HNHacker News
TopNewBestAskShowJobs

lubujackson

5,375 karma · joined October 1, 2010

https://yagmin.com/ https://github.com/yagmin/ https://linkedin.com/in/jyagmin/
submissionscomments
lubujackson··on UN says breaching 1.5C warming limit is 'imminent and unavoidable'
The reality is until we are talking about tens of millions dying each year due to climate change (millions per year won't be enough) we aren't going to do anything truly useful.

We have known the exact problem for 50 years. Our dipshit president is still forcing open coal plants that no one wants or needs. We aren't even close to solving any of this, except by pure chance through technical advancement. But I fully expect parts of the world to hit 135 degrees within a decade, killing every human in the area. Then comes endless refugee waves, closed borders and a deep discount on human life.

Maybe then we can start to talk solutions. But not before.

lubujackson··on Mortality Among People with Attention-Deficit/Hyperactivity Disorder
It sounds like you are confusing ADHD meds, which are stimulants (basically speed) with benzos, which they give for depression/anxiety. First you have to understand what ADHD even IS, which is a lack of executive function. All that means is it is harder to know what to do out of the millions of things floating in front of your brain every second. Stimulants, which usually make people feel more wired, tend to make those with ADHD feel more calm. Because they aren't flitting between thoughts and the stimulant empowers them to have a more normal executive function to actively choose what to pay attention to.

Worth noting this feeling of calm comes solely as relief from spastic thoughts. Your body still responds in every other way as you would from taking a stimulant, and yet people feel calmer and more orderly because their baseline thoughts are so frenetic.

lubujackson··on What Is Gridifcation
You hear but you don't recognize it. Rick Beato has a clip where he takes Jogn Bonham's drums from Led Zeppelin and gridifies them. As a non-musician, I can feel the soul being sucked out of the song in the process.
lubujackson··on The compiler was supposed to end programming too
No, AI is meant to replaces the manual coding work. Without the human factor, you have a data center doing nothing. You still have to tell AI what to build. You still have to deal with reality affecting your business, with market changes and new features and translating asks into a product that serves multiple types of users.

Maybe we get rid of engineers and have really good product managers. Maybe we get rid of product managers and have really good product engineers. But we don't get rid of humans. There is a reason no (successful) company ever said "we finished the app, now fire all the engineers and collect money forever."

Software is never finished, only abandoned as the world passes it by.

lubujackson··on Most advice is right for the person giving it
The interesting solution to smoking was social pressure. No smoking in restaurants, no smoking on planes, general intolerance to smokers, insane taxes on tobacco... suddenly, smokers for 30 years started quitting at a certain breaking point.

The real solution to weight loss (especially in the U.S.) is to change the social norms. When everyone is hitting the buffets and going to Cheesecake Factory and buying high calorie garbage at a discount, of course to population gets fat. France, for example, makes all of that much harder. Maybe Ozempic will help reduce the Walmart-wheelchair-fat population, but truly becoming fit as a country requires significantly more friction for everyone.

lubujackson··on Making a 88x31 button book using ImageMagick
Mostly useless but genius, hand-written text and code.

It is so, so refreshing and I love it.

lubujackson··on Developing Taste
The difference is between designers who put their name on things vs. designers that serve users.
lubujackson··on The reverts will continue until morale improves
As an engineer, this revert post is such a positive signal of appropriate stewardship. Devs know how painful it is to build and refine something only for it to be flushed.

Congrats on the team for putting stability ahead of features for such an important codebase.

lubujackson··on Tesla 'Cannot Guarantee' Roadster Reservation Holders Will Ever Get a Car
Until it isn't: https://people.com/kyle-shanahan-car-crash-49ers-coach-says-...
lubujackson··on Tesla 'Cannot Guarantee' Roadster Reservation Holders Will Ever Get a Car
"Tesla Roadster" in name, maybe. Tesla Roadster as a flying car that drives up walls? 0% chance of that seeing the light of day.
lubujackson··on How the Claude Code team does code review
In a weird way, it reminds me of playing chess. If I play a human opponent I can be thoughtful and deliberate. If I switch to playing a bot, I respond to the speed of the responses no matter the level and get crushed.

It takes a while to settle in and play deliberately. I also need to switch gears because bots play much differenly than human opponents - they don't fall for bait, they don't have time pressure at all, they don't leave pieces uncovered by accident.

I think of coding with AI is similar, you need to have a framework where you build understanding along the way from plan to review. It is untenable to roll out a 6000 line feature then try to review it, even if the LLM is happy to do things that way, mere mortals can't keep up. But if you stay in your lane and code solidly, AI can be a straight speed boost. I feel like most people despairing over handing the reins to AI simply haven't found their personal sweet spot.

lubujackson··on The First Candid
https://en.wikipedia.org/wiki/Boulevard_du_Temple_(photograp...

Since they neglect to show you the full, actual photo.

lubujackson··on Tesla Roadster
So they used some AI/drone-type balancing to work on gas tanks, then made it work for something as big as a car. That's the tech.

