HNHacker News
TopNewBestAskShowJobs

perrygeo

3,193 karma · joined April 28, 2014

submissionscomments
perrygeo··on The Mysterious Syndrome Destroying Endurance Athletes
Nope. Lactate and lactic acids are by-products of fatigue, not causal agents. Far from being toxic, elite athletes are even intentionally ingesting lactate as fuel these days.
perrygeo··on Incident with Github.com
Github's, COO Kyle Daigle, claimed that the number of commits has increased 14x since last year. So shear scale has something to do with it. I don't know of too much software already operating a planetary scale that could easily absorb a 14x bump in a year.
perrygeo··on Our Reality Is Shifting and It's Just the Start
Mostly I'm seeing breakthroughs in the ways we use LLMs. Agentic harnesses, MCPs, etc - its the wild west still but we've come a long way from a basic chatbot. Gains are now coming from tools that make better use of the LLMs existing capabilities, and put guardrails on their worst tendencies.

I personally feel like we've plateaued in raw model intelligence (even regressed, I find sonnet 4.6 to perform better than Opus 5) but we've gifted them new skills that allow them to run for longer and explore the search space more thoroughly, making them more effective at the same level of "intelligence".

Take two smart people and a problem to solve. Give one person the tools, the other person nothing. The one with the best tools wins. At some point its more about abilities than raw intelligence. Watching a "frontier" model fumble with basic syntax is still common, but not if you give it treesitter.

perrygeo··on NP-overrated
When people are initially confronted with TSP and similar optimization problems, the reaction is usually awe at the shear number of potential combinations. "A 48 city tour of the continental US has 10^61 possibilities!!".

Well, not really. Massive chunks of the search space can be eliminated through clever (but non-optimal) algorithms. The rest of the search space can usually be explored through heuristics. In practice, we can solve gigantic TSP problems "well enough" and "fast enough".

It reminds me of the Midwit meme. Both the low IQ and high IQ folks say "Heuristics are good enough". Only the mid IQ guy cares about NP-hard.

That's not to take away from the research into theoretical limits of computation. Just noting that's a completely different question from the practical concerns of actually solving those problems IRL

perrygeo··on Why Target Common Lisp for Code Generation?
The counterweight is that Lisp syntax is tiny and regular, the language is concise, and it encourages the composition of small pure functions. There's very little context to get confused about; the code can be reasoned about locally. Then verified in the live REPL, which gives the agent much tighter feedback loops.

So in practice, Lisp uses tokens very efficiently and tends to get things right, fast. The syntax just is not an issue one has to worry about (and your harness should be checking your model on this btw)

I suspect that Clojure which adds a little more syntax - vectors [] and maps {} - is slightly better than other Lisp dialects since the data structure literals provide a stronger signal than parens.

perrygeo··on 2,085 Tests, and None of Them Opens the Front Door
Anecdote time: I worked on a project with a senior dev who was really into unit testing and only unit testing. The software had to be fully tested as tiny little lego bricks with mocks galore. Because "if all the pieces were correct, the combination had to be correct" - his words. (I guess he'd never considered emergent behavior)

I'll never forget the cycle, after staunch objections to integration/e2e/manual tests and zealously insisting on full unit test coverage, the senior dev deployed changes ... boom. The process failed hard minutes after start because there was a bug in main. Full production outage. A bug so obvious we should have caught it by eye. Clearly, in the weeks developing the feature, he had never bothered to even run the application. Each part was fully polished in isolation but utterly failed when integrated into an application.

The way I've come to think of it: Unit tests are for libraries, a closed world where you can neatly define all inputs and outputs. Integration/E2E tests are for applications, an open world which require systems thinking, simulations not proofs. Know which world you're working in!

perrygeo··on AI is removing the middle class of software engineering?
Rather than good or bad engineers, I'd focus on the quality of the idea. Every software engineer has good and bad ideas. Historically, bad ideas were kept in check by the effort it would take to code the implementation. With that barrier nearly gone, there's no friction - a bad idea in the morning gets deployed to prod in the afternoon. No thinking required!

So yes, garbage in -> garbage out, but framed in a way that makes it clear what is garbage. The ideas, not the engineer themselves :-)

This suggests we need to be doing more designing and planning; introducing that friction intentionally to make sure bad ideas get culled, viable ideas get refined. Critical thinking becomes the bottleneck; the quality of the idea becomes the deciding factor in success.

perrygeo··on LLMs reward expertise
LLMs are language models. So much of this can be reduced to a simple heuristic: If you can't think clearly, LLMs will not help you. If you don't know what you're asking for or how to express it precisely, it really should not be a surprise that the output is garbage. Failures of LLMs are more often failures of our own brain to consider the problem clearly. That's harder to admit than just blaming the AI.

It's important to note though, from the perspective of the LLM's objective function, that this is not a failure at all! LLMs are designed to match patterns. You give a jumbled mess of incoherent ideas, it will faithfully reproduce a token stream of incoherent ideas. It's only when the model output is subjected to the real world that it fails.

perrygeo··on Wind and solar overtake fossil fuels in Germany for the first time
True. Less than 40% of the world's fossil fuel energy goes to generating electricity. The rest goes to transportation, industrial processes, material feedstock, and direct heating - which are not accounted for in this metric!

