frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Stop Talking

https://gurkan.in/2025/12/stop-talking/
28•npstr•45m ago

Comments

giancarlostoro•30m ago
I get it. This is roughly me, I don't always have the best answers, but I know most things can always be done better. I've coined a few different terms over the years such as "marketing driven development" when I wind up working in places where the marketing team is driving the devs off a cliff, and pushing new features at the expense of ever having time to deal with technical debt. The industry really needs "Tech Debt Thursdays" or something.

There's always way more work to do and those key enhancements or research stories that could improve everything get deprioritized.

leetrout•21m ago
I didn't RTFA - just responding to you:

> Tech Debt Thursdays

Yes, "Fix it Fridays" is another alliteration.

Have you ever heard the phrase "man your battle stations"? Turns out in the US Navy there is also "cleaning stations" and there is a call for all hands to cleaning stations on the regular. I have proposed something similar on a few teams I've been on. Daily won't work and quarterly is too long. The problem is the sprawl that comes from cleaning up things that have unintended side effects. But yes, paying the interest on the tech debt needs to be normalized across our industry.

https://www.navy.mil/Press-Office/News-Stories/display-news/...

https://www.youtube.com/watch?v=fyJH8VbFE6g

shadowgovt•14m ago
As of late, I've been thinking about how "debt" may not be the right metaphor.

Fiscal debt is a one-dimensional number that becomes higher or lower from some offset, but it can't change direction. There's no "complex numbers debt."

But software engineering is only one-dimensional if your problem domain is so constrained that the only roadblock to execution is time-at-keyboard, and that's rarely the case in most software (especially startups and hacking). I've too often seen that debt just "evaporates" when the company pivots or the entire system is replaced by another system or rendered completely irrelevant to continue accepting the notion that debt works as a metaphor. Even in the small, too often I've seen things flagged as, for example: "debt - we should consolidate these two pipelines on top of a smaller set of helpers" only to see the use of the pipelines diverge over time such that it turned out to be a great first step to keep them separate and duplicated.

Sometimes things to be improved / cleaned up are obvious, but cleanup assumes taking disorder and making order out of it, and that requires us to know what order even looks like.

catigula•28m ago
This is an obvious AI slop post for anyone interested in these things.

Pangram: fully AI generated, confidence high.

CharlesW•21m ago
"AI detectors" are snake oil. https://mitsloanedtech.mit.edu/ai/teach/ai-detectors-dont-wo...
catigula•19m ago
1. I determined the post was AI myself. The detector was merely to satisfy someone disingenuously asking for "proof".

2. This one works. It's peer-reviewed.

mdorazio•15m ago
Determined how?
catigula•13m ago
By having native English language proficiency and an IQ above 100.

This isn't even particularly good slop. If you can't identify this, we're entering a not so good space.

Anyhow, I'd recommend you roll away from this hill because it's really not worth dying on. Common sense and peer reviewed slop detection aren't working for you. I provided my opinion and backed it with evidence. That's why I posted what I did.

hyperhello•17m ago
Yes, but I wonder if learning that part of a paper is likely AI and part of it is likely not, might be a signal.
shadowgovt•19m ago
That might be because the thing is whole-cloth generated from AI, but given that the author has a .in email address, it also might be that English is not their first language but they want to reach a broader audience with their message so they used AI to translate it into the lingua franca of modern technical and scientific discussion (which is, this is always funny to me, not franca).

AI generators do a better job of conversational output than traditional language translators (although there is a risk that if you can't read the target language as a non-native speaker, it can distort or destroy the message).

catigula•11m ago
Regardless, presenting AI generated content without the disclaimer *I AI GENERATED THIS CONTENT. IF YOU THINK MY MOST LIKELY BRIEF AND LOW-EFFORT PROMPT JUSTIFIES YOU READING POTENTIALLY LARGE AMOUNTS OF AI OUTPUT, CONTINUE* as if it's worth reading is something a lot of people aren't okay with.

I actually find it an insulting imposition on my time. I don't think this is unreasonable or even unusual.

theamk•25m ago
> The difference between “annoying senior sysadmin” and “good consultant” is often just whether you’re in a room that opted in.

So much that. No one likes "drive-by advice" - if you want something to be fixed, there should be a person responsible for that. Maybe it's you doing all the work, or you convincing management, or management who is asking for an advice... But if you are just saying "we should fix FOO by doing this and that" with no plans as to whom those "we" are, it's only annoying.

shadowgovt•21m ago
Communication bandwidth is a finite resource, as several years of managers have reminded me.

(Although, it's worth noting that in this era of more remote work, perhaps a little more read-in and context is useful to avoid burning time on back-and-forths that used to take minutes in front of someone's desk but can now take hours over Slack).

vrnvu•20m ago
Everyone can talk and give opinions. The real question is if you can actually make a difference. I tell people there's a gap between knowing how to do something and actually doing it. And that gap is a big part of our engineering skills.

If I'm not going to change something, I'd rather not talk or give opinions.

Related: https://strangestloop.io/essays/things-that-arent-doing-the-...

jetru•5m ago
This is a better way to say it.

Talking at the right place at the right time on the right topic is.

hyperhello•19m ago
Strategically, “stop talking” means nothing unless you would otherwise be slamming out ideas. You don’t need people who don’t talk, we have plants for that. You need your silence to say something.
pizzafeelsright•17m ago
Most problems have been solved except the ability to align incentives.

Until the desired outcome is defined and documented, holding off on solutions and effort would benefit both parties.

