patrick451

The optimal control framework Casadi has had the ability to compute sparse jacobians and hessians for a long time (maybe a decade?), which come up all the time in trajectory optimization.This not only provides massive speed ups in both the differentiation and linear solver time, but also greatly reduces the memory requirements. If this catches on in machine learning, it will be interesting to see if we can finally move past first order optimization methods.

This only works if you don't need real time performance. For anything that does need to run in realtime, you have to be damn sure that whatever your initial approach is can be optimized to run at the required sample rate. Otherwise, you will be ripping out the entire design and starting over. You might as well optimize first.

Same. After learning that Wikipedia is not, in fact, about to go off the internet for lack of funds, I came to hold a very dim view Wikipedia organization.

Theoretically, it could have been an advantage, but in practice it's not. What we got instead is a software ecosystem of slow, bug infested garbage that requires constant updates.

If I waste 5-7 minutes multiple times every day, that quickly adds up. It's easy to waste an hour or more fiddle farting with these unreliable tools. It's easier to just do it myself.

The frustration of using an LLM is greater than the frustration of doing it myself. If it's going to be a tool, it needs to work. Otherwise, it's just a research a toy.

You have to balance this time savings with all the times the LLM is just making shit up. I can't the number of times I ask "how do I do X?", and says "call function Y" or "Use command line argument Z". Turns out Y doesn't exist and CLI doesn't take argument "Z". The frustration from that lying is immense. I'd rather take a little longer on some tasks and never deal with this lying.

Previous layers of abstraction were largely reliable and deterministic. Compilers don't just randomly generate assembly. Yes, all abstractions leak, but most of them just drip, while whatever "abstraction" AI provides leaks like a sieve. It would be like if CDK just created or deleted random resources 50% of the time.

Contractors have to do the leetcode monkey dance for prospects?

Sometimes I write code like this. Then I delete it and replace it with a for loop, because a loop is just easier to understand.

This functional style is what I call write only code, because the only person who can understand it is the one who wrote it. Pandas loves this kind of method chaining, and it's one of the chief reasons pandas code is hard to read.

I never use them to generate code for languages that I know well. Google has started trying to answer my programming related searches with LLM results. Half the time it points me a useful direction, the other half it's just dead wrong.

I have found them pretty helpful writing sql. But I don't really know sql very well and I'd imagine that somebody who does could write what I need in far less time that it takes me with the LLM. While the LLM helps finish my sql task faster, the downside is that I'm not really learning it in the same way I would if I had to actually bang my head against the wall and understand the docs. In the long run, I'd be better off without it.

> We have observed a ~13.3% slow-down, which is significant enough to warrant a second thought on whether the bounds checks were a good idea in the first place.

Wow. I will think twice the next time I reach for .at().

>Natural England has developed a detailed licencing regime and application process to make sure that stakeholders are engaged and landowners are supported.

What a bunch of dissimulating bullshit. If they actually wanted to support landowners, they wouldn't be re-introducing this destructive species.

>whoever comes across your clever code next will probably want to hit you, especially if it's not a dev from your immediate team, and especially if you have succumbed to blurring the lines between your idk code and your app code.

If you want to maximize the number of people who have a chance of understanding what is happening, python is your huckleberry. They are going to want to hit the guy who wrote everything in a bizarre language called HCL that nobody outside of infra has ever seen or heard of.

> If they instead come across some bog-standard terraform that maybe has a bunch of copy-paste and is a bit more verbose... Who cares? Its function will be obvious, there is no mental overhead needed.

"bog standard" is doing a lot of heavy lifting here. You can write simple python or esoteric python and you can write simple terraform or esoteric terraform.

Do most individuals who oppose AI doing so out of "just" self interest? Maybe. Not wanting to starve is inherently self interest but it's fundamental to not just human nature but every living thing to avoid starvation.

AI also has the potential to cause unimaginable harm to the entire globe.

If they are so easy to identify, why not just make it a JIT error. Manually inspecting all of this sounds awful. I'd rather my compiler just do it for me.

Let's forbid the police from lying to everybody.

Software devs are not alone here.

Have you ever tried asking a doctor how much something will cost? Or if the thing they want to bill you for will even work? They have no fvkin clue.

I still find seatbelt laws to be one of the most ridiculous examples of government overreach. People think the government shouldn't be in their bedroom, but for some inexplicable reason have no problem with the government being inside their car.

The only thing more frustrating than a rustacaen redefining "safety" to suit their purpose is a Stallman discipline redefining freedom to mean something totally weird. Something in the public domain is not free? Give me a break.

Calling writing for just humans irresponsible is quite the claim. Personally, I'd rather AI did not injest anything I write whatsoever. Not that there's much I can do about it.

We use some of these terms in the U.S. too, but they are less common and/or mean something slightly different. Some carpenters and millwork companies will call casing architrave instead, but this seems be mostly used for classically inspired, high end work.

I think I've heard some carpenters call baseboard skirting, but the most common usage of "skirting board" is restricted to the baseboard like piece on the sides of stairs.

We have a term cove molding, but it is a particular style of crown molding.

A rail at 3/4 height we also call picture rail. I don't think I've heard of dado rail though.

I'm a controls guy, not a quant guy but I've found the SDE perspective and this blog post to be incredible in helping me understand how a diffusion model works.

> The new-school guys want to make things better in a new way.

The new guys definitely want to make things different, but it seems there is a lot of debate over whether it will actually be better. Really, they should just write their own their kernel. If rust is really that much better, they'll will.

I think obscuring details to foment emotion is lying regardless of what the chart is about.

I agree data visualization be used for emotional manipulation. But that is unethical. Dressing up rhetoric in the trappings of science is one of the many reasons science has lost tons of credibility in the last 20 years.

Charts should be about hard data, not rhetoric.

This is a fixable problem if we can get congress to roll back the insane KYC laws.

Evoking emotions is the role of art, not data visualization.

> It is completely reasonable to give up some precision to:

> - Reduce unnecessary cognitive load and emphasize a feeling, as in the Climate Stripes example.

> - Encode a much higher volume of data, as in the Climate Globe example.

> -Add delight and use a visual metaphor to show trends, as in the Coffee Shop example

This the most absurd claim I have ever seen regarding data visualization. Give up precision for delight? Give up precision for a feeling? This article should have been tile "How to lie with graphs". The article is advocating for how to use graphs in the service of rhetoric, not science. This is exactly why visualizing data should NOT be treated as art. The first and third bullet point are obsolutely not reasons to give up

Depending on the context, dot and times may have different meanings.