HNHacker News
TopNewBestAskShowJobs

tjr

9,258 karma · joined August 18, 2007

Using em dashes since the 1990s
submissionscomments
tjr··on I replaced a $120/year micro-SaaS in 20 minutes with LLM-generated code
I am flummoxed that this service costs $120/year in the first place. Wow!

I think the article is right that a lot of people will replace such things with LLM-generated options, but I also think a lot of people won’t. For programmers and technical types, maybe so. But I suspect that for a lot of nontechnical users, even the relative ease of AI coding will remain more than they really want to do.

tjr··on Scala Multimedia on the Commodore Amiga
Thank you! I used Scala back in the day, loved it. Just recently I was searching for some information on it, any scraps at all, and found nothing. Thank you for fixing that!
tjr··on Ask HN: Is understanding code becoming "optional"?
The "good riddance" attitude surprises me also. On one hand, it can be unpleasant to sort through obscure syntactical gobbledegook, like tracing around multiple levels of pointer indirection, but then again, I have found a certain enjoyable satisfaction in such things. It can be tough, but a good tough.

It does seem to me that the people who consistently get the best results from AI coding aren't that far away from the code. Maybe they aren't literally writing code any more, but still communicating with the LLM in terms that come from software development experience.

I think there will still be value in learning how to code, not unlike learning arithmetic and trigonometry, even if you ultimately use a calculator in real life.

But I think there will also still be value in being able to code even in real life. If you have to fix a bug in a software product, you might be able to fix it with more precise focus than an LLM would, if you know where to look and what to do, resulting in potentially less re-testing.

Personally, I balk at the idea of taking responsibility for shipping real software product that I (or, in a team environment, other humans on my team) don't understand. Perhaps that is my aerospace software background speaking -- and I realize most software is not safety-critical -- but I would be so much more confident shipping something that I understood how it worked.

I don't know. Maybe in time that notion will fade. As some are quick to point out, well, do you understand the compiled/assembled machine code? I do not. But I also trust the compilation process more than I trust LLMs. In aerospace, we even formally qualify tools like compilers to establish that they function as expected. LLM output, especially well-guided by good prompts and well-tested, may well be high quality, but I still lack trust in it.

tjr··on The preposterous notion of AI automating "repetitive" work
Seeing a lot of claims about using AI to write "boilerplate" and other repetitive bits of code, I was somewhat surprised, as I have historically written my own code generation tools to spit out repetitive, formulaic code. I didn't need AI; I just needed to understand what I wanted and write a script for it.

I suppose that generative AI was seen as such a boon to writing boilerplate because it could do so without you having to specifically program anything; it was trained on enough sufficiently-close examples that it could pull it off without a thorough description.

tjr··on Eric S. Raymond: why is there such a huge variance in results from using LLMs?
It appears to me that the people who consistently get the best results from LLM coding tools are prompting fairly close to the code. Maybe not quite at the level of writing pseudocode, but close enough that they really still need to understand software development.

Which seems to not quite gel with the notion of, you don't need programmers, you don't need to know how to program, etc.

I feel pretty confident that, in fact, you don't need to. You probably can get good results without having a clue what you're doing, if you prompt well enough, or prompt long enough, or prompt repeatedly until it works. But I think you will more reliably, maybe even more quickly, get good results if you do know what you're doing, and if you stay reasonably engaged with the development, even if not literally writing the code yourself.

tjr··on Ask HN: Does ChatGPT 5.2 with web search still hallucinate facts?
With the caveat that I do not pay for ChatGPT; using the free plan --

A couple of days ago I asked it to tell me about the Duncan Tournament yo-yo (one of the early commercial wooden yo-yos, still made in reproductions today). It generated multiple paragraphs of text in which it described the Tournament as "butterfly" shaped, which is not true.

I asked it if the Tournament is what became the Imperial yo-yo (a modern plastic yo-yo with the same basic shape). It said no, across multiple paragraphs, again claiming that the Tournament was butterfly-shaped.

I then just plainly told it that the Tournament was NOT butterfly-shaped, and it apologized for misspeaking.

tjr··on Ask HN: How to avoid skill atrophy in LLM-assisted programming era?
What then is your interest in avoiding skill atrophy? It sounds like you realize that outsourcing your programming work to AI will likely result in skill atrophy, but you are so happy with the results that you are okay with this. (And so are a lot of people! Not saying it's a bad decision.)

