sagebird

a human would have noticed something different about the requests it was getting, or the frequency of requests, and as soon as it noticed a shift, it would have carried that knowledge forward and intensified the scrutiny if something seemed off- eventually communicating it up the chain.

- instead of the ai context dying.

in the ai case, information only survives to the extent where the ai is empowered to store a note or notify a manager of an observation. Anything that does not result in sending a message/storage is wiped

should have submitted it to 5 independent fact checkers - would have deflated nonsense before it began by showing that you are going to see trivial and non trivial shifts among them.

mostly true and misleading counting as separate buckets while also being somewhat orthogonal conceptually is also stupid.

a better output format might be true|false|unknowable, confidence where confidence is 0..1

at least then you can compare agreement among models as a distance measurement and not a moronic bucket agreement

the conclusion is actually obvious: llms are good enough for most of this work and it is definitely cheaper, so you should use llms for fact checking at least as a first pass

same, without context it is speculation. even with context, if the person creating the story controls the window, a window can be found which supports the story.

Does running the command to remove the anti-goblin ask from local prompt increase performance slightly because there is less "cognitive load" to have to hold its tongue?

If you work at open ai or another llm company, I have a clear message I want you to hear:

I don't give a shit if my agents say goblins or not.

They are coding monkeys to me, researchers, etc.

I only care about their performance. perf per token / cost.

If you load their context with a bunch of style rules or safety theater shit, really - please don't - the context is for me.

Do you de-goblin before you run all the benchmarks, because that is what i am paying for, the performance as benchmarked - please don't benchmark then ship a bunch of one shot context mods to my install by default.

The article is cute and interesting but doesn't rise to the level of a thing I give a shit about for my use.

I think that what is missing in this discussion is the latent background that one needs to actually create "ai art" that has merit.

It is a strange combination of skills that some have developed to create styles, pipelines, techniques to create bespoke, recognizable, artistic output from these tools.

Anyone who is typing a prompt into a publicly available generative model is likely not getting anything of high value from it.

I will commend parascene project for trying to court this special sort of advanced users who can create custom pipelines and then connect it to the parascene system for anyone to use.

That process, the implied skills to create a custom generator and host it, can all be broken down so that more people can do it, but I don't think enough people realize it is even something they can do. We are so trained to be consumers of ai services.

I am not an expert but this seems like model distillation could work to get the behavior you need to run on a cheap end-user processor (Raspberry Pi 4/5 class). I chatted with claude opus about your project and had the following advice:

For the compute problem, you don't need a Jetson. The approach you want is knowledge distillation: train a large, expensive teacher model offline on a beefy GPU (cloud instance, your laptop's GPU, whatever), then distill it down into a tiny student network like a MobileNetV3-Small or EfficientNet-Lite. Quantize that student to int8 and export it to TFLite. The resulting model is 2-3 MB and runs at 10-20 FPS on a Raspberry Pi 4/5 with just the CPU - no ML accelerator needed. For even cheaper, an ESP32-S3 with a camera module can run sub-500KB models for simpler tasks. The preprocessing is trivial: resize the camera frame to 224x224, normalize pixel values, feed the tensor to the TFLite interpreter. The CNN learns its own feature extraction internally, so you don't need any classical CV preprocessing. Looking at your observations, I think the deeper issue is what you identified: there's not enough signal in single frames. Your validation loss not converging even after augmentation and ImageNet pretraining confirms this. The fix is exactly what you listed in your future work - feed stacked temporal frames instead of single images. A simple approach is to concatenate 3-4 consecutive grayscale frames into a multi-channel input (e.g., 224x224x4). This gives the network implicit motion, velocity, and approach-rate information without needing to compute optical flow explicitly. It's the same trick DeepMind used in the original Atari DQN paper - a single frame of Pong doesn't tell you which direction the ball is moving either. On the action space: your intuition about STOP being problematic is right. It creates a degenerate attractor - once the model predicts STOP, there's no recovery mechanism. The paper you referenced that only uses STOP at goal-reached is the better design. Also consider that TURN_CW and TURN_CCW have no obvious visual signal in a single frame (which way to turn is a function of where you've been and where you're going, not just what you see right now), which is another reason temporal stacking or adding a small recurrent/memory component would help. Even a simple LSTM or state tuple fed alongside the image could encode "I've been turning left for 3 steps, maybe try something else." For the longer term, consider a hybrid architecture: use the distilled neural net for obstacle detection and free-space classification, but pair it with classical SLAM or even simple odometry-based mapping for path planning and coverage. Pure end-to-end behavior cloning for the full navigation stack is a hard problem - even the commercial robots use learned perception with algorithmic planning. And your data collection would get easier too, because you'd only need to label "what's in front of me" rather than "what should I do," which decouples perception from decision-making and makes each piece easier to train and debug independently.

