emorning3

I dont understand why anyone would even think that the current crop of AI is capable of planning or reason on its own.

Transformers with memory would be different story.

But, no memory, no capability to reason. End of story, right?

Yknow, sometimes I think that the best thing for America would be to just wipe the slate clean.

We could do better. Maybe we just need a fresh start.

And I honest to god think that Trump and Musk are gonna burn our world down to the ground, piss on the ashes, and then demand payment for doing so.

And I dunno, somehow it all just feels so right.

I hope Musk has got double B working on a SOMA recipe for us...

PS: Not voting is voting.

No real American has the right to sit on the sidelines anymore.

What the insurers are doing is all legal, what do you propose exactly?

Do you think that the current administration is going to 'hold executives accountable'?

Trump could literally drive over to your house and shoot you in the head for even suggesting such a thing.

BUT BUT BUT Floridians voted for DeSantis, so its all good.

So now he's busy protecting insurers from being sued.

So nothing to see here. Just the government that you voted for collecting their bribes and protecting their friends.

Floridians are gonna get exactly what they deserve from the new administration.

LLMs cannot reason, they can only say things that sound reasonable, there's a difference. Duh.

Guns are an inefficient/stupid way to kill people anyway.

Just ask Russia and Ukraine.

Look around, human beings are quite clever.

My Trumper brother loves his union and Trump.

I've asked him to explain how this can be and he cant, he just accommodates the contradiction with a lot hand-waving.

So I think that there are enough Trumpers that are OK with unions to make a majority that support unions.

Basically, Trumpers can choose to believe anything they want, anytime they want, for any reason they want, or for no reason at all.

They do that in order to be able to rationalize their irrational, and often detestable behavior to themselves.

Looking over the list I determined that I already regularly follow all the rules but the first three.

But, as a C#/C++/Typescript developer, I don't have the first clue how any I might go about implementing those rules.

Anybody know of patterns for this kind of safe programming that I can use?

Does just using a language like Zig automatically make my code 'safe'?

Anybody know if generating code from a proof checking language like Lean would satisfy all these rules?

Liars lie, news at 11

Yep.

There's gonna come a time when Elon's gonna tell the AI to tell us to push all the buttons, just to see if we'll do it. And I'm pretty sure we will.

My conclusion from all that I'm reading lately is that LLMs cannot do deduction but they can fake it real good.

I mean, you wouldn't use this brand of AI to plot your path to Mars. Well, you could, BUT you'll also want to validate the path or risk dying.

But this AI is good enough for Elon and his ilk. Because Elon's not gonna get into the capsule, you are.

Because you are not the master of this AI, you are the validator.

No. I'm gonna murder the shit out the next trump fuck that lies to my face with that trademarked shit-eating smile.

I'm a software developer with just a basic understanding of LLMs.

But its not surprising to me that LLMs wouldnt be good at composition and planning. To me, LLMs seem like half a brain, the right half, that does pattern matching and such.

To me, the current crop of AI seems to be missing the other half of the brain, the part that's good at planning and composition and such.

LLMs can tell me all I want to know about logic and deduction, so its funny that LLMs cant understand their own words.

I think you meant its not a crime if you can afford it.

Excellent point. Multi-methods are awesome.

I remember Apple giving away Dylan CDs when Java was first released. I learned them both and went with Java. But I should have taken the opportunity to also learn the Common Lisp Object System.

These days we like to bitch about OOP languages, but we could have had better.

I was on a team at Citibank, in the late 1990's that built an ATM that was a full-on virtual bank that you walked around within and transacted your business.

Nobody liked using it and we were eventually all fired. The end.

The DOJ is planning to sue RealPage...

https://www.politico.com/news/2024/07/12/justice-department-...

"Oh we failed to design the perfect solution, so we are not going to change anything until we find that perfect solution."

Did they think that generics were perfect? I doubt it. Its a community that's been burned by its past pragmatism. I think its wise to be more cautious in response.

Odata is 80% of GraphQL with only 20% of the hassle.

In the late 1980's I rewrote a very large pension actuarial forecasting system, written in Fortran, to Basic.

I'm very not sure, so I'm guessing that the system had several hundred thousand LOC.

The idea was to stop spending a fortune on mainframe time and run pensions valuations on a microcomputer.

The Fortran system, running on a mainframe, might take 60s to run a valuation.

The same valuation would take a full day to run on a microprocessor, but it worth it because it reduced costs by a lot.

Thank you for pointing that out.

I just happen to be sitting here reading HN instead of fixing a deadlock I've run into.

I have a ton of async code and I just now noticed that my projects don't use this analyser.

I dont know why it's not included by default but I'm gonna add it for sure.

If AI was a real thing, instead of phony baloney probabilistic bullsh!t, then we'd already have AI-driven tools that take crappy, bloated, slow, but correctly working systems built by humans and turn them into smaller, faster, correctly working AI-enhanced systems.

Right?

Using Tailwind is having an effect on how I build components.

I've really fallen in love with Tailwind and atomic design.

And I've discovered that Atomic design can be more than just for styles, but for behavior and content generation too.

Here's a snippet of an annotation-driven version of Razor that I'm working on. This snippet composes a table of weather data using atomic annotations that add styles and behavior.

```

@@<Spacing MarginTop="@Theme.Spacing.Large" />

@@<Placeholder Show="@(!(forecasts?.Any()??false))" Variant="@Placeholder.Skeleton" />

@@<DataAnnotations All="@true" />

@@<PlQuickGrid />

<QuickGrid Items="@forecasts">

    <PropertyColumn Property="@(c => c.Date)" />

    <PropertyColumn Property="@(c => c.TemperatureC)" />

    <PropertyColumn Property="@(c => c.TemperatureF)" />

    <PropertyColumn Property="@(c => c.Summary)" />

</QuickGrid>

```

Here's what the annotations do...

@@<Spacing> - adds TW class style equivalent to 'spacing-[--my-spacing-token-l]'

@@<Placeholder> - displays a skeleton placeholder, instead of the table, when loading data

@@<DataAnnotations> - extends the QuickGrid with formatting and label metadata found on forecast Type

@@<PlQuickGrid> - Applies custom component styles to a Blazor Quickgrid

The first three annotations are reusable across a wide variety of components.

Tailwind and atomic design have really taken hold of me, I'm gonna go all the way down that rabbit hole :-).

bwahahaha, I love that!

The Circle of Descent continues....

You weren't alone, but we just used more hands.

I usually did navigation, and my grade school aged son did the shooting while sitting on my lap.

Now he's 33 and I'm 64, we'll have to switch places.

I've played around with Lowdefy, https://lowdefy.com/. I kinda liked it. A similar concept but uses YAML instead of JSON.

At the end of the video the narrator states that a 3x3x3 structure can support 9000 newtons.

I don't know what a newton is but google tells me that's 2000 lbs.

Do I understand that correctly? That would be incredible!

And to me, would make the design of the voxel more interesting than the robots.

>> Can we get to no hoops? That's the intriguing challenge! <<

That's definitely a challenge :-)!

I'm unfamiliar with the concepts of institutional analysis but definitely interested in learning more.

I'm also interested in the Functional Source License.

I'm skeptical though! I live in a country where corporations will literally make people work for food if they can. It's very hard for me to imagine how an open source project gives away the thing of value that it produces and only afterwards gets the money it needs to survive.

Solving the open source sustainability problem would actually be world changing.

For now I'm following you on Twitter and look forward to future posts.