steve_adams_86

Thinking In Systems: A Primer by Donella Meadows

My 17 year old son bought it for me for my birthday. I've been enjoying it a lot. I do in fact like thinking in systems and learning more about it, and reading it in a more generalized fashion than I tend to in the computer science world has provided a refreshing perspective. It's all very approachable, nothing very complex, but gives names and discreet patterns to things I intuitively understood but previously wouldn't articulate easily. Definitely one of my favourite gifts in a long time.

Stories of Your Life and Others by Ted Chiang

I read bits of this to my wife at night occasionally. One of the first trips we went on together was camping on a little island, and I had Kafka on the Shore (Haruki Murakami) with me which I suggested I read to her in the evening. She really liked it and I've done it since. I wish I did it more often. Anyway, Stories of Your Life and Others is a wonderful collection. I've read it a few times, not always completely, and I enjoy it. Story of your life is the standout. It's what the movie Arrival is based on, which I consider one of the best scifi movies of all time. It's a beautiful story.

Silo series by Hugh Howey

I've been watching the show and decided the books deserve a shot. If the show is as good as it is, the books must be great, right?

I had a lot of fun figuring this out and getting my system working, but ultimately the UI is too buggy (sometimes up to 3 clicks of the same button to finally get it to do anything, whether it be ordering chemicals, switching a tab, dispatching...) and it was chewing through my battery incredibly quickly. I gave up because the rough edges are a bit too rough.

This is kind of what it was supposed to do, in this case

> who cares how many sit-ups LLMs have to do?

At large enough scale, I suspect it'll become a problem. At the moment part of the appeal is how effortlessly models absorb menial work, but I do wonder if eventually it'll be too wasteful at large scales to sustain it. Maybe not, though.

It's the framing. Wolves, sheep, and gypsies. "Standing idly by while your livestock is devoured by predators or your neighborhood is taken over by migrants"; migrants are not inherently predators at all. North America is built by them. DHH implies anyone from outside is only arriving to violate and destroy what existed before them. This is not at all how it works.

Alongside this framing is seemingly no admission or acknowledgement that crimes have been and still are perpetrated by "normal Brits" as well; the focus is entirely on the barbarians coming from outside to maraud. It's as though he attaches criminality to foreignness.

He also steps all over the line of what a "native" Brit is. Seemingly not someone born there, but, according to statistics he is certainly referring to white Brits.

The list goes on. The man is racist. I don't think being anti-large scale immigration is inherently racist at all. As a Canadian I have substantial concerns about how immigration is conducted, and I don't consider that racist. But very, very little of what DHH thinks, or how he frames his ideas, ever crosses my mind. I don't think about preserving 'real' or 'normal' Canadians. I think a lot about how immigrants have made my country better. I don't internally or explicitly associate crime with immigration. It's much different.

Humans have assisted each other medically, supported injured and disabled tribe members, and sought out medicinal herbs since long before history was written. I'm not sure how you'd delineate between modern health care and when humans never had some notion of health care and/or support when bodies enter failure modes. We don't survive well without a lot of assistance and care, and living long lives required plenty of bodily maintenance regardless of how fortunate and healthy we were.

Yes, extremely worried. It seems we're on a path to brand new kinds of weapons of mass destruction, and arms races in mass parallelization. How could anyone slow down?

I'm neurodivergent (got all the proper stamps and papers and everything) and I think it's very useful in the right places

> The LLM likely needs to be reminded of its abilities

Like when it tells you something is 3 days of work but it can do it with some degree of guidance in a couple hours

From a practical perspective this makes perfect sense to me. But we do know the photons travelled over a period of time, right? Where those photons came from, ‘now’ is something else. Though it doesn’t matter practically, those events did occur in the past, before they were a part of our now. Or am I missing something?

As I recall, I went with it because it supported the best dive features. I used it for that quite a bit for a while, but eventually realized my basic cheap dive watch worked fine, could take a beating, and was actually better rated for deeper dives. Having physical buttons is a huge advantage over the Ultra. The dial was never enough to get the control I wanted, even if the UI only required it.

