ajani

I'm more impressed with his giant signature at the end of the post, than the post itself.

Rules, if taken as written, are never clear.

Try to define "furniture". You may say "Something to sit on". But what about a table. Ok, something to keep things on. But then is a soap dish furniture?

Furniture is a concept. It has fuzzy boundaries of meaning. And the meaning is only ever clear in context. And context is not just what is said alongside the concept. It is also the exchange itself in a particular situation.

When 14 years olds play ball in a "garden" area of a park, it's clearly disruptive. It can hurt someone, destroy the foliage etc. But when you play ball with a couple of 4 year olds in a garden, no one will suggest you stop. It is understood that a 4 year old in a play area with teenagers is at risk of being harmed, and so better to keep them in the garden area.

Rules, and the standardization of some of them into a system of law is problematic only if taken literally. As long as the what is written is understood to be a scaffold for actual meaning derivation from context there is no problem. This is the reason why judges, juries and courts exist - to interpret the law. And in the absence of formalization or systematization, common law applies. And at a very simplified level, common law is mostly common sense. Common as in shared among many. Common sense as in the sense and meaning shared among most of us implicitly.

I think this is a great idea! I'm surprised by the vaguely shrouded negativity in the comments. It's so easy to poke holes. This is great work and whatever problems arise can be dealt with at that time.

Here are some of the lazy thinker comments:

- If there's money involved, things go askew!!

Yes, well look: Google, FB, "open"AI etc. etc. What's new? The difference is this is a new approach and could potentially work.

- You make me sign up without showing me the stuff?! The gall..

Sigh...

- It's not pixel perfect on my special unique Moto-galaxy-iphone!

Really? You realize it's in the process of being made right? Soft launch it says. Recall how shit ChatGPT UI was, and it's not significantly improved since.

> This is one of the great things about the Internet. It’s so large that no matter how weird and unconventional you are, there’s always more poeple just like you. ... The central thesis is that what these villages can’t tolerate is a sustained large influx of strangers. A stranger in this context is a nothing more or less than an unfamiliar face.

How can I meet my mirror image freak if he's in another village?

You know what’s really truly private?

It’s what you said. The real world. Meet a friend. It’s as private as it gets.

Pick a small friend for true micro experience. :D

Jokes aside, unplug. And it doesn’t have to be that we build things at a world scale. But isn’t it amazing that we can and do build stuff at this scale.

Devs often don’t use tech as much as regular users! They build, and forget the use.

Aw come on now, don’t play coy, you know what I meant :)

That just as we went from hunting and gathering to farms and now cities, internet grew from a few people having access in universities and some tech corporations, to now phones in the hands of children in villages.

Let’s be more inclusive and dare I say progressive even?

I, for one, am not hankering to gather and hunt.

In ye olden days the whole world didn’t exist. Maps indicated ‘here be dragons’.

I feel you though, marginalia, and appreciate your work. I wrote a search engine too once and after a few months of me not using it myself, gave up. My results were too specific and not broad enough to include the world.

I wish! I wish!! I so wish this would happen. And the ‘type’ attribute to the script tag would finally mean something!

But nope. It won’t be. Just as var remains to support old webpages. Use strict to support newer features without affecting old ones. Imports. Etc.

But the post is factually off. In fact there is now an ordered map in JS. I think whoever is responsible for the spec and the implementations need to be complimented. Its actually shocking that it works so well. I have written painting software (much more computationally expensive than CAD software) using Canvas and WebGL apis and it’s pretty darn performant. Never native. But still very very good.

Although I don’t know about the vitriol in the comments toward old crocky. And his good parts book was good and useful to me in parts.

And to think ASD was entirely hand rotoscoped!

Ahh, most of the cost is CPU. It’s so expensive to have register memory, that you have l1 l2 caches. Each further away and cheaper. RAM is the cheapest which is why you can have so much of it. Try having 16G of register memory. UMA is a step towards that, but not quite.

Yes it is definitely worse. I submitted feedback a few days ago saying exactly what is being said here, that the model responses look like 3.5.

There are also very telling patterns of response that indicate a pre gpt-4 model.

1: All previous models suffered terribly if your chat got too long. After 20 or so responses it would suddenly start to feel less attentive to what is being said and output superficial or incorrect responses.

2: If you stop a chat midway and come back later to continue (after a refresh or a different chat interaction), it would often respond with code or suggestions that have nothing to do whatsoever with your prompt.

Both these patterns are sometimes evident in the current model. Likely then, there is some clamping down on its capabilities.

My suspicion is, this probably relates to computing resources. The 25 messages cap must mean that it’s difficult to scale its performance. And the only way to do so is to simplify the model activations with heuristics. Perhaps analyzing and preprocessing the input to see how much of the model needs to be used (partial model use can be architected).

This seems to be the simplest explanation of observed state and behaviour.

Assuming the start was a singularity is an interesting idea. Unlikely though. Light speed is constant in all frames of ref, even in singularities. So if a bang was accompanied with a flash then the flash must preexist at a constant speed. Which makes no sense since we are saying light came into existence along with everything else at the ‘start’. The ‘start’ must have a fundamentally different nature, where even time came into existence. You see?

> Passwords are productivity, not preferences.

Surely passwords are security?

I was excited to see this. My results were not convincing.