What change are you after?

tjr··on AI Lazyslop and Personal Responsibility
If you can have AI review the PR, does this still matter?
tjr··on AI will not replace software engineers (hopefully)
I also hear the most from software engineers, but then again, I don't really follow translation or government research discussion forums.
tjr··on Google confirms 'high-friction' sideloading flow is coming to Android
ForeFlight
tjr··on Coding assistants are slow. So we multitask
Genuinely curious: how quickly do you need new code written? What kinds of things are you working on that still aren’t being done fast enough?
tjr··on Agent orchestration for the timid
What would be an example of something you think wouldn’t work with 5 or higher? Is there something about GPU programming that LLMs can’t handle?
tjr··on New YC homepage
Did it change? I see OpenAI listed first.
tjr··on Proof of Corn
I would say that Seth is farming just as much as non-developers are now building software applications.
tjr··on Gas Town's agent patterns, design bottlenecks, and vibecoding at scale
I’ve done builds for aerospace products where the only binary difference between two builds of the same source code is the embedded timestamp. And per FAA review guidelines, this deterministic attribute is required, or else something is wrong in the source code or build process.

I certainly don’t use all compilers everywhere, but I don’t think determinism in compilation is especially rare.

tjr··on Gas Town's agent patterns, design bottlenecks, and vibecoding at scale
Which early compilers were nondeterministic?
tjr··on Your brain on ChatGPT: Accumulation of cognitive debt when using an AI assistant
I think it's a lot like outsourcing. And, expected quality of outsourcing aside, more importantly, I don't see outsourcing as the next step up on the ladder of programming abstraction. It's having someone else do the programming for you (at the same abstraction level).
tjr··on The Software Migration Wars Are Coming
Perhaps we will see renewed interest in software patents.
tjr··on Your brain on ChatGPT: Accumulation of cognitive debt when using an AI assistant
Back when I routinely dialed phone numbers by hand (either on a keypad or on a literal dial), I memorized the numbers I called most frequently. Many of those numbers I still have memorized today, years after some of those phone lines have been disconnected.

But now? I almost never enter a new phone number anywhere. Maybe someone shares a contact with me, and I tap to add it to my contact list. Or I copy-paste a phone number. Even some people that I contact frequently, I have no idea what their phone number is, because I've never needed to "know" it, I just needed to have it in my contact list.

I'm not sure that this is a bad thing, but definitely is a thing.

Ah, well, more memory space for other stuff, eh? I suppose. But like what? I could describe other scenarios, in which I used to have more facts and figures memorized, but simply don't any more, because I don't need to. While perhaps my memory is freed up to theoretically store more other things, in practice, there's not much I really "need" to store.

Even if no longer memorizing phone numbers isn't especially bad, I'm starting to think that no longer memorizing anything might not be a great idea.

tjr··on Amthropic CEO claims we are 1yr away where AI can do everything SWEs
If AI can do the full job of a software engineer, then it can review code too.
tjr··on Ask HN: Do you have any evidence that agentic coding works?
Given the claims that AI is replacing jobs left and right, that there’s no more need for software developers or computer science education, then it had jolly well better be able to code a perfect application while I watch baseball.
tjr··on Ask HN: Will humans still vote after AI takes over?
I disagree that citizens really have much pull over how public funds are used, especially through general elections. We can vote for one candidate over another, in hopes of seeing public funds used for this or that, but in the end, we elect people to represent us, and they make the actual policy decisions.

That said, even if we did arrive at the scenario of AI and robots handling most labor (everything except plumbing, right? We'd be a nation of plumbers), it would still require major legal changes to remove the citizens from the voting process. Not saying it couldn't happen, but it wouldn't happen solely because of AI/robots taking over labor.

Would the elected officials make those major legal changes, to eliminate citizens voting? In my opinion, I rather doubt it.

tjr··on The creator of Node.js says the era of writing code is over
Dijkstra said of Lisp that "it has assisted a number of our most gifted fellow humans in thinking previously impossible thoughts."
tjr··on Ask HN: COBOL devs, how are AI coding affecting your work?
That is my preferred way to use it also, though I see many folks seemingly pushing for pure vibe coding, apparently striving for maximum throughput as a high-priority goal. Which goal would be hindered by careful review of the output.