I was on the Ultra for 18 months or so and came away from it feeling like a much less powerful and capable Garmin is much nicer as a daily watch. I found myself wanting to find reasons/excuses to justify the Ultra. Ultimately I couldn't, and it wasn't worth its cost, charging schedule, and lack of battery life.

I can see why others like them (they want WatchOS and Apple-specific capabilities). That stuff is useful, but not often enough for me to care.

Really, the watch was a regular reminder of how functional life was before the watch existed, if that makes sense. It's a genuine marvel, and yet... I got by (and get by) great without it.

> Providing ready-made UI components and templates has never been a million-dollar business, nor should it be.

But what if people were already selling them, and tailwind showed up, did it better, and people went directly to them to buy what they were offering?

Why shouldn't that exist?

The folks over at Tailwind offered curated, hand-built components that were readily adapted and tuned to your own needs. They did an excellent job making it so you could pick and choose what you wanted, then theme all of it from your tailwind settings with ease. It was excellent.

Sure, similar things existed for less or at no cost. I was content to pay for this, though. It made my work meaningfully simpler at times. It paid for itself.

> Tailwind's business model existed due to a massive inefficiency in the frontend industry not being able to come to terms with CSS or native components.

I don't agree. The pain it addressed on my team was more so that very few people I worked with understood CSS, but tailwind was easy enough for them to work with (particularly in a React context). It was a useful tool in a context where a lot of the people didn't fully understand the tool. They could just pick and choose components and they'd slot into the app well enough. It addressed real pain there.

They surfed a wave for sure, but I think it was mostly that a non-trivial number of people working on the frontend don't understand HTML and CSS very well. That's not their fault at all.

I agree, more context would be helpful.

In my mind there are two components of this that are in direct tension though, regardless of numbers:

1. Humans having to review ad content is legitimately traumatizing and awful

2. Automated systems can miss obviously harmful content, which should never be allowed

To me this means we probably shouldn't permit these systems to scale as they do, because the company profits from harms to its own workers and/or ad viewers. I worry that we've normalized a business that shouldn't exist in its current form, but it prints so much cash that no one is willing to stop it.

I'm torn on this.

I do want safer vehicles and I do see a path to get there with this technology.

What I don't want is companies training their technologies with legitimate risk to the public, with a clear pathway to profiting off of this risk and public data, with no benefit to the public _unless_ they happen to succeed, and to some degree, unless you can afford their products. Something feels off about that.

Of course, everyone benefits if vehicles are safer, but it does feel like there is an extractive factor regardless. These companies benefit from public resources, are wholly enabled by public resources, but are trying to create a situation in which the technologies they create are more or less required to be used in order to utilize the public resource. Does that make sense? I suppose it's inferred, but, I don't think it's unreasonable to believe any car manufacturer with a working FSD technology would LOVE to have that technology mandated to be on every new vehicle. Imagine leasing that to every manufacturer who didn't develop it themselves so they could continue to sell in the USA.

On one hand, safer vehicles are a MUST because we allow an absurd number of people to die due to unsafe driving every day. On the other hand, I don't feel right about private companies benefiting so much from roads the public built, while training their technology with some degree of risk imposed on the public, and with no clear desire in these companies to give back. They tend to pay paltry sums in taxes, if any. Rivian was at a net loss of over 4 billion for 2024. I understand they're trying to build something that requires massive capital, but, I also understand that these companies are experts at operating at a loss for prolonged periods of time. Amazon is a stellar example.

This definitely cuts both ways, but I'm not sure exactly where I stand. I need more information to decide, but I'm confident that there is some degree of exploitation of public data and resources here. I'm not trying to imply Tesla, Waymo, or Rivian are the only ones guilty of this either. The tragedy of the commons is a well-known phrase because it's such a typical phenomenon. I'm also not suggesting that there's a simple solution. There can't be because the fragmentation of the polities in which these companies operate and train their technology within.

I wonder if this pertains to crop trees. A lot of crop trees are trained to stay smaller or grow in helpful shapes, such as with espalier training.

> Considering the amount of ads they handle that seems... reasonable?