- I used "create voice", and the page refused to allow me to create anything because the big green button at the bottom was disabled. Only 1 checkbox shows up out of the 3 labels, and I was unable to check even the box that was visible. I used console tools to remove the disabled property from the element and it worked. (I'm using Safari so maybe it doesn't work there properly)

- The generated voice did not sound like me (I used my own voice). It did have some familiar tones, but not really.

- I fiddled with top-p, temperature and voice guidance, but the improvement was minuscule.

- Also recording the voice did not work (did record, but couldn't replay it to verify). So I recorded it on my computer and uploaded a file and that did work.

But it took no time for the universe to come into existence. In a bang and flash.

Making art of any form is about you, not about the art. When you learn to paint well you are creating you, the piece is just a side effect. Getting good at something affects the way you see the world. It creates meaning in you.

The one sided focus on the finished piece is the outcome of our culture valuing the art more than the creation of the artist. (The creation of you, not your creation.)

Agency. It’s about agency.

Social media, while not permanent, can be long lasting, and can have a chilling effect on a child’s future agency.

Guarding it until they can make their own decisions is what all good parents ought to do.

How is this not obvious and common sense?

I had heard of it, but never checked until you mentioned it in this context. It seems that Elm is an entirely new language written atop JS (or maybe not, and targets JS?). A DSL.

DSLs also usually are an incomplete solution. Unless they themselves allow adhoc language creation using the DSL as a base. Common Lisp with its macros allows writing DSLs quite off-handedly as you program. You grow your program and language together as you get deeper into your problem.

I can't help but sigh exhaustedly. There was a moment in time angular was the thing. And then x. And another. And React.

The problem each of them tried to solve, and continue to try to solve is purely lexical in the end. It's about writing things differently. Creating abstractions (terms) beyond what the underlying language/platform allows you to do.

And this is not a bad thing. The solution really is to create your own language so you can express the problem you are solving in the terms of the problem and not the underlying language/platform/hardware.

And then I sigh again... aah but if only Common Lisp and it's macros were the underlying platform.

Thanks!

I suspect the work is still ongoing. But maybe not on priority.

This diff from around 3 months ago shows it's likely possible to run it in the browser currently but with a lot of caveats (pretty great nonetheless!).

https://gitlab.com/embeddable-common-lisp/ecl/-/commit/6af4b...

Electricity is a natural phenomenon, yes, in the clouds, maybe elsewhere. But using giant turbines to generate electricity by manipulating the electromagnetic field mechanically, thus harnessing electricity to flow into wires and into our homes required a lot of invention.

We also invented a method to refine electricity into nanometer sized channels and sinks that enable bits that enable computing.

Nuclear fission is also a natural phenomenon. And we invented bad things with it. But also some power plants. Meltdowns aside.

Inventions aren’t good or bad. People are. If all soldiers choose to lay down arms the weapons lose their badness.

I don’t see ML systems as they currently are, to be nuclear bombs i.e. the end. But rather akin to the nascent forms of harnessing electricity with basic inventions and their basic functions.

Invention of electricity contributed necessarily to the creation of computing. And ml systems will contribute to many developments. Time will tell.

The coming apocalypse has always been us, not the things we create.

I tried looking for the webasm version of ECL, but couldn't find it. The official ECL homepage doesn't mention a wasm target. Do you have a link maybe?

We will get better at using ML systems. I find myself regenerating chatGPT's responses just to see if it generates similar responses. I ask it questions phrased differently to see if it is consistent. Discussing what code needs to do before asking it to implement it. And I get tremendous use out of it.

Electricity must've been scary when it was invented (harnessed?).

I agree. I too am scared by the idea of some random person playing around with genetics.

We don’t allow anyone and everyone to play around with nuclear technology. I wish no one had ever played with nuclear technology. We would have avoided a lot of harm already caused. And a lot of harm that is surely yet to come, because ‘yet’ is a very long time. Infinite even.

And maybe nuclear genetics will nonetheless develop in the hands of power seeking governments. But it’s still better than Steve from 2 streets over accidentally unleashing a strain of resistant bacteria on the neighborhood.

Hmm.. the ‘myth as an antidote to fear’ idea only makes sense if humans were on the ark. It makes more sense interpreted as a warning as to the coming future. The earth will survive no matter what, you may not. Making minds and hearts fear, not fear antidote.

Summary of the story:

Garmin to the rescue: Buy Garmin, and if ever you fall overboard, Garmin’s intuitive touchscreen interface will ensure that someone will find you.

Buy Garmin, save your life !

How does such obvious SEO spam (or maybe just an advertising piece for a gadget) get so much interest here?

I tend to do the exact opposite. Only when the program is fully done (if ever) do I document it (beyond peppered lines of documentation that explain something that is not obvious from code). There is good reason for this: Premature abstraction. As others have pointed out when I start to work on an idea, I’m not sure about everything. Working on it will reveal misassumptions and blind spots in my current thinking. I have also learned to abstract out behavior into functions lazily. Often having created defined functions with an immaturely developed idea is a hindrance to flexible experimentation.

Yes. Acute observation. It’s actually two separate indexes. A trigram index for sub word searching. Word* is exactly right and the other is a Unicode61 index with porter stemming (also remove diacritics). Trigram tends to work well with abbreviations and such. Whereas porter works well for general searches.

Thank you for the appreciation :)

It’s a linode. Shared cpu Plan.

1 CPU Core 50 GB Storage 2 GB RAM

It’s just $10 per month.

With linode even shared cpus are powerful and I’m yet to hit any overload. I’m sure it will at some point, I might upgrade then.