perrygeo

Creatine is a health fad? No other supplement has been this widely-used for such a long time. It was considered safe and effective when I started using it 30 years ago. The benefits are/were well established, the mechanisms well-known, the safety is impeccable. If anything, it's the most well-established supplement on the market - the exact opposite of a fad!

Ironically that's where I started my career and now I'm hoping to get back there. I was a data-oriented environmental scientist who happened to code. But at some point, I slipped into 100% software development roles and I never got over how strange it felt to be siloed, isolated from the domain knowledge on purpose. It was far too easy to "win" the tech struggle of the day but produce nothing of actual value.

Dedicated coders were somewhat necessary to wrangle the tech stack (which itself grew more complex to accommodate more devs, a positive feedback which nearly spun out of control in late 2010s). But now? A single coder with domain knowledge has a shot at wrangling their own tech stack as a part time job, the rest of the time focussed on IRL value. This is the future I want.

My heaviest LLM usage month came in at $45. So far this July, $6 and counting (it's been a light month). I can certainly imagine increasing my usage by several orders of magnitude but ... why? If I have something that needs to get done and an LLM can do it, great. But I'm not sitting here inventing reasons to waste money, which is apparently what the tokenmaxers are doing - as evidenced by the astounding lack of value produced by all this vibe coding.

It's not just "the code itself looks LLM generated" - it's also LOC/hr by a particular author which suggests vibe coding. You could look at the author's github contributions to identify time periods when the author was generating code at super-human speeds. Combine the two signals and you might get something better than a pseudoscience?

Love it. Another great site along these lines: https://solar.lowtechmagazine.com/

Point datasets have two distinct modes of visualization. First is an aggregate view which serves to show you the trends and spatial distribution. Second is the individual view showing details about the point itself, its attributes, etc.

Clustering (for all its faults) is the only off-the-shelf technique for seamlessly switching between these two modalities without having to change the underlying data representation. Need more detail? Zoom in. And the zoom level is adaptive so it works with any scale.

There are better aggregation techniques (summing to a hexagonal grid, heatmaps, etc) but they generally require a separate calculation (possibly server side) and then switching to the raw source for the individual point view, either manually or at some hardcoded zoom level. It's not the same experience - it feels like two separate map layers instead of one integrated clustered layer.

This is mostly a matter of what's available in the mapping libraries. You could imagine building an alternative to clustering that calculates a heatmap on the fly when zoomed out, eventually revealing points as you zoom in. But presently this is something you'd have to DIY. For now, clustering is the only thing that works right out of the box.

Ask your favorite GPT to generate manifests, ... Before LLMs writing consistent YAMLs was PITA but today on low/development scale it's pretty much free lunch.

Writing manifests seems like a trivial thing to focus on. Who operates the k8s cluster in production? Who runs upgrades? Who's on call to monitor the system? Of course if someone else is doing all the work for you, it feels like free lunch!

This is a consequence of our "AI accusation culture" that has arisen in the last year. An obvious reaction to the "AI slop culture" (seriously, stop doing that) but its a shame to see legitimate creators caught in the crossfire. This is not the first human-written article that's been wrongfully accused and it won't be the last.

I don't think people fully grasp the scale of a trillion dollars. A trillionaire is a million millionaires.

It's impossible to efficiently manage that much wealth in one brain. Distribute it!

I'd run an essay contest. Every year the best thousand ideas are selected by yours truly and get funded at $100 million each. Even then, it would take a decade to burn through one trillion. It's a tough job, but I accept.

At what point does this become an issue for data quality and global epistemology?

It seems inevitable that we ask for more AI assistance on topics we don't understand. And therefore have the least context to correct. Result: a flood of poor quality information.

In areas we DO understand, we'll either not ask AI at all, or treat its results with a higher degree of skepticism. Result: a lack of high quality information.

Inevitably this means a higher volume of non-expert prompts gets translated into the next generation of internet content. AIs are pumping out more novice-level text and less expert guidance.

The result will be an internet full content written from the perspective of an ignoramus; not addressing any complex issues, staying surface level on every topic. Which will cascade into future models, etc.

Learn SQL (because it's basically the only option) but much more importantly, learn databases. Know why atomicity, consistency, idempotency, and durability matter. Understand the wire protocol and the client-server model. Do relational data modeling; think beyond databases as a dumb store. Join. Know when to normalize. Internalize indexing strategies. Think deeply about what work belongs on the database server (work that can leverage relational set theory) and what work stays in the application. Once you figure the true capabilities of databases, SQL as the language interface is a side note - about as important as the leather on your steering wheel.

Navigating and manipulating code as a tree, directly in your editor. There is nothing like it.

Going back to code as bunch of carefully arranged ascii chars feels like a regression.

Friction is the mechanism by which mammalian brains acquire skills. This is as close to proven as we get in cognitive neuroscience; without a struggle, we literally don't learn. This is not a controversial statement. The only way to improve our cognitive skills is to intentionally add friction - aka practice. Use it or lose it.

This isn't necessarily anti-AI. Imagine a tool that could quiz you, provide context for decisions, and make sure you're up to date on your knowledge of the codebase - instead of just writing code for you. IOW an AI-based system could intentionally add the right kind of friction to improve understanding.

yes, sorry its an important distinction. Especially raw whole fruits since they are packed with fiber and nutrients and hard to overeat.

Yep. If there was one single thing that literally every person should do for their health, that is to greatly reduce or completely eliminate sugar. The evidence is overwhelming.

The evidence against seed oils is not quite as convincing. I see seed oils as a low quality food to be avoided - goes rancid too easily, requires chemical processing, etc. - but it's not strictly poison. These oils are in virtually every industrial "food product" which makes them unhealthy by association. Stop eating highly processed crap and you'll see the benefits - cutting out seed oils is a side effect.

