Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Lines of code got a better publicist

curlewis.co.nz

Post reply on HN
Print view View original

281–290 of 309 posts

Re: Lines of code got a better publicist

#281

Post by zombot »

selicos wrote: post #219
sunaurus wrote: post #6

I'm constantly thinking about that Microsoft guy who posted something like "we want 1 million LoC per engineer per month", which basically read as satire to most engineers I talked to, except apparently it was not satire at all, and indeed seemed to reflect the position of many CEOs etc when it comes to LLM code generation. I do think that over the past few months, it feels like the hype around producing unmaintainab…

1000000/25/8/60 = 83+ lines of code per minute. 100000 LOC per month /25 days per month /8 hours per day / 60 minutes per hour That seems...problematic for anyone doing code reviews.

> That seems...problematic for anyone doing code reviews.

No, it's incentive to let LLMs do the reviews, supporting your tokenmaxxing efforts.

Re: Lines of code got a better publicist

#282

Post by dijksterhuis »

mpyne wrote: post #256
dijksterhuis wrote: post #249

Earlier quoted context omitted.

sorry, but this just sounds like a rationalisation for "we built the wrong thing". we should just be straightforward, say "we built the wrong thing" and then ask how we built the right-er thing.

> sorry, but this just sounds like a rationalisation for "we built the wrong thing". If it were so easy to decide what the right thing is to build before you build it then business would be easy. That's the whole reason options have value. Having 3 shippable products ready to go when you can only effectively ship 1 puts the whole team in a much better position than choosing 1, focusing everyone on it, and hoping you…

> If it were so easy to decide what the right thing is to build before you build it then business would be easy.

It's not easy. That's why it's important to be straightforward and just move on without all the navel gazing.

> ... hoping you hit the lottery with product-market fit.

There's this thing called "research" where you talk to real people, instead of guessing.

> That doesn't retroactively make their effort worthless

No-one said building the wrong thing was worthless. Life is one continuous mistake.

---

we are saying mostly the same thing i'm pretty sure, especially in your other comment reply (https://news.ycombinator.com/item?id=48498912). although i feel you're dressing it up a little too much for my liking. i prefer being a lot more plain and direct about it (and probably a bit arsey).

ruthlessness is an asset when it comes to we built the wrong thing. ruthlessness gets us moving on faster.

Re: Lines of code got a better publicist

#283

Post by adamddev1 »

> I’ve watched this industry absorb higher-level languages, IDEs, autocomplete, agile and devops, and there were always crusty hold-outs reminiscing about the good old days before X came along and ruined everything.

But all of those things were consciously built, deterministic and transparent tools. LLMs/AI are something fundamentally different.

Re: Lines of code got a better publicist

#284

Post by trashb »

adamddev1 wrote: post #283

> I’ve watched this industry absorb higher-level languages, IDEs, autocomplete, agile and devops, and there were always crusty hold-outs reminiscing about the good old days before X came along and ruined everything. But all of those things were consciously built, deterministic and transparent tools. LLMs/AI are something fundamentally different.

yes it does seem fundamentally different. Most claims about AI are unmeasured marketing and favoritism. Similar to how management wants you to use eclipse or codeblocks to a developer that uses Emacs skillfully because it is shiny and new and they don't understand the requirements of the developer.

Re: Lines of code got a better publicist

#285

Post by trashb » 

  When a company says “AI made everyone more productive, so we need fewer people”, I want to see the evidence - and I don’t believe it exists today.
I think these companies doing "AI layoffs" do actually see improvement though it is a placebo and not caused by the AI usage. Don't we know for a long time already that leaner software teams perform more efficiently?

  don’t read any of this as anti-AI
I am not afraid to say I am anti-AI it surfaces a rot that in this industry marketing ideals and anecdotes have more impact then measured performance and that many people still find it very hard to estimate a developers performance, impact.

AI is shit, doesn't speed up my work. Only 10/20% of programming is typing and AI can do that fast, but the whole process no. If you disagree show me a proper study where actual improvement is measured.

Probably you could get a cheaper and more constant improvement if you make sure the developers are properly trained in the IDE's and environments they are already using. For example give everyone a Unix programming course and a course in their preferred IDE.

Re: Lines of code got a better publicist

#286

Post by nhinck3 »

monkpit wrote: post #221
getnormality wrote: post #5

This weird trend reached an apex in a Feb 2026 OpenAI blog post [1], recently on the front page [2], which describes the process for building... something... written 100% by agents. There is no description of what the thing is, no indication of what value it provides its users. The closest it gets is "the product has been used by hundreds of users internally, including daily internal power users". But the fact that t…

Sorry, was it not this? https://github.com/openai/symphony

That does not appear to have a million lines of code

Re: Lines of code got a better publicist

#288

Post by sanderjd »

dijksterhuis wrote: post #249
sanderjd wrote: post #186

Earlier quoted context omitted.

It's definitely that, which is very valuable, but it's also the optionality value additionally. You had the option to launch the thing, which you wouldn't have had if you had never worked on it at all. It's notoriously difficult to properly value optionality, but it definitely has value, and often a lot of value.

sorry, but this just sounds like a rationalisation for "we built the wrong thing". we should just be straightforward, say "we built the wrong thing" and then ask how we built the right-er thing.

No, optionality very literally has value. If I buy an option to purchase some commodities at some price, and the market moves against me, I lose the price of the option. By your logic here, the thought should be "buying the option was wrong, because it didn't go my way and I lost the price of the option". But it's often the case that actually buying that option was a good move.

The kind of optionality I'm talking about in software projects is not so clean to account as the financial instrument, but it has real value in just the same way.

Re: Lines of code got a better publicist

#289

Post by adamtaylor_13 »

By this argument, I guess we should write our code slower too. Maybe we should only use a single finger as we type. And add an extra day between email responses.

Because by the inverse of their argument, slower MUST be better, right?

This is silly. LLMs are a diesel powered keyboard. If I've got a backlog of features, why shouldn't I ship all of them? If they're bad ideas, I can also just remove them.

Nothing about SDLC best practices has changed EXCEPT the ability to increase volume.

Re: Lines of code got a better publicist

#290

Post by sanderjd »

dijksterhuis wrote: post #282
mpyne wrote: post #256

Earlier quoted context omitted.

> sorry, but this just sounds like a rationalisation for "we built the wrong thing". If it were so easy to decide what the right thing is to build before you build it then business would be easy. That's the whole reason options have value. Having 3 shippable products ready to go when you can only effectively ship 1 puts the whole team in a much better position than choosing 1, focusing everyone on it, and hoping you…

> If it were so easy to decide what the right thing is to build before you build it then business would be easy. It's not easy. That's why it's important to be straightforward and just move on without all the navel gazing. > ... hoping you hit the lottery with product-market fit. There's this thing called "research" where you talk to real people, instead of guessing. > That doesn't retroactively make their effort wor…

I don't understand why you interpret "we should recognize that it's worth paying for optionality and sometimes we decide not to exercise the option and that's normal and fine" as naval gazing. I think it's the opposite, that feeling butt hurt about the projects you work on not shipping is what is naval gazing.

To me all your "I prefer being plain and direct" just sounds like someone who hasn't thought much about why building the wrong thing is not worthless, and why those continuous mistakes in life are worthwhile, and isn't really interested in thinking about things at more than the surface level.

It does seem like you aren't really disagreeing with us here. But you're just saying "don't make me think about why we agree about this!"

Post reply on HN