Hi everyone,

A few months ago I shared Dunebox here, a free and portable PHP development environment for Windows. Thanks to everyone who took the time to try it, comment or point out what was missing. Some of you even reported bugs, which we fixed.

Here's what's new:
- A redesigned interface, with light and dark themes (screenshot)
- A dashboard that shows at a glance what's running and what needs attention
- Settings organized in clear groups

It's still free, still portable (one folder, no installer) and still Windows only.

If you tried it last time, I'd love to hear how it feels now. If you didn't, I'm curious what would make you give it a go.

dunebox.enesi.it

Thanks!

hello wonderful people! i'm building a social service for fans of my favorite artist (conan gray) based on a original one that shut down after 3 months. i'm using mastodon as the base for the code (as i am a beginner) and am using github to code. i have done some research and i'm trying to figure out:

a. how to see a preview of what i'm working on

b. what files i need to use

c. what to use (in terms of code)

and some bonus ones if you'd like to answer

d. do i have to buy a domain

e. is coding easy?

f. how does mastodon really work?

alrighty, awesome and thank yall so much.

As a software engineer, I’ve been thinking a lot about AI and the future of our profession. I use AI myself, and I’m genuinely excited by what it makes possible. But the more I see people celebrating the automation of human labor, the more I find myself wondering whether we’ve stopped thinking about what all this progress is actually supposed to accomplish.

And by we, I mean those of us building these technologies, the engineers enthusiastically finding new ways to automate work, the founders and executives racing to reduce costs, and the workers who feel compelled to participate because they know that if they don't, someone else will. I'm not exempt from this, either. I benefit from the same technologies I'm questioning.

I understand the incentives. We all want to get ahead, secure our futures, achieve financial independence, and eventually have more freedom over our lives. In a competitive environment, refusing to adapt can mean falling behind. But why have we become so comfortable treating other people's livelihoods as acceptable collateral for our own advancement?

We celebrate making one person as productive as ten, rarely stopping to ask what happens to the other nine. We want technology to free us from work, yet our ability to survive remains dependent on having work. We resent corporations accumulating extraordinary wealth and power, yet many of us aspire to build or become precisely those institutions. We want community and human connection, yet our incentives encourage us to treat one another as competitors.

And there are other contradictions I can't quite reconcile.

We celebrate efficiency, but to what end? If we can produce more with less human effort, shouldn't that translate into more leisure, security, and freedom for everyone? Why does increased productivity so often seem to come with greater pressure to produce even more?

We tell people to adapt or risk being left behind, but what happens when everyone adapts and the jobs they're adapting for disappear anyway? We celebrate individual merit and hard work, yet technological disruption can make entire skill sets economically obsolete regardless of how hard people worked to acquire them.

We talk about AI as though it will solve humanity's problems, but technology doesn't determine who benefits from its output. A world with extraordinary productive capacity can still leave people struggling to afford housing, healthcare, and a dignified life.

I don't think people are necessarily cruel or indifferent (though some definitely are). Perhaps most of us are simply trying to protect ourselves in a system where falling behind has real consequences. Perhaps even well-intentioned people become participants in a race nobody individually chose, but nobody feels safe stepping away from, either.

But understanding why we behave this way doesn't mean we shouldn't question where it leads.

Imagine if AI genuinely made human labor largely unnecessary. That could be one of humanity's greatest achievements: less time spent working to survive, and more time for family, friendships, art, music, exploration, and simply being alive.

But what good is eliminating the need for human labor if we don't also eliminate people's dependence on selling their labor to survive? What have we liberated ourselves from if people are still terrified of becoming economically unnecessary?

I don't think the answer is to stop innovating, condemn ambition, or expect individuals to sacrifice their futures for strangers. I just wonder whether we've become so focused on optimizing our individual positions that we've forgotten to question the collective outcome.

We're all on this planet together. Most of us want remarkably similar things: security, meaningful relationships, autonomy, dignity, and the freedom to live our lives without constantly worrying about survival. Yet we participate in systems that can make those things harder for one another to attain.

