you are. I can vouch that there are harsh urban environments which prevent, by design, even progressive disclosure from being a safe option. This effectively kills any chance at real unity in the community, and drives up crime statistics, further justifying the continued disunification tactics.
It would be cool to catalog, categorize and analyze these kinds of social algorithms. It seems like an interesting cross-disciplinary field, involving psychology, sociology, game theory, cultural anthropology, etc.
If I meet somebody that immediately skips the progressive self-disclosure small talk and jumps right in to a big discussion… I’m going to withdraw. Even if I agree with everything you’re saying, it comes off as aggressive. like youre trying to speed run forming a relationship by skipping the small talk
No, I'm just not going to progressively disclose my nature. I'm just going to be myself, regardless of how others in my environment might react.
I can field the small talk, several of my friends have commented on my ability to break the ice quickly with strangers. But after a minute or two, the conversation is either over or we're moving onto more interesting discussions.
Come to one of the conservative towns I grew up in and you'll understand the need for such a mentality. Progressive disclosure can lead to things like accepting racism, sexism and other injustices.
It's a good mentality to carry forth into other environments as well, because at the end of the day, the less masks I have to carry, the better.
You don't sound like someone I'd like to speak with even if we might agree on things. You writing has a very aggressive tone to it for no reason.
I'll be honest if someone tries to get into politics and other such things very soon after I start speaking to them it really puts me off. I might not disengage right away, but I'll probably never choose to speak with them again.
I think both of you are projecting an aggression tone onto my words. Poe's law, maybe.
It shouldn't offend you that I don't personally enjoy continued small talk and prefer to form deeper connections at the risk of losing superficial connections, by not engaging in drawn-out progressive disclosure.
You're making assumptions about our compatibility without knowing much about me at all. But, this was my point: Now I know that we don't need to continue the discussion and we can both spend our energy elsewhere.
As a neutral 3rd-party who wants to help you speed-run your self-growth (because I like your energy):
- Yeah you come across as aggro. That's okay, sounds like you went through some stuff.
- Sounds like you've identified you grew up in a weird situation. That sounds bad, sucks you had to go through that.
- But it also sounds like there's a piece of you that's trying to overcorrect. I understand, it's common among us nerds -- you grow up in a situation where you aren't as appreciated as you should be and you try to turn off that feeling entirely somehow. Unfortunately these types of attempts to hack our own feelings are usually worse than the problems in the first place. Usually the best course is to slowly try to remind yourself (over years) "That was a bad situation, it was bad luck, it meant nothing, and it's not the norm. I don't need to fundamentally change to not have that happen again."
I do appreciate you taking the time to share your thoughts. I still believe that you and the two other commenters here are misreading or misinterpreting me.
That's a lesson for me to choose my words more carefully, but only to avoid misinterpretation in an online forum; everything I said, I 100% stand by, and it's honestly unsurprising that a comment about being provocative and to-the-point, not always progressively self-disclosing, has made some people uncomfortable.
> Unfortunately these types of attempts to hack our own feelings are usually worse than the problems in the first place
I want to stress that I mean this in the most constructive, positive way possible, but it feels like your comment projects a bit onto me in an attempt to find common ground. I welcome the attempt but I do want to point out that I don't try to "hack my feelings", and I don't organize my life and behaviors under some fear that my childhood is somehow going to happen again. I am very in touch with my feelings, I value emotional intelligence and reflection. I don't pathologically worry about others appreciating me.
I brought up my past to show how such a perspective might form, but the perspective is not some kind of defense mechanism. It is a playbook for how to live my life in a way that aligns with my ideals and goals, and it's one of the only good things to have come out of my childhood. I cherish my perspective and how it's allowed me to help both myself and others.
Well the reason I said it sounds like you were trying to "hack" feelings was that you mentioned you came up with your own social "algorithm" for testing people you meet that could make them uncomfortable. It's been my observation that most attempts to optimize conversation backfire.