Why reasonable? Don't these ads pass through pre-publication review, and Meta makes money off of it? Any number above zero is a serious problem. If they can't run an ad business that doesn't publish and profit off of CSAM, they shouldn't run it.

I don't think we should distinguish between 'fake' and real CSAM based on if kids are hurt; this ad machine shouldn't be able to publish it, regardless. The moral question of if it's real or kids were directly harmed seems separate to me.

Not OP but my guess is that they're already using bad AI images for their kebab business, and OP wants them to use better AI for their images (so they may actually lose fewer customers because the AI imagery isn't so awful)

Yeah, important to note that the way the label is used has deviated from its historic meaning. Colloquially it’s something like “someone who rejects new technology” with variable reasons as to why, but historically there’s more nuance and it’s grounded in rejection of technology which worsens working conditions, detracts from craft, and disrupts cottage industry/workers rights/wages/etc. They made more sense than the label presently suggests they did. I suppose you could think of them as activists protecting their crafts and trade privileges that were being undermined by specific machines and their owners

PMTiles is the underlying technology I believe. We use it on oceanconnect.ca and in my opinion it’s pretty awesome. Very efficient to store and retrieve, and cut our S3 PUT request costs by something like 90% across the project.

I wrote a while back about some coho fry populations I found which would have benefitted tremendously from dams. Each day I returned I would find more and more empty pools along the river, picked clean of fish where I’d previously seen dozens or hundreds of fry. Beavers used to live here in great numbers, but these days it’s fairly uncommon to see them. I bet they saved a lot of coho back in their day.

https://steve-adams.me/coho.html

There are only papers which suggest this could be true, none that conclusively prove it. That I’m aware of.

There are some cool studies done where people with a without ADHD are tasked to forage berries. The people with ADHD do better and adhere to foraging patterns which existing theories suggest would be more optimal.

A lot of the studies like that indicate the individual with the condition might benefit more from ADHD than the group they’re in, but I’m not sure that studies have explored group benefits very much.

There’s an often cited paper in which nomadic men who carry a certain ADHD-related allele (DRD4-7R, a dopamine reception variant with ADHD traits and novelty seeking) tend to have better nutritional status. I can’t recall much about it or how high quality it is.

I wonder why Cloudflare didn’t think of this

I was only 2 in ‘88, but the story is the same from my perspective. I think so much about how to get my city to incentivize more manufacturing industry. The small amount we do have is the most lucrative use of the space in the municipality on a square foot basis, yet the city makes it far more difficult than is necessary to start these types of businesses. Worse still is that this type of space, relative to population, is decreasing when it should be increasing.

We have SO MUCH potential in Canada. It’s incredible. The only thing stopping us is ourselves.

What drives me nuts too is that my kids find work in places like tourist traps and Tim Hortons. When I was a kid I got to learn to build things, use tools and machines, use my body, plan projects (albeit very menial ones in my teens), and figure out what the hell all this abstract stuff in school was really for. I was applying math, drafting simple designs, writing to customers about orders, and using skills I hadn’t found real-world uses for up until that point. They’re missing out so much. But our city is, too. My kids are worth far more than serving coffee and frying fish and chips for tourists. And they’d be worth even more in ten years if real opportunities existed here. They’d have knowledge and skills they’ll never find in those places.

It really drives me crazy. So much wasted potential. We can do better, but we need policies that enable us to do it.

Even 10% would concern me.

The comment is a question; I’m not sure why one answer or another would make it irrelevant.

Haven’t they actually done a relatively good job in this regard in many ways? I can still use my 12 year old MacBook Air. It’s not super current and not all apps work, but it’s actually still a decent Apple experience. Same with our old iPad. They kept it updated way longer than I expected.

Definitely. I was only responding to the statement that sugar is a necessity for energy if you’re moving around a lot. You can do this without sugar.

Kind of; humans do adapt remarkably well to low-glucose diets, and we can produce glycogen without sugary inputs via gluconeogenesis. They do make it much easier, though.

It’s quite poor given that you don’t get to have a life outside of work, too. I’d start to think that sounds great if it was a 4 day work week, but 7… Bizarre.