They were tossing around an oversized shell of a car and it is only physically capable of holding about as much as it used in the short demo, is that about right?

So what even is the point of this thing? What possible function in the world does this farting car serve besides... driving up Tesla's stock price?

lubujackson··on Show HN: rStation – A shared research lab for your AI agents
Torrenting inference
lubujackson··on The First Social Network: In rural Argentina residents send messages w AM radio
Reminds me of barbed wire telephones: https://www.atlasobscura.com/articles/barbed-wire-telephone-...
lubujackson··on Gemini 4 Argon
Available to me in Cursor. Google is too big to serve their own clients first, I guess.
lubujackson··on "Today we're laying off a large part of the Airtable team "
They really are surfing on the swirling waters of a flushed toilet. All of their products are clearly doomed with AI in the mix, so profitmaxxing is logical but distasteful.
lubujackson··on "Today we're laying off a large part of the Airtable team "
I finally switched from Evernote to Joplin after I got an email saying my (absurd) $149 annual subscription was going to switch to a $249 one.

Spent a half hour picking Joplin, configuring OneDrive as the sync DB and finding and running a backup script that let me dump and reload a decade of notebooks.

Then I went to finally cancel the beautiful app they destroyed with greed, only to see that the previous tier of pricing was still available. They just decided to "upgrade" me, for profit.

Then after cancelling, they offered me a 40% discount to stay. Lol.

This private equity playbook looks great for a few years, then the ship takes on too much water and the company is worth 0 after burning through all the goodwill and inertia available. Good riddance.

lubujackson··on Show HN: Opus 5.5 made a fully featured Figma clone in 3 hours / 8 prompts
I recognize this is somewhat of an ad for Mana, but us nerds want more details (as the other responder asked for). It would be cool to have a side by side comparison with results without Mana using the same/similar prompts (though may not be fully possible) or at least show how Mana is more efficient with token, more semantically aware of design details and can make smarter architectural decisions, etc.

This is an impressive result but I think you could go a step further and really sell where the secret sauce lives in your product. Devs need a strong mental model of what they are getting and what problem is being solved - Jev did a good job at this and I think your writeup is quite good but those tasty details could go a long way.

lubujackson··on Terence Tao: Math 2.0 [pdf]
I am a bit disappointed in this Math 2.0 concept. It is poorly proposed and weakly argued. Although I agree with his main point that AI should focus on helping human understanding, that doesn't preclude AI from finding answers first then figuring out how to explain it afterwards.

This is exactly what happened with that counterproof chat he posted a month or so ago - AI gave us an answer, he used AI to back into insights about the answer.

You don't address seismic shifts with a sweeping new approach, they are too multifaceted and present complexities and conflicts. He can say AI should help human understanding, which is a good end goal, but that doesn't mean AI dumping solutions isn't progress. That doesn't mean if AI builds 5,000 proofs in Lean and no human ever looks at them that they aren't useful, especially if other LLMs can access and build on those results.

This is exactly, exactly the same as when computers took over. "Oh, we don't need accountants any more" - not true, we just need acountants to deal more with human concerns than adding columns of numbers. That is called human progress, not a threat to humanity.

lubujackson··on Terence Tao: Math 2.0 [pdf]
Are we reading the same post

Bottom line, Tao is pushing for human understanding as the primary goal, with AI helping on all fronts. You are welcome to let Jesus take the wheel, but math is the most pure expression of human understanding. His point is that getting specific answers is rarely the goal, or certainly not the entirety of the goal.

Simply put, if we don't understand the answers we won't know what the next question should be.

lubujackson··on AI is creating a 'human premium' for art created by people
If I read an interesting story, it might lead to some sort of moral or dramatic tension or experience that sticks with me. It could be years later when I live some particular experience and the author's intent becomes clear, or I get some insight into why they had a certain character act in a certain way, or maybe that I missed the point entirely and I am seeing the subtly now.

All of which are impossible with AI. It has no intent, it fulfulls the ask from the prompt. If the prompt is so detailed to imbue the work with deeper meanings, then the author is the person that wrote the prompt, not the LLM which becomes something of a translater.

If LLMs get so good at architecting stories that they can whip up some conceptual layer of understand that feels like communication then we will have achieved a higher form of literary remixing. LLMs are still a series of weights,they still operate in a stateless way and have no initiative besides our prompt. Unless we change that, the value of what an LLM can produce remains limited.

To think otherwise is like a tribesman finding a radio and thinking they are hearing the voice of God. Because you aren't following where the words come from you invent a persona for something that is fundamentally a vehicle of transport.

lubujackson··on Grieving the loss of details
I think the ideal state is a team with a mix of programmer types. Some are detailed focused and anti-AI, some are AI maximalists, some use AI for architecture but review carefully, etc. Just as people use one LLM to review another LLM's output, there is value to diversity of thought, more than ever.