Can you please design a version for kids to ride on?

With a seat and handle similar to "wooden bee ride on" by b. toys?

I want a vacuum that kids can actually drive, ride on, do real vacuuming and has minimal levels so safety, like turning it over halts vacuums, stairs/ledges are avoided, and lack of rollers or items that could snare a kids hair, etc.

There may be benefits of fusion of child input signals with supervisory vacuums route goals. Would be age dependent, older kids would want full manual I think.

Kids like to do real jobs, and as a parent I prefer purchasing real items for my kids rather than toy versions if practical.

> To misquote Kennedy, “we chose to focus coroutines on generator in C++23, not because it is hard, but because it is easy”.

Appreciate this humor -- absurd, tasteful.

This is an interesting project, congrats. I have a similar project and goals (free, light, fast portable os-like-dev environment) at http://shiro.computer, http://shiro.computer/about ( https://github.com/williamsharkey/shiro ).

You may find bits of Shiro's code useful as it has massive shimming work to get Claude Code, npm/node, git, various grep tools and isomorphic git and git diffs to work, and some weird features like virtual servers that create virtual ports to communicate to frames.

All the unix tools that Claude code are supported as well. It is also a typescript project and has similar architecture, and MIT license so there may be parts you can just straight import without much hassle.

Probably the hardest part to keep architecturally clean is the shimming required in js eval environment to make Claude Code and non-browser-native packages to run. But it is very nice once you have an agent able to work inside your browser os.

Great job and thanks for sharing Lifo. I am certain this will catch on once the implementations become more solid.

ok do pornhub next

it's not agi until we have browser browsers automating atm machine machining machines, imo

Wes Anderson is positioning cinematographers on American soil with extremely powerful telephoto lenses, filming actors performing in meticulously designed miniature sets across the Canadian border. The film will be titled "The Asymmetrical Tax Avoidance" and will star Bill Murray as a customs agent with daddy issues.

i wouldn’t be surprised if many insults to human health are not relevant in population that exercises vigorously 5 times a week and have good body composition.

for the average american maybe we are looking for straws that break camels backs that are on the edge of breaking anyways

You are exactly correct - thank you for reading and letting me know, appreciate your curiosity!

Sorry - I believe I am off topic as this is not relevant given:

"This indirectly enforces the idea that sets cannot have duplicate elements, as set membership is defined purely by the presence or absence of elements. For example:"

So there is a constraint on what sort of trees are allowed in this -forrest- which would preclude most finite rooted trees.

A more compact and beautiful relation exists between integers and finite rooted trees exist, imo.

David W. Matula found a correspondence between trees and integers using prime factorization, and reported it in 1968 in SIAM: "A Natural Rooted Tree Enumeration by Prime Factorization", SIAM Rev. 10, 1968, p.273 [1]

Others have commented on it before, search the web for Matula Numbers

I independently found this relation when working on a bar code system that was topologically robust to deformation. I wrote a document that explained this relation here[2].

I created an interactive javascript notebook that draws related topological diagrams for numbers. [3]

[1] http://williamsharkey.com/matulaSIAM.png

[2] https://williamsharkey.com/integer-tree-isomorphism.pdf

[3] https://williamsharkey.com/MatulaExplorer/MatulaExplorer.htm...

