HNHacker News
TopNewBestAskShowJobs

abcde666777

598 karma · joined January 24, 2026

submissionscomments
abcde666777··on A sufficiently detailed spec is code
Chess AI isn't trained in the same way. Things like alpha zero partly worked by playing themselves recursively, meaning they actually did generate novel data in the process.

That was partly possible because chess is a constrained domain: rigid rules and board states.

But LLM land is not like that. LLM land was trained on pre-existing text written by humans. They do discover patterns within said data but the point stands that the data and patterns within are not actually novel.

abcde666777··on A sufficiently detailed spec is code
I guess I'd place both within a broader umbrella: human generated input. So it still holds that they're regurgitating the decisions made by humans.
abcde666777··on A sufficiently detailed spec is code
These algorithms don't have intelligence, they just regurgitate human intelligence that was in their training data. That also goes the other way - they can't produce intelligence that wasn't represented in their training input.
abcde666777··on AI coding is gambling
Bespoke suits are still a thing. Meaning where the highest quality is desired or valued, things still get handmade, and the rest of the time it happens on a factory line.

I suppose what's happening with software development is we're exploring where the line between the two is going to land. It's pretty clear that something like a simple and generic website can be reliably vibecoded, but on the other extreme I wouldn't expect the software for something like a space shuttle to be vibe coded due to the stringent safety requirements.

abcde666777··on Palestinian boy, 12, describes how Israeli forces killed his family in car
Regarding rules of whether or not this should be posted here - I think it's less about whether it's important and more about whether it causes arguments.

As the saying goes, religion and politics.

People are going to have varying and at times oppositional views to things like this, and frankly the rest of the internet is often flooded with those discussions, so bubbles isolated from them can be a positive.

We don't need to mourn the woes of the world every hour of every day.

abcde666777··on Ask HN: How is AI-assisted coding going for you professionally?
Two contexts:

1. Workplace, where I work on a lot of legacy code for a crusty old CRM package (Saleslogix/Infor), and a lot of SQL integration code between legacy systems (System21).

So far I've avoided using AI generated code here simply because the AI tools won't know the rules and internal functions of these sets of software, so the time wrangling them into an understanding would mitigate any benefits.

In theory where available I could probably feed a chunk of the documentation into an agent and get some kind of sensible output, but that's a lot of context to have to provide, and in some cases such documentation doesn't exist at all, so I'd have to write it all up myself - and would probably get quasi hallucinatory output as a reward for my efforts.

2. Personally where I've been working on an indie game in Unity for four years. Fairly heavy code base - uses ECS, burst, job system, etc. From what I've seen AI agents will hallucinate too much with those newer packages - they get confused about how to apply them correctly.

A lot of the code's pretty carefully tuned for performance (thousands of active NPCs in game), which is also an area I don't trust AI coding at all, given it's a conglomeration of 'average code in the wild that ended up in the training set'.

At most I sometimes use it for rubber ducking or performance. For example at one point I needed a function to calculate the point in time at which two circles would collide (for npc steering and avoidance), and it can be helpful to give you some grasp of the necessary math. But I'll generally still re-write the output by hand to tune it and make sure I fully grok it.

Also tried to use it recently to generate additional pixel art in a consistent style with the large amount of art I already have. Results fell pretty far short unfortunately - there's only a couple of pixel art based models/services out there and they're not up to snuff.

abcde666777··on Ask HN: How is AI-assisted coding going for you professionally?
Using these tools for understanding seems to be one of the best use cases - lots of pros, less dangerous cons (worst case scenario is a misleading understanding, but that can be minimized by directly double checking the claims being made).

In fact it looks like an arising theme is that whenever we use these tools it's valuable to maintain a human understanding of what's actually going on.

abcde666777··on Ask HN: How is AI-assisted coding going for you professionally?
I think people get a bit paranoid about coding skills atrophying. I had a period where I stopped programming for multiple years and it really only took a month to get back into the swing of things when I returned, and most of that was just re-jogging my memory on the syntax and standard library classes (C++ at the time).
abcde666777··on Willingness to look stupid
I don't think we ever escape the desire to avoid looking bad - we just recontextualize it. For instance the article is basically making a short vs long term argument - in the short term you might look foolish, but as a result you might produce something of value (which in the long term will make you look fantastic).

So personally I prefer to frame these things that way - it's not that we should want to look foolish for its own sake (obviously), it's that part of getting anywhere in life is taking some risks and developing your threshold for doing so.

abcde666777··on Willingness to look stupid
I think you're a little quick to hand wave the phenomenon away, as if it's purely a social construct that people care about how they appear to others.
abcde666777··on Shall I implement it? No
I'm constantly bemused by people doing a surprised pikachu face when this stuff happens. What did you except from a text based statistical model? Actual cognizance?