But I'm not really trying to convince you, I don't have a horse in this race. If you want maybe ask an AI and see what it thinks, they are great neutral tool for being a judge on human tone or being a social mirror.
> you mentioned you came up with your own social "algorithm" for testing people you meet that could make them uncomfortable
I didn't say that, reread my comments. I have no interest in "testing" people like some sort of sociopath. You mentioned the algorithm of progressive disclosure. I said that I specifically do not do this whenever I can help it, and mused about why that is, and why it might be so for others raised in certain communities.
I said I quickly like to discover who other people are and communicate who I am, to skip all of the progressive disclosure crap and either come to terms with the fact that we aren't compatible, or to find a thread to start pulling and weaving into a relationship.
> It's been my observation that most attempts to optimize conversation backfire.
Progressive disclosure is an optimization. It's just optimizing for different things. I don't walk into a random conversation with someone planning to control how the conversation unfolds, or "optimize" it. But anyone with experience in public speaking, or leadership, sales, political organization or other environments which necessitate the ability to navigate and calibrate conversations, will learn a few tricks for keeping things on track or avoiding dull moments.
Conversational speaking is a skill, and getting better at it for the sake of improving your ability to communicate is not "hacking" or "optimizing" the conversation. I think you have decided on a bunch of behaviors in your head that I simply do not engage in.
> If you want maybe ask an AI and see what it thinks, they are great neutral tool for being a judge on human tone or being a social mirror
I have dumped my entire HN history into chatbots to study my conversational approach and learn from it. Self-betterment is always a work in progress, but I simply do not engage in the behaviors you've decided I engage in without even meeting me.
This thread has turned into a series of misunderstandings from multiple users, none of whom ever stopped to seek to understand or ask for more detail before making assumptions. Instead, I had to field several bad assumptions from people who were ironically claiming that someone whom they've never met, but simulated in their head based on a single comment, is aggressive or annoying to be around. It's ironic because, from my perspective, all of these assumptions represent missed chances for us to seek understanding from each other, and shift this from a conversation to a debate, which to me is aggressive.
I simply shared my perspective. This thread did not need to evolve this way. If I were the first user replying to my post, I would ask more questions to clarify my understanding before just deciding for myself that someone is annoying to be around because they said they like to be themselves from the jump when meeting others.
I love the idea but what keeps me in the browser is things like uBlock Origin + uMatrix + a bunch of other extensions that I know keep me safer. On top of that, Firefox has anti-fingerprinting.
I don't necessarily have a ready solution to offer, but these are the obstacles preventing someone like me from being able to use apps like this comfortably and safely, especially knowing we are entering a transitional period where new apps are being vibe-coded every day and formal verification has not yet caught up.
Even if a given app has had every line of code reviewed by a human, or has well-defined interfaces that allow for sloppier internal code, how do I know that without cracking it open myself or asking an agent to help me audit it?
Well, I suppose the app could offer a proxy service. Funnel everything through a VPS, apply ad and tracker blocking there.
That opens the door to lots of additional features… Cache responses so you can still read stuff when it gets the HN hug of death. Do a full-text index and offer a secondary search capability over article contents. Maybe build an API for all that so you can have AI Agents ground themselves on articles that got strong quality signals on HN. Maybe sign agreements with publishers like LWN, The Information, or whoever else shows up on HN behind a paywall frequently.
These are possible solutions, but offering a VPS/VPN won't convince anyone who is already on the fence over privacy or security issues. They probably have their own already and don't want their browsing data running through someone's servers. HN clients should interact directly with HN as another normal client, and not proxy incoming traffic.
Even if it enables lots of other features, that's not why I come to Hacker News and such a feature would be an immediate pass. Maybe others feel differently, but the fact that HN's design and featureset have not followed other trends over the years is part of why many of us still come here.
Maybe there is a market for what you're thinking, but I'd continue to do more market research to make sure you understand your user demographics before making the wrong move early on.
https://www.youtube.com/watch?v=jmys2abx4co
reply