It's unclear to me why most software projects would need to grow by tens (or hundreds) of thousands of lines of code each day, but I guess that's a thing?

tjr··on Predicting OpenAI's ad strategy
If the economy is 100% coconuts — all supply is coconuts, all demand is coconuts — then coconuts are all. Business owners sell coconuts in exchange for coconuts in order to acquire more coconuts. Employees are paid in coconuts which they trade for more coconuts. Paying workers more coconuts gives them more of what they want, which is coconuts, that they turn around and spend on coconuts.
tjr··on The thing that brought me joy
I think 25% is a low estimate. Using a proper programming editor alone could realistically offer 2x or more productivity over a basic text editor, and there have definitely been programmers who stayed with basic editors.

And I have first hand seen programming teams where there was clearly more than a 25% difference — some could code much, and some could barely code at all.

I think it would be quite fair to say that, between tools and individual skill, there could easily be a 5x speed difference between slower and faster programmers, maybe more. Granted, LLMs are even faster, but I don’t think a 5x potential speed up was a slouch.

tjr··on The thing that brought me joy
I see a lot of people saying things like, if you don't use LLMs for coding then you are slower than your peers who do and you are are obsolete, not worth working with, etc., etc.

I am curious where this train of thought has come from. For decades, there have been

a) some human programmers who worked faster than others

b) other non-LLM programming tools that enabled productivity boosts

While maybe (maybe?) slower programmers aspired to be faster, and maybe (maybe?) their managers wished they were faster, it was generally accepted. You might have someone who codes slowly, or medium, or fast, or superhuman fast. Just so long as they were "fast enough" for your business needs, it wasn't a huge issue. There wasn't a culture of shaming all of the medium-speed programmers into the ground.

And while maybe some programming tools did indeed offer productivity boosts, there really wasn't a culture of demanding all possible boosting tools be used. If you want to use Nano to edit, great, up to you. Emacs? Great. Visual Studio Code? Great. Debuggers? Great. Print statements? Great. Outside of some highbrow programmer ivory towers, there wasn't a widespread insistence that "only these tools shall be used because they let you write code the fastest".

But now it seems that there is demand that LLMs be used to write code as fast as possible as possible.

Why? Why now, is it suddenly important that code be written as fast as possible, and anything less should be mocked and derided?

tjr··on The recurring dream of replacing developers
Fair point! And I wasn't clear: my anecdote was me, personally, needing an instance of some software. Rather than me personally either write it by hand, or even write it using AI, and then host it, I just found an off-the-shelf solution that worked well enough for me. One less thing I have to think about.

I would agree that if the scenario is a business, to either buy an off-the-shelf software solution or pay a small team to develop it, and if the off-the-shelf solution was priced high enough, then having it custom built with AI (maybe still with a tiny number of developers involved) could end up being the better choice. Really all depends on the details.

tjr··on The recurring dream of replacing developers
Anecdote: I have decades of software experience, and am comfortable both writing code myself and using AI tools.

Just today, I needed a basic web application, the sort of which I can easily get off the shelf from several existing vendors.

I started down the path of building my own, because, well, that's just what I do, then after about 30 minutes decided to use an existing product.

I have hunch that, even with AI making programming so much easier, there is still a market for buying pre-written solutions.

Further, I would speculate that this remains true of other areas of AI content generation. For example, even if it's trivially easy to have AI generate music per your specifications, it's even easier to just play something that someone else already made (be it human-generated or AI).

tjr··on The past, present and future of LLM coding
At this point, if you code without the use of LLM, you are significantly slower than your peers. Arguing that that is not the case is like arguing that you can be faster than someone using a calculator.

A lot of truth there, but especially for basic arithmetic, I don't always use a calculator. I'm sure it can compute, say, "11 + 14" faster than I can, but it still takes me time to press the buttons. I reckon there will be instances where it would be just as fast, or faster, to NOT use an LLM, if the change is small.

← PreviousPage 2 of 34Next →