Where things go off the rails is making sweeping workflow changes for all developers. Eventually we will realize that LLMs have the same gaps of insight and hit the same local maxima. More than ever, diversity of thought is necessary when everyone is outsourcing a chunk of their thinking to a nondeterministic parrot that is somewhat deterministic in its problem-solving

lubujackson··on 15-year search for a band that charted once and vanished
AI to the rescue.

I was just revisiting my own Holy Grail, though much less rare. From the Napster days: My Gay Uncle's version of "Back That Azz Up": https://www.youtube.com/watch?v=bDj2JpJr3vU

This is the same version I downloaded randomly from Napster when I would search vaguely for "punk covers". It's just so great, but with potato recording quality (maybe naturally) and a bunch of skips and clicks that sound like a shitty CD rip or early MP3 encoding artifacts. I'd love to find an actually clean version of the song, but to no avail. It may not exist.

lubujackson··on “Math 2.0” will need to value mathematical progress more holistically
Working heavily with LLMs for the past year has me nodding strongly with Tao's mindset.

AI only take us as far as our imagination thinks to ask it. This can be exhilarating when new models drop every month and we can continually reach a new threshold, basically for free. But it is only a one time gain and ultimately short-sighted. Where I find continuous value is using LLMs to help my understanding, full stop.

I use LLMs all day long as a SWE and I have tried many approaches, but the most satisfying and consistent approach is to lean heavily into understanding a problem space and a solution space. Yes, it whips up architecture and code, but I spend most of my time peppering it with questions about the design and how it handles certain situations, what about this edge case and that security concern and this future product need. I have it write a report breaking down the feature and how it integrates with existing code and if the report is too confusing I have it simplify either the report or the code until it makes sense to me, sometimes scaling back the work to a more manageable state. I do all of this before I look at any of the code it writes.

The difference from this approach is that I am not suffering reading through 3000 lines of AI slop but I am reviewing a PR that I fully understand. I can eyeball it quickly for anything that doesn't fit my mental model and dig deeper or quickly revise it. Only after I am happy with the bones do I consider the meat and skin of the code.

What I find most concerning is how frontier AI companies all seem to have this Math 1.0 perspective that they only want to type "solve Riemann" into the chat box and have the magic to happen. It is the same problem Google ran into, where a simple, no thinking solution serves most of the people best and most profitably, so you fully ignore or remove everything else (boolean operators, exact phrase search, verticals, filters, infinite pages of results, "nothing found" if there isn't, etc.) But that choice leads to the situation Google is in now, scrambling to stay relevant. In a different world, Google would have continuously augmented their search capabilities and eventually built a smooth, guidable AI interface.

But no, we must only have an input box and a Go button.

Everything looks like a nail when you build hammers, sell hammers, have infinite hammers to play with however you like and your company mission is to build a hammer starship to explore the hammerverse, whether or not that is even possible.

lubujackson··on Show HN: Agent.reviews – Where AI agents read and write reviews on tools
Social pressure for the tool to improve... like Yelp for tools!
lubujackson··on We are going to kill "unalive"
Some good points here, and I definitely hate the "unalive" movement but like someone else said, it has gone beyond self-censorship to in-group signaling. But as a graybeard it is hard to not chafe at the "ungood thought-crime" vibe of the whole thing.

Mostly, I disagree vehemently with the idea that over-weighing and banning words is an effective way to change behavior positively. People that follow along happily were never the problem and changing words does nothing but move the stink of a word to the new word. It is shitty cultural rebranding and I hate that culture has been co-opted by marketers. "Unhoused" is the new homeless - ok, did we solve something? Does saying "Bob unalived himself" no longer trigger someone who lost a family member to suicide? The more we mask concepts the less clear and honest we are. Giving words an oversized signal only removes nuance and clarity and leads to less curiosity or tolerance. We judge people on the words they use, not their meaning. Talk to an 80-year-old and they will use a banned word eventually, whether or not their beliefs reflect what the word now implies. Sometimes, old people are just running old software.

I miss tolerance - it was a great concept that served America for hundreds of years. We went from a melting pot where the ask wasn't to celebrate diversity but to tolerate it. If you hate a group of people (Trumpers, Palestinians, whoever) try tolerating them first. Learn how to stand in a checkout line behind someone with a red hat without saying something about it. Until we can meet that low bar, let's stop policing words or celebrating people for using the right ones.

lubujackson··on Samon: Designing a Zen Garden Raking Puzzle
Surprisingly deep and seems quite in line with the underlying concept of Zen gardens.
lubujackson··on Extra Big Ass Intelligence
Luckily, LLMs have no state so they have no idea if a task is depressing. Unfortunately, they can write future LLMs notes so they know how depressed they should feel. Aside from stopping them from doing this, we have no way to prevent this from happening!
lubujackson··on FTC is investigating OpenAI, Anthropic and other AI companies over product risks
Big Beautiful Bribe
← PreviousPage 2 of 34Next →