Hey you’re welcome to your view. In my view that’s an insanely disgusting devaluation of human life that’s been put into peoples brains by propaganda. Capitalism and materialism sure is great huh.

And Christians founded a nation where Black people were counted as a 3/5th of a person and my still living parents grew up in the segregated south - that was upheld by a Supreme Court made up of people who were religious.

Well here's a thing - I don't think it's capitalism which made me think this way, or even living in the first world. The first world in my experience is big on preaching the dogma that we're all special and valuable (whether it practices what it preaches aside).

Rather my view actually came from observing violence - between humans, but also in nature, and deeper still as a product of reality.

We exist in these fragile bags of flesh and bone. To a hungry lion we'd just be food.

And we behave the same - take for example the eating of meat. I'm a meat eater myself so this isn't anything preachy, but consider this - many animals are quite beautiful beings, possessed of their own personalities (we see it in our pets).

But that instrinsic value they have doesn't stop us from turning them into steaks.

On the one hand I wish you were right - I wish the world were beautiful and our souls along with it, and that I could sing that song.

But it doesn't always look that way. And worse still, I think it all started at the level of nature, not at the level of capitalism. We attribute too much significance to the human race when we lay such woes only at our own feet.

Curious why the model architecture wasn’t talked about at all? Did I miss that part?

pretty basic

``` from torchvision import models

# Avoid downloading pretrained weights; we load trained checkpoint weights. model = models.resnet18(weights=None) model.fc = nn.Linear(model.fc.in_features, num_classes) ```

https://github.com/mixfont/lens/blob/main/lens_inference.py#...

Nobody should have these types of problems in the age of AI agents. This kind of clean up and grunt work is perfect for AI agents. We don’t need new abstractions.

Version control systems are more important than ever with AI.

Meta doesn’t even need to use this, they’re just going to be constantly recording all video and audio from those rayban glasses ;)

Smart move, Zuck.

I haven’t seen tinygrad used for any mainstream production project or thing of value, yet.

Besides a lot of self congratulatory pats on the back for how elegant it is. Honestly, when I read it, it looked confusing as all the other ML libraries. Not actually simple like Karpathy’s stuff.

All that to say, I do really want it to succeed. They should probably hire some practical engineers and not just guys and gals congratulating themselves how elegant and awesome they are.

> Compliance isn't that hard once you stop looking for shortcuts and start spending time doing it correctly.

Trying to understand how someone can have this perspective when it’s usually someone’s full time salaried job in a lot of companies.

Maybe they meant "Not hard != quickly done". I don't think many people think bureaucracy is especially difficult. It's just time consuming.

But frankly if they meant that, the statement doesn't really say anything at all. Because what in this world is hard if you stop taking shortcuts and spend time doing it correctly?

Cold fusion.

A lot of that comes down to the costs associated with not being compliant and/or the requirements of existing contracts/insurance policies, where having dedicated FTEs to compliance is a requirement. Compliance might not be hard for the person/people managing the program, however it might seem difficult or complex to the FTEs that have to build to those standards if they do not have a security or governance background.

I assume they mean "getting a SOC2 report", which is the part that Delve attempts to automate. The maintenance of controls, adoption of new policy as the company evolves, etc, is what someone will do in the full time role and that Delve et al would do nothing to assist with.

There are more productive ways to vote with your money than tax evasion.

You can make tax-exempt donations, or start your own non-profit organization.

Some people hoard money without building businesses, without participating in government, without contributing to welfare. People who take more than they give are assholes.

> People who take more than they give are assholes.

100%. I hate welfare leeches too

Most tax money goes to social programs. Especially at the state and local level.

You're conflating "taxes" with federal taxes.

In my state (NY), I pay income tax to the feds and NY state. I pay property tax to my county and town. This pays for things like roads, cleanup and maintenance, the school district, the library, the parks and sports recreations. The community trails and wildlife preserves.

The federal government can basically print money. The only reason they "need" your tax money is to limit inflation.

You’re not wrong, unchecked inflation is bad for most people though. Stable currency is pretty important for trade and economic stability. Unless you prefer heating your home by burning stacks of cash

Oh, I agree. I never said unchecked inflation was at all desirable or even ok.

My point was that local and state governments do need your tax dollars, in the sense that that is literally their income. But for the federal government it's different. If federal tax revenue declines, they can just sell more treasury notes and continue to spend as much as before. In that sense, federal tax revenue has no direct effect on federal spending.

You know that’s not the entire budget right? You’re being an asshole by denying funding for disaster relief, schools, healthcare, roads, scientific research, all the public goods and services that don’t work on a profit driven model, but you still get a direct benefit from.

If you want to play concerned citizen get out and protest, vote with your dollars by not throwing them at big tech companies who kowtow to politicians and fund their campaigns. But if you think you’re sending kind of message by withholding your taxes, it’s really just that you’re a selfish asshole.

> vote with your dollars by not throwing them at big tech companies

Abstaining is not voting. If you want to vote with your dollar, spend it actively undermining big tech companies. Get out there and blind some cameras or something.

> Abstaining is not voting.

Fair if you’re already not giving them money. But if you manage a sizable chunk of cloud spend at AWS, GCP, Azure etc, you can send a meaningful signal by taking away that revenue and shifting it to a company that’s not aiming for neo-feudalism.

This reads like a scattered mind with a few good gems, a few assumptions that are incorrect but baked into the author’s world view, and loose coherence tying it all together. I see a lot of myself in it.

I’ll cover one of them: layers of management or bureaucracy does not reduce risk. It creates in-action, which gives the appearance of reducing risk, until some startup comes and gobbles up your lunch. Upper management knows it’s all bullshit and the game theoretic play is to say no to things, because you’re not held accountable if you say no, so they say no and milk the money printer until the company stagnates and dies. Then they repeat at another company (usually with a new title and promotion).

My friend, there’s tons of evidence of all that stuff you talked about in hundreds of papers on arxiv. But you dismiss it entirely in your second bullet point, so I’m not entirely sure what you expect.

I’ve read dozens of them and find them unconvincing for the reasons outlined. If you want a more specific critique, link a paper.

I personally like and use tests, formal verification, and so on. But the evidence for these methods are weak.

edit: To be clear, I am not ragging on the researchers. I think it's just kind of an inherently messy field with pretty much endless variables to control for and not a lot of good quantifiable metrics to rely on.

Let me guess, it’s ok if they do it, but if you handed their crappy ticket to Claude and shipped whatever crud came out, you’d be held accountable? ;)

Funny how that works out.

I've heard a great thing recently, more or less - If all you're doing is writing prompts, maybe you're not needed anymore. Stay behind the intent, own the output and understand it and then maybe it makes sense. sloppy prompt + c/p doesn't bring value and will be treated as such. As with anything in life, outcome is usually proportional to the effort put in.