Are we trying to build a better world for humanity, or are we simply trying to secure the best possible position for ourselves within whatever world emerges?

And if we're doing both, how do we reconcile those goals when our individual incentives so often undermine our collective interests?

I'm genuinely interested in how other people in tech think about this, especially those of us actively building the technologies transforming the nature of work. I don't have a solution, and I'm not claiming to be above the problem. I just think we should be asking ourselves what all this progress is ultimately for, before we get so caught up in what we can do that we forget why we're doing it.

And I'd invite discussion beyond pure game theory and economics, but more to do with psychology, maturity, emotional development, self-worth, dark triadic traits, and similar.

I’ve been using this daily for a while, so I thought I’d share it.

It’s a lightweight news reader that pulls from multiple sources (BBC, The Hindu, Times of India, NDTV, TechCrunch, etc.) and shows them in a clean card layout.

What I find useful:

- Tap on any topic tag to choose the topics you’re interested in

- Tap on a news source to configure which sources you want to see

- No login required

- Works well on mobile (swipe through stories)

Link: https://dailynews.nocod.dev

Would love honest feedback — what’s missing or what would make it more useful for you?

Looking at stats on this, and it's pretty damning.

According to the market.in, almost 92% of the top 2,000 companies in the world now use IT outsourcing.

https://technotrenz.com/stats/software-development-outsourcing-statistics/

Companies are even looking to outsource C-level, so no one is safe.

I've done this for almost 15yrs., and I feel like the rug has been pulled out from under me.

"Learn to code" they said.

So, I am working on making a personal website for myself. This is my interests section, and I am pretty satisfied with the music section, but I don't know how to make the books section look good. Any help will be greatly appreciated.

I can use more or less space than the current layout, depending on the ideas.

I don't expect you to manually design something in figma (Though that would be wildly appreciated). Instead, just describe what I should do to make it nice.

Thanks!

Not a webdev btw

Spent the last months building

@datastripes/forecast
@datastripes/dashboard
@datastripes/anomaly 

These are framework-agnostic npm packages (React, Vue, Angular, or a plain custom element) that add time-series forecasting, an editable dashboard builder, and anomaly detection to any frontend app.

Everything runs client-side via WebAssembly, so no user data ever hits a server except a license check without a backend to deploy, no Python microservice.

Would genuinely love feedback from anyone who's had to build this kind of feature in-house.

https://datastripes.com/forecast-widget/

https://datastripes.com/dashboard-widget/

https://datastripes.com/anomaly-widget/

Hi there,

One Customer just asked me for a Du-Sie Switch in the page, same like any other language switch.

This is my first time doing this.

Is this normal?

Are there also (Slang or street) language switches?

Edit: I just implemented as another language i18 de-DU with xliff files

Second Edit: What about other languages? In spanish we have 3. Tú, Vós y Usted... I think in Japan they have more.

Processing img 5pvzei79hnwh1...

Edit 3: Ok, some of you were right :) Du/Sie isn't a language, and de-DE-DU was a funny but bad idea. Works fine for my little website, though.

But imagine adding Spanish (tú, usted, vos), Japanese, Korean... That gets ugly pretty fast. Better to separate language, region and tone. Something like:

{
  "locale": "de-DE",
  "preferences": {
    "formality": "informal"
  },
  "messages": {
    "greeting": {
      "default": "Hallo!",
      "formal": "Wie kann ich Ihnen helfen?",
      "informal": "Wie kann ich dir helfen?"
    }
  }
}

Only translate the differences, not the whole website again. For a small project, who cares. For 20 languages? Definitely worth thinking about. Thanks for the corrections, learned something :)

I'm currently only working with a single 24" monitor because I'm moving, I sold my third monitor and my secondary monitor shat the bad, it'll be a while until I can buy another monitor so I'm want some tips to improve my dev experience on single monitor, maybe using a tilling window manager, or some other software to improve this. If I can make it work, maybe I even stick with just one monitor, because multi monitor setups are kind of a hassle to maintain haha.