I watched people ask LLMs for linting/refactoring help, burning easily 5 minutes for something that could be completed deterministically, locally, in ms using any modern editor.

Quite frankly it was embrassing. We've had tools for static analysis for ages. Use them.

Someone with better knowledge could work 100x faster using 100x fewer resources. They did it the slow, expensive way but at least didn't have to think? Odd flex.

Another framing is "Prioritize". It's not yes or no; it's what are you focusing on right now. There may be plenty of good ideas that are "no" today because you can only work on one at a time.

The unit you plan in is the truth you will or will not face.

This 100x. "Story points" are evil, simply because they can't be measured. The qualitative nature of story points means that planning assumptions are NEVER checked against reality. It's a strong incentive for management to avoid any responsibility; stay "agile", use "story points", and never ever assess your actual performance or velocity.

The worst managers I've seen straight up forbid talking about the story points. Anyone asking for more information or discussion gets shot down; story points are "just an estimate" after all. Then when that lack of information becomes a problem, employees are forbidden from referring back to previous estimates. Even calibrating story points in a relative sense is expressly forbidden. There is no feedback mechansim to correct poor planning. And that's by design; the system protects and shields mediocre managers from accountability, period.

No, this is quite backwards. I find myself reading code more, to gain context and build the theory for the feature I'm working on. The time it takes to write the code is now trivial compared to designing a good idea. Reading/thinking about code absolutely dominates now that writing code is cheap.

Muddled prompting by humans gets you the Homer Simpson car you wished for

Well put! Now that we have a magic tool that can generate tokens on demand, the quality of the underlying idea gains enormous importance relative to the code. Tokens are cheap. Good ideas are not.

I would like to hope that some people take advantage of this newfound agentic power to create better theories. But there's a sizable population that seems intent on generating more and more code, regardless of quality.

Super strange framing.

The plant still need solar energy. They still need electricity within the tissues of the organism to survive (ATP and krebs cycle). Humans have always burned organic matter for light.

Not trying to be a pedant but "Light without electricity" falls down when examined from any angle. It's not a serious claim.

Partly true, but don't blame GeoJSON. Blame the data model.

GeoJSON is firmly rooted in the "simple features" model of spatial data. Sometimes called the "vector data model", this is ubiquitous in GIS. Each geographic entity (aka "Feature") has a single geometry and many non-geometry attributes. Each feature is independent.

The vector data model (for better or worse) is the basis of many systems because it fits the tabular/relational style so closely. What is a feature but a row in a table plus a special column describing its geometry? Topological relationships are ignored by design.

TopoJSON, ESRI coverages, the internal OpenStreetMap data model, and a few others are notable exceptions. They explicitly handle spatial relationships, at the cost of making the model unintelligible to row-based processing.

14x is insane, especially since the quality and quantity of IRL software has barely budged.

One could hope that we'd use these newfound agentic coding powers to actually realize value, improve quality, etc. Instead I see enshittification and stagnation. What are we even doing with all these tokens?

Good advice generally. But please, not at the gym. All gyms have a different vibe but mine is almost strictly no talking. We go there to workout, not to chat. Everyone locked in, headphones on, no nonsense. I've been going for years and I can count on one hand the number of conversations I've witnessed.

But the flipside is, I see the same gym crowd at the coffee shop next door and we always have a good chat there. Context matters.

Valuing quality over velocity is not selfish

Fully agree. I never mentioned velocity or advocated for lower quality. In fact, this statement very well sums up my point: we should care about the thing we're producing, not our personal experience of coding it.

The "typing ascii characters" angle is a bit hyperbolic, I admit. But my intention is to get people to think about their software, not their personal experience of it.

BTW, there's nothing preventing you from using AI agents and staying in the flow state. If you want that, the universe is not stopping you. In fact, not dealing with the minutia of source code may well free us up and allow even greater flow experiences.

I tend to agree with his point.

But I found myself laughing at the style; just ranting about software like a cartoon villain in his bathrobe. No fucks given.

Thanks for this perspective. Do we really miss typing ascii characters into an editor? That seems to me the least consequential and least interesting part of building software systems. Always has been.

Dare I say those stuck on nostalgia for pressing keys are demonstrating that they cared more about their own personal experience than about the outcome of their work? Now that coding is automated, we have to elevate our ambitions.

Ironically, Phish's music emerges from egoless expression (to paraphrase keyboardist Page McConnell). Giving up your own personal stake in the process is literally what brings something as beautiful as Phish's music into existence. We need to do the same with our software; give up the notion that "our" code is meaningful.

Running models locally using LM Studio, you can use a shell function like

    claude-local () {
     MODEL=$(curl --silent localhost:1234/api/v1/models | jq 'first(.models[].loaded_instances[].id)') 
     ANTHROPIC_BASE_URL=http://localhost:1234 ANTHROPIC_AUTH_TOKEN='' claude --model $MODEL
    }

Seconded. I learned a ton from it, as well as his previous book The Ends of the World on mass extinctions. His writing style is wonderful, he turns an academic subject into a page-turner.

The Story of CO2 taught me something I had never considered. It wasn't exactly that photosynthetic life started pumping out O2 and chilled the planet. Snowball earth happened way later. It was photosynthetic life that got buried in sediment and locked it away from aerobic respiration. The amount of carbon stored in the earth's crust is insane. Fossil fuels are just a minuscule fraction of that.

This has some implications for our current climate: If we want to use biology to sequester carbon (growing trees, algae, etc), it's only a temporary sink unless we lock it away for eons. Once it's eaten/burned, the CO2 is right back in the atmosphere. In short, we gotta physically put it back into the earth's crust if we want to draw down carbon.