I can intellectualize

But I cannot truly physically believe how much there is here in our universe.

It’s one thing to see zooming out animated diagrams

It’s a whole other thing to see these photos.

Imagine having a mind capable of holding every planet in every galaxy as familiar as we know our own.

The optimus bots seemed laggy. I thought the non verbal communication really suffered compared to a human using fingers and eye contact. The smoothness of the optimus movement was at odds with human movement - delicately but quickly darting around.

When a human makes eye contact with you and signals something with their hands - it is so responsive that you are certain that they are talking to you. With the robots, it was ambiguous.

Did someone put a low pass on the movements on top of a laggy remote control loop?

Is a fine not another way of saying - “no fair!- we want our cut!”

Please don’t do this. Reddit exists, keep HN high quality less jokey. If there is humor it should be hacker culture, not low hanging.

What are the dimensions of the devil's matrix?

Satan by Satam

It is interesting that operating systems exist for server applications at all.

What is the problem they are solving?

What is the difference between what an operating system contains and can do and what you need it to do?

Why would I want to rent a server to run a program that performs a task, and also have the same system performing extra tasks - like intrusion detection, intrusion detection software updates, etc.

I just don't understand why compiled program that has enough disk and memory would ever be asked to restart for a random fucking reason having nothing to do with the task at hand. It seems like the architecture of server software is not created intelligently.

Ai girlfriends are best positioned for tutoring success because they can motivate, withhold, etc.

Parents- get your son an AI girlfriend with a strong degree.

IO Monads solve this by passing the world around.

Yeah, the best world for multiple users is a database right?

So it would seem that if apps have stores to buy things one of those things should be a store?

If you could buy a database place then you can dispense tickets. Share the tickets with friends as photos. Then your local - first app can store take the tickets to the places your friends shared with you. Some of them go offline, fine.

I don't like the experience of having to setup a dropbox externally from apps. Why should a database place not be a one-off purchase like an item in a game?

Nevermind the security and ethical issues.

The implementation will be a slog, annoying, and distract from something that Microsoft ought to care about: creating a delightful, snappy experience.

Why isn’t recall a piece of software instead of an operating system integration?

Can apps spawn apps? Like, if you want chrome session recalled- open recall, then open chrome from there?

Why does every new feature need to enshitificate the operating system? Isn’t that an indication that the operating system doesn’t give app developers enough expressive power to create tools that they must go up the hierarchy and reach for OS modification?

Perhaps the whistle itself is contaminated.

If urelements may be distinct from each other, or the same, it seems like you could place the universe under a type and name it urelement, without creating a new axiom -- Except for, a urelement defined in this way could never be equal to the empty set - hence the new axiom? Am I understanding this correctly?

I'd be interested in if there was a gene that significantly reduced covid infection likelihood - but somehow didn't affect much else, like other viral infection rates.

That would almost be a natural sort of experiment, to compare the IQ of these and other population.

Though, you would have to have enough people in this group, not be too rare. And it would help if there was already a large population that was IQ measured prior to Covid, and could be re-measured.

It seems unlikely that such a scenario exists.

Interesting. The about page starts with a quote from Dada Manifesto 1918. The quote is changed, as explained in a footnote: "Updated to use modern pronouns."

Here is the original quote:

I speak only of myself since I do not wish to convince, I have no right to drag others into my river, I oblige no one to follow me and everybody practices his art in his own way, if be knows the joy that rises like arrows to the astral layers, or that other joy that goes down into the mines of corpse-flowers and fertile spasms.

changed to :

I speak only of myself since I do not wish to convince, I have no right to drag others into my river, I oblige no one to follow me and everybody practices their art their own way.

dada-lang about: https://dada-lang.org/docs/about/ Tzara, Dada Manifesto 1918: https://writing.upenn.edu/library/Tzara_Dada-Manifesto_1918....

This is a lovely exploration of color. Thank you for sharing.