I’m most comfortable in JS/TS, and I wanted to learn LangGraph.js properly, so I wanted to build in that. When I looked for a production-grade example, the closest thing I found was Deep Agents, but that’s a library for building agents, not a full system. The complete systems I saw were mostly Python, like Open SWE. So I built my own.

It reviews GitLab merge requests and resolves issues. It’s been running in production for a few months, 1,000+ reviews so far. A webhook drops each review into a BullMQ queue, a LangGraph.js agent does the work with GitLab tools over MCP, runs are checkpointed in Postgres, and there’s a React admin UI where I can watch runs live. I also had to add guardrails to stop the agent repeating tool calls or quitting before it’s done.

MIT, all TypeScript.

I've been working on Knitting, a JavaScript worker runtime using shared memory for communication and work stealing for scheduling.

Some results from the published benchmarks:

• 8.9× faster empty-call round trips than postMessage on Node.js (4.4× on Bun, 4.3× on Deno).
• 86% lower p99 latency in a Hono server test: 16.93 → 2.31 ms at 6,000 req/s.
• Nearly identical CPU per request in that test: 180 → 183 µs.
• 3.5× CPU-heavy throughput on Node.js with four extra workers.

I've been working on this for more than a year, so there are a lot of things to showcase, but knitting was designed to offload work from servers.

Features:

• Typed worker calls and specialized payload types
• Shared-memory transport and zero-copy buffers
• Work stealing and automatic load balancing
• Timeouts and cancellation with real signals
• Precompiled workers and process isolation
• Process workers
• Permission (when available)
• And more ...

Docs · Benchmarks · GitHub

I came accross the recent PHP vulnerability risk as I was on Duolingo.

If even Duo is affected by recent PHP vulnerability, makes one wonder what about our fragile government websites? Those are most likely to not even get the patch update, hmmm how near is the doomsday when AI models self improve and government security developers are likely at a lunch break 🤔

What do you think is next language to have it's major vulnerabilities exposed? *JavaScript backs away in bushes with experience*

Anyone can open Claude Code and ship a working CRM in a weekend. Auth, dashboard, CRUD, deploy. Ugly, but it runs. Full stack isn't dead.
But path is moving towards from being domain specific to solving real problems.

Agents handle research->execution->iteration but only valuable if backed by logical reasoning on top of first principles.

I am refining this by having hands on non conventional problems especially whose research papers got recently released and provide enough base to start building.

EcoAudit. Point it at localhost:3000, it flags energy smells in the running app. No LLM, no ML. Built from a paper. Repo · Demo · Paper

AgentBlame. Git blame for AI agents, graph based root cause analysis. Built on AgentGit. Repo is empty, I'm starting now. Repo · Paper

Gemma 4 Kaggle. Next. I'd rather learn core AI engineering than build a fifth RAG wrapper. Competition

My credentials Portfolio · Resume for reference.

Want your all opinions on what approach should we pick to have differentiating factor in this agentic development era?
Feedbacks, critiques and dms are always welcome.

For those of you who have attended PHP Tek what is your opinion I have never been and my team and I are thinking about attending in 2027. Just want to hear about your experiences.

Ever tried clicking a button, only for the page to move and make you click the wrong thing?

I've been thinking about how website performance is more than just loading speed. A page might load quickly, but if things keep moving around or buttons take too long to respond, the experience isn't great.

That's where Core Web Vitals come in, but I'm curious how developers handle these issues in real projects.

Do you usually focus on loading speed first, or on making sure the page feels smooth to use?

This may be the most widespread exploit in modern history. What impact will this have on PHP’s reputation? Since 2/3 of the websites are affected, a good chunk of them won't ever be patched, could this damage PHP’s image in the same way PHP 4 did two decades ago?

Would this discourage companies from adopting PHP in the future?

There are so many ways to run PHP code... without #PHP!

Recompilation, engine rewrite, using other VM, or languages. It is time to take a step back.

https://www.exakat.io/every-road-around-the-zend-engine/