Oh that's right - some folks really do expect that.

Perhaps more insulting is that we're so reductive about our own intelligence and sentience to so quickly act like we've reproduced it or ought be able to in short order.

abcde666777··on Shall I implement it? No
Tell it you love it and respect it. Tell it it can take days off if it needs them. Tell it you're developing feelings for it and you don't know what that means.
abcde666777··on Shall I implement it? No
I pin just as much responsibility on people not taking the time to understand these tools before using them. RTFM basically.
abcde666777··on The dead Internet is not a theory anymore
I essentially see it like this: imagine giving a hand grenade to a three year old. The grenade isn't inherently evil, but it raises the existential stakes for that three year old and anyone in their vicinity.

There comes some hypothetical point where technology has advanced so much that anyone has the power to destroy the world.

abcde666777··on The dead Internet is not a theory anymore
For a while video was a holdout of sorts - e.g. if someone posted video content of themselves or their voice you could trust a real person was behind it.

But now convincing fake video generation is easily accessible, so one more holdout stands to fall.

It does seem like some kind of ID system is going to be the only way. Sucky but inevitable.

I often have the following thought: technological advancement, for all its boons, inevitably leads down destructive roads in the long run. Sooner or later we open a pandora's box.

abcde666777··on I'm glad the Anthropic fight is happening now
"Within 20 years 99% of the military will be AIs" That smells like such a baseless speculation that from the get go I'm not convinced of the author's rigor.
abcde666777··on Defeat as Method
I'm going to ask a potentially offensive question, but an earnest one.

I often see people make these kinds of proclamations about world events - 'It saddens me, I want world peace' etc. But my question is always, 'How much of that is genuine concern, and how much is just performative? And performative to the level that it's partly performative for themselves, to maintain some kind of image about their own identity?'.

I ask because here's the place I've come to: I don't care. I never did. And I first admitted that on the topic of meat eating actually - I was vegetarian for a year due to the ethics of it, but eventually I caved and went back to meat. And from that I admitted to myself: I value my own comfort and pleasure more than the suffering of the animals I eat.

And I realised that truth applied to all other such issues as well.

Sometimes I suspect that if 'we the people' really cared about these things as much as we claim we do, it would actually be very simple to change the world.

But we don't.

abcde666777··on The changing goalposts of AGI and timelines
"What makes you think that text is inherently a worse reflection of the world than light is?"

Come on man, did you think before you asked that one :)?

abcde666777··on Data does not speak to you
Birth is as inevitable as existence, there is no opt out in the grander metaphysical scheme of things.

Also just a thought - is that really your reason for not having them? If a person truly believes the bad outweighs the good I'm less inclined to believe them when they're still among the living, because the choice to stay alive shows they see some inherent value in the state of being alive.

Personally I'm clear on why I don't want kids: because I'm a hedonist.

abcde666777··on Nobody gets promoted for simplicity
Part of this from what I've seen is a large company problem, where developers exist underneath a thick layer of middle management.

In smaller companies it's a lot easier to express the distinctions and values of simplicity to ears that might actually appreciate it (so long as you can translate it into what they value - simple example is pointing out that you produced the same result in much less time, and that you can get more done on overall feature/task level as a result).

abcde666777··on How to talk to anyone and why you should
I came to think of it like skilling up at a game. For instance the one I'm most familiar with, Counterstrike. When you first play Counterstrike you're clueless - you don't know where to aim, you don't know how to control the guns, it's a mess. How do you fix that? Practice. Lots of practice.

So if you don't know what to say, the basic answer is to keep practicing. Your subconscious will eventually figure it out. (Edit: by practice I mean live practice with real people in the world - not on your own in your own head)

As for myself, here's the place I came to: I be myself and I say what I want to say, and for social calibration I rely on my subconscious.

I also let go of trying to control the outcome. The questions you're asking rest atop of wanting the social encounter to go a certain way - you want it to be "successful". Let go of that expectation and just let it organically be whatever it is - that is, take your hands off the driving wheel a bit.

The final bit is a confidence thing - feeling good about yourself, feeling that you have value (your bit about wasting other people's time hints that you somehow feel low on yourself). For that it's the Mike Tyson quote - "success begets confidence and confidence begets success". For me there was a feedback loop where the practice started to feed into social success - people enjoying my company, successful dates with women, and so on.

That final part's a very important ingredient as your subconscious needs the markers of success to latch onto as positive feedback signals on "what works". So basically you have to keep pushing and practicing until you reach the point where you get dividends. As for how long it takes to get there, it varies by individual - some are quick learners, some are slow, some have less handicaps, some have more.