GuinansEyebrows•4m ago
what this misses (and unfortunately is not always an option, especially in larger orgs) is that instead of talking (read: complaining), just fix the damn thing and present the solution on a platter (on company time, of course). more often than not, if you've already addressed the issue and it's ready for primetime, people will not refuse the change.

if they do, there's an equal chance that you either didn't understand the situation to begin with, or you work in a team with poor leadership and strategy. learn from the former, leave the latter.

LLMs as a new alien subset of all possible embodied intelligences

https://advanced.onlinelibrary.wiley.com/doi/10.1002/aisy.202401034
1•inshard•2m ago•0 comments

Devtools Just Became AI Infrastructure

https://www.nibzard.com/bun-acquisition
1•nkko•3m ago•0 comments

Show HN: Informiton – Intelligent Healing Signals

https://informiton.com/
1•milabr•3m ago•0 comments

Show HN: GPTShirt – Custom t-shirts powered by Nano Banana Pro

https://gptshirt.app
1•nliang86•4m ago•0 comments

Bio-essential sugars in samples from asteroid Bennu

https://www.nature.com/articles/s41561-025-01838-6
1•ca98am79•5m ago•0 comments

Remodeling My Home Proved One Thing: We Need Robots to Build Houses

https://spicermatthews.com/blog/remodeling-my-home-proved-one-thing-we-need-robots-to-build-houses/
1•cloudmanic•7m ago•0 comments

New Elf-Stats Malware Campaign on NPM

https://www.npmjs.com/search?page=0&q=elf-stats&sortBy=published_at
1•eyberg•7m ago•0 comments

33 days, new WR in speedrunning a fan translation of a long-lost game

https://aeongenesis.net/news?date=2025-12-03T10:10:49-08:00
1•jordigh•8m ago•0 comments

OpenAI Agrees to Acquire Neptune to Improve AI Model Training

https://www.bloomberg.com/news/articles/2025-12-03/openai-agrees-to-acquire-neptune-to-improve-ai...
1•world2vec•11m ago•1 comments

Power Outage: On the Thermodynamics of History

https://www.laphamsquarterly.org/energy/power-outage
1•bryanrasmussen•12m ago•0 comments

The Metaverse Only Has 900 Users [video]

https://www.youtube.com/watch?v=ntPGl8UyIq4
1•thm•12m ago•0 comments

Sunflower Basic

https://wiki.xxiivv.com/site/basic
1•rickcarlino•12m ago•0 comments

The Triumph of Logical English

https://worksinprogress.co/issue/the-logical-triumph-of-english/
1•fanf2•13m ago•0 comments

US pressures Mexico for violating the 1944 Water Treaty

https://english.elpais.com/usa/2025-12-03/us-pressures-mexico-for-violating-the-1944-water-treaty...
1•geox•14m ago•0 comments

Ghostty Is Now Non-Profit

https://mitchellh.com/writing/ghostty-non-profit
5•vrnvu•15m ago•0 comments

Formally Verifying Advent of Code Using Dijkstra's Program Construction

https://haripm.com/blog/aoc-day-3-without-thinking/
1•seafoamteal•16m ago•0 comments

Micron to exit consumer memory business amid global supply shortage

https://www.reuters.com/business/micron-exit-crucial-consumer-memory-business-2025-12-03/
5•djkoolaide•16m ago•3 comments

Private employers shed 32,000 jobs in November

https://adpemploymentreport.com
9•JumpCrisscross•16m ago•1 comments

A central hub for LLM API config info: model-api.info

https://www.model-api.info/
1•rizzy_o•17m ago•1 comments

The Quiet Tensions of Specialized Roles

https://sachlabs.com/the-quiet-tensions-of-specialized-roles/
1•abnercoimbre•18m ago•0 comments

Authors retract Nature paper projecting high costs of climate change

https://retractionwatch.com/2025/12/03/authors-retract-nature-paper-projecting-high-costs-of-clim...
1•docdeek•19m ago•0 comments

Towards a Language for Optimization

https://theahura.substack.com/p/towards-a-language-for-optimization
1•theahura•21m ago•0 comments

Aerosol induction-removal system for mitigating airborne disease transmission

https://www.sciencedirect.com/science/article/pii/S0360132325010418?via%3Dihub
1•PaulHoule•21m ago•0 comments

Antiidle – A Free, Open-Source Anti-Idle Tool

https://github.com/uppnrise/antiidle
1•uppnrise•22m ago•0 comments

From Moderation to Mediation: Can LLMs Serve as Mediators in Online Flame Wars?

https://arxiv.org/abs/2512.03005
1•kelseyfrog•23m ago•0 comments

Waymo Now Autonomous in Dallas

https://twitter.com/Waymo/status/1995976302864580611
2•xnx•24m ago•0 comments

Kilo Deploy: Ship Apps Directly from Kilo

https://blog.kilo.ai/p/kilo-deploy
1•janpio•24m ago•0 comments

TSA's New $45 Fee at U.S. Airports Unfairly Punishes Families in the Fine Print

https://www.thetravel.com/new-tsa-45-usd-fee-for-americans-without-real-id-or-passport-penalizes-...
9•stn8188•24m ago•20 comments

Kiro Powers

https://kiro.dev/blog/introducing-powers/
1•janpio•25m ago•0 comments

Noninvasive imaging could replace finger pricks for people with diabetes

https://news.mit.edu/2025/noninvasive-imaging-could-replace-finger-pricks-diabetes-1203
1•mikhael•25m ago•0 comments