"Renewable alternatives cover half of 40% of fossil fuel demand" would be more accurate. Still a win, but it's not "overtaking fossil fuels" by any definition.

perrygeo··on So you want to use plants to reduce CO₂
Good point. I didn't mean to imply that carbon was the only value. The "services" provided by these ecosystems are incalculable.

The myopic focus on carbon is an easy trap to fall into; it's the quantifiable part of the system. In this world, that which gets measured gets managed.

I would hope we'd conserve these ecosystems because of a deeply-held spiritual connection to life. Or at least a realization of the value they provide to humanity by simply existing. But carbon remains the objective metric by which conservation interacts with the financial market.

perrygeo··on I fucking love this; you don't
> Just keep building... If you don't want to use AI, don't. If you want to, great.

I love this sentiment. Quality should transcend the tooling used to create it.

The by-hand artisan coder vs the LLM-assisted artisan coder - not that big a difference in practice. The important part is actually giving AF about your work.

Ultimately it's about skin in the game. Do you care about what you're building or not? Show me, don't tell me.

perrygeo··on So you want to use plants to reduce CO₂
Forests and grasslands are critical for the climate because of the carbon they lock away. So long as the plants stay alive, they store carbon in the cellulose structures of roots and stems.

Conserving these ecosystems is less about their daily respiration and more about keeping existing carbon out of the atmosphere. It's a defensive strategy. Not "carbon capture" per se but more "avoid loosing the carbon we've already captured". We're interested in the sink, not the flow.

Of course when the plant dies and decomposes or burns up, it all goes back. Long term, it's almost exactly net zero. The ONLY way to use plants for long term carbon sequestration is to make sure that plant matter gets treated or literally buried so that aerobic microbes and fungi can't break it down.

perrygeo··on `bun init` automatically creates a Claude.md file by default
At least this was planned ahead of time, intentionally, by the human maintainer. https://x.com/jarredsumner/status/1934822081591681353

The fear with these coding agents is that, unsupervised, one will merge a sneaky pro-AI changeset like this without a human ever signing off on it.

But this looks like run-of-the-mill bad human decision making, no singularity yet! This is fine.

perrygeo··on Maybe Intelligence Ain't All That
I largely agree with the premise. I was just honing in on the coding claim because the statement is so obviously wrong that it distracts from an otherwise solid post.

We should put software squarely in the "complex system" category, ie "The only way to find out is to build the software system and try to break it". Writing code may be cheap but running the software IRL is still constrained by both computational bounds and the laws of physics. Not to mention social and market dynamics.

If anything this strengthens the overall argument. "The limit is contact with reality".

perrygeo··on Maybe Intelligence Ain't All That
> Coding, math and most administrative work are similarly benign. What makes them easy is that they have relatively smooth solution spaces and are tractably verifiable.

Code most definitely does NOT have an easily verifiable search space. P!=NP? Halting problem?

It's easy to brush this off and think those are obscure computer science trivia, but these things come up all the time. If we could prove software correct, we wouldn't need to test it.

> Most of what matters doesn't behave like that. The universe is mostly the emergent behavior of complex systems..... In systems like these, no amount of reasoning delivers the answer, because there's no shortcut hiding in the gaps. You have to run the thing

Software is clearly a complex system, with plentiful examples of emergent behavior. To understand the behavior, you HAVE to run the thing, in prod, on the open network, with real users, under real conditions, with real data. You can simulate these in testing but you can never fully reason about them beforehand - they are computationally irreducible.

perrygeo··on Quality non-fiction books are the antithesis of AI slop
Billions of conversations. Zero lived experience.

The difference is huge. An LLM conversation can contain within it a batshit crazy model of the world and still produce a plausible stream of tokens. No skin in the game. If you don't test your ideas against reality, you don't have a model of reality; you have a model of written human thought. It's critical to understand the difference.

perrygeo··on Does creatine make you smarter?
My point is that creatine has had significant attention on it for decades. OP perceives an increase in attention but that's just a reflection of their knowledge (Baader–Meinhof effect)
perrygeo··on Does creatine make you smarter?
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!
perrygeo··on Ask HN: Alternative Careers in the Age of AI
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.

perrygeo··on The LLM Critics Are Right. I Use LLMs Anyway
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.
perrygeo··on No LLM Code in Dependencies
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?
perrygeo··on Open Source Low Tech
Love it. Another great site along these lines: https://solar.lowtechmagazine.com/
perrygeo··on Map Clustering Is Not My Favorite
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.

perrygeo··on What job interviews taught me about Kubernetes
> 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!

perrygeo··on Removing my Nix flakes vs. guix post
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.
perrygeo··on Ask HN: What would you do with a trillion dollars
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.

perrygeo··on "Don't You Just Upload It to ChatGPT?"
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.

perrygeo··on Learn SQL Once, Use It for 30 Years
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.
perrygeo··on My thoughts after using Clojure for about a month
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.

perrygeo··on Friction Was the Point
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.

← PreviousPage 4 of 34Next →