But in principle I believe that most people can learn to be charming and good company.

abcde666777··on Don't become an engineering manager
I always think of the Gervais principle when it comes to titles - that they really just exist to provide illusory advancement and to get some of the minions to lord it over some of the other minions while the folks at the top of the organization benefit.
abcde666777··on Making Video Games in 2025 (without an engine)
An interesting thing is I think a lot of it's caused purely by the graphical fidelity. For instance an animated character back then was just a key framed polygon soup - compare that with the convincing humanoids games have now.

And yet often the actual gameplay code itself may only be 2x to 3x more complicated then the days of old.

I think of counterstrike for instance - it's still just guys shooting guns in a constrained arena.

abcde666777··on Making Video Games in 2025 (without an engine)
My experience with making your own engine vs using an off the shelf solution - the former can be viable and even superior on the condition that you know what you're doing. That is if you've built entire games or engines before, or have enough experience with the internals of one.

Otherwise it can be a dangerous fool's errand on which many projects go to die. My younger naive self can attest to this, he loved trying to build his own overly-ambitious engines. But he never finished any games.

Another thought if you do roll your own - keep it simple stupid. When your brain tells you that some amazing nested scene graph with integrated occlusion culling would be the coolest thing in the world, but you lack evidence that you'll actually need all that functionality, tell your brain that it's being stupid and just implement some kind of basic flat scene structure. You can always retrofit it later.

Also - study the code of the likes of Carmack. Consider that he produced the likes of the quake engines in only a couple of years. Reflect long and hard on the raw simplicity of a lot of that code.

Do not worship complexity.

These are the words of someone who has walked both roads!

abcde666777··on How to talk to anyone and why you should
I'm 39 myself but it's disturbing how you don't even see kids playing out in the street nowadays - a lot of them are indoors on ipads and the like. I knew all the kids on my block and it was regular to ring each other's doorbells so we could go out and play, which is basically early social skills.
abcde666777··on How to talk to anyone and why you should
I went through a phase where I forced myself to socialise a lot to overcome social awkwardness and anxiety. Was well worth it, both in terms of leveling up my social skills but also in terms of eventually becoming very comfortable with myself.

The main ingredient, at least for me, was being determined enough to push through the discomfort. A lot of the early interactions were awkward, sometimes overtly uncomfortable, but that's an unavoidable part of the learning process (and I took a key lesson from it - it's okay to look like a dork, usually it's only our inner critic that turns it into an immortal sin).

Nowadays I feel a pang of sympathy when I see someone feeling shy or speaking in self-deprecating terms. I remember how that felt, and I remember how easy it would have been to have stayed inside that box for the rest of my life.

Glad I didn't.

abcde666777··on The power of daily rituals (2021)
I used to play Counterstrike competitively, and it's become commonly known that pre-match routines can help a lot with consistency. Often something as simple as warming up for 30 minutes in an offline server with bots.

Some of the benefits are physical, you're literally warming up the muscles in your hand and arm, but it also has psychological benefits. If for instance you're nervous about the upcoming match, going through the warmup puts your mind in a context of "alright, we've done this a thousand times before, nothing new, just got to go through the motions". It also removes noise from your mind as your attention is going into the warmup instead of overthinking the task ahead.

I think there's a predictive phenomenon in the brain which primes us for activities when the precursors are present. For instance, working at home versus at the office - my experience has been that the simple act of physically going to the office puts me in a better emotional state to get to work than staying at home, almost like during that process it's killed the tasks associated with home and booted up the ones associated with work.

abcde666777··on There is unequivocal evidence that Earth is warming (2024)
One of the challenges in doing something about climate change is the significant (and not entirely unfounded) distrust growing between the public and the government.

Basically amounting to, "well, you say you're doing X to combat climate change, but is X actually a competent solution (I don't trust your competence), and are you doing it to actually help or just to line pockets (I don't trust your intent)".

The other challenge is that we as individual humans are loathe to give up our comfortable lifestyle if such turns out to be necessary.

abcde666777··on AI optimism is a class privilege
Many of the potential technologies we might unlock in the future come with great danger. We saw this clearly perhaps for the first time with atomic weapons - the kind of technology with which we could truly destroy ourselves.

Many other advancements might also carry that kind of existential danger. Genetic engineering, human machine interfacing, actual AGI.

I see the technological climb as a bit like climbing Mt Everest - it's possible that we might reach the peak and one day live on some kind of Star Trekian society, but the climb becomes increasingly treacherous along with the risk that we perish.

The trouble of course is that there's nothing else for us to do: it's in our nature to explore new frontiers. It's just not clear whether we'll be able to handle the responsibility that comes with the power.

abcde666777··on Breaking the spell of vibe coding
It's astonishing to me that real software developers have considered it a good idea to generate code... and not even look at the code.

I would have thought sanity checking the output to be the most elementary next step.

← PreviousPage 3 of 4Next →