OhSoHumble

> Code is a beautiful form of creative expression, as rich as literature or music

Something I'm trying to do right now is to build something and avoid using LLMs to write any code. I still use it to consult. I'm writing a Dota2 tournament match aggregator in Elixir that takes tournament streams and chronologically orders them in a format that makes it easier to watch them sequentially since I find YouTube hard to use for ingesting series of videos.

I'm building it because... I like programming. I like making things. I find that LLMs are making me intellectually lazy and making things with them feels unfulfilling. I want to build. It's human to want to build.

Oh yeah, no shade against the language itself. I had fun learning it and using it for some toy development years ago - and the TEA still exists in multiple library implementations. Just wild to see the creator of the language emerge from the fog like that.

Oh my God, I had no idea this project was still alive. I don't mean to throw any shade but I had assumed that the lid was on this turkey.

Yeah, I apologize for how restrictive "want" is - I'm not sure what the appropriate word is in this context. It's more that content engagement is what drives web traffic regardless of whether or not users express desire for human connection. The social part of social media doesn't drive website expansion - content does. We love it. Yum, yum, yum.

I think if you sit a lot of people down and ask them "would you rather have cat videos or the potential to meet new friends (or maintain connections with old ones)" and people will say the want the latter but reward platforms that give them the former.

Old Facebook was maybe closer to the ideal of connecting people but Meta realized that it just doesn't drive ad revenue or engagement so they refactored it to better suit ad delivery streams and put more effort into Instagram (because it's better suited for that type of revenue model).

I actually strongly disagree with this. This is how many social media sites have failed.

I don't know if alternative social media can ever scale as large as Reddit. Tildes.net and Lemmy both have similar activity to Hacker News and I'd consider them both "successful."

The truth is that Reddit, Instagram, Facebook, etc are not websites used to drive connections between human beings; they're media machines meant to create ad distribution channels. And that's what users want.

When someone uses Instagram, they don't want to see their friends vacation photos - they want to see Cardi Bs latest post. Meta attaches ads to that interaction and that's how they drive revenue.

When someone uses Reddit, they aren't seeing what their neighbor has to say - they're consuming a long flowing rage bait story or a web comic. Sure, there are smaller, niche communities but that isn't the bulk of the site traffic.

And, honestly, content is addictive. I'm personally not on Reddit to make friends or deepen my connection with other humans. I'm there to scroll through memes and cat pictures.

I think that smaller social media sites fail because they don't emphasize media generation and distribution. That reframing still presents difficult problems to solve but I feel like that simple perspective pivot helps with growth. Like, if you wanted to take the "OG" Reddit approach to bootstrapping a social media site you could just programatically generate massive amounts of fake content. Rage bait stories, product placement, etc. Just firehose as much media to drive engagement until you get actual human beings to take part.

That is is growth is what you want. It's arguable whether hyperscaled social media platforms actually make society better or individuals happier.

This is kind of how I feel about any business. Any new initiative requires an IMMENSE amount of capital. If not to build the product, then to buy the exposure.

It's just hard launching a project in general - especially a B2C app. Someone could write the best software imaginable and never get a single user.

I've actually been turning it over in my head recently about whether or not it's possible to launch a software business without venture capital because without large amounts of ad expenditure or "growth hacking" I'm not sure how you would even go about getting a software business off the ground.

Cursor is a good enough product to be purchased for 60 billion. It launching unsuccessfully 8 times kind of shows how hard it is to pitch an idea to consumers.

Gitea has artifact support?! We are truly blessed.

Mmm, this is for an Etsy-lite website ( https://plukio.com ) though honestly I've used Gitea for YEARS - even back when it was Gogs.

I personally just self-host. The project that I'm working on right now uses Gitea internally with an action runner.

I use Digital Ocean and run a Nomad cluster on top of that. Gitea and its action runner builds container images and then pushes nomad job definitions to the cluster. I have zero downtime deployment and rolling deploys. Dagger.io is in there somewhere to make local CI mirror what happens in the action runner.

Setting up Gitea was honestly just an afternoon of work. The sqlite database is backed up to Cloudflare R2 and the code is mirrored. It's unlikely that my project will take off to the point that I'll need to upgrade Gitea to something else, but it's extremely easy to setup, maintain, and build CI/CD on top of.

I don't know if this falls under good ethical behavior but the one time I needed to mass scrape sites was to build an affordable housing directory for the Bay Area. Whether or not a unit of housing was under rent control was only available through private entities or apartment hosting sites and it was scattered and hard to find.

In my opinion, a directory of subsidized housing should have been provided by local governments and not through a plethora of real estate websites.

I tried out NixOS a few years ago but recently transitioned back to Rocky Linux and Ansible. I know that Nix is treasured by some but it always came across as an esoteric tool for functional programming idealists. I found the community to be split between people who were genuinely helpful and people who were just... not.

I found Nix just really hard to work with. The documentation was just so poor and every aspect of Nix just seemed to be divorced from pragmatism.

An example of this, years ago, was that I wanted to do something VERY simple: codify the creation of a directory in NixOS. It took me 6 HOURS to find the relevant code for doing that. I couldn't even get an answer out of the Discord server.

I don't know if I'll ever pick it up again. The learning curve was incredibly steep and it's just not on job descriptions and I've never worked in a shop that has used it. I tried it out as a curiosity, found that it was hair pullingly frustrating to use, and moved on.

I'm working on https://plukio.com. A lightweight alternative to Etsy built in Elixir. I saw a lot of people on the Internet (and a couple of people in real life) complaining about how there really is no equivalent for "old Etsy" and how a lot of people who sell handmade pottery and jewelry have turned to custom websites and Instagram.

I have a mailing list for when I launch and some local makers in Portland are interested so I'm hopeful.

Oh, the alt tab developer! I actually love your product.

I'm pretty smitten with Phoenix as a framework. A lot is built into the BEAM by default and the Elixir ecosystem is quite lovely.

I've been building an Etsy-lite replacement in my free time with it: https://plukio.com/

It's pretty enjoyable. I also like using Ash for data modeling and business logic expression. It's also quite lovely.

I've caught Claude Code generating some pretty egregious security vulnerabilities. I'm using it to build an AI RPG site and the goal is to use web assembly as a bridge between author submitted code and LLMs in order to help shore up state management at the game level.

The language that I picked for the game runtime is Python. Claude really thought that the best way to validate user submitted Python was to bypass the WASM sandbox and execute it within the application container using shell exec - essentially opening up an RCE vulnerability.

I also find that the quality of Claude Code degrades substantially. Claude really wants to implement every feature in as bespoke way as possible. This is fine when you first generate the project but over time you'll find that every web modal is implemented differently. Every button is different. Business logic is disconnected. It's why agentically produced codebases are MUCH larger than they should be; every feature is developed in a vacuum.

Then I'm trying to shove stuff in my AGENTS.md or CLAUDE.md files like "ALWAYS look for existing patterns within the codebase to keep it consistent." But the harness doesn't always work and it'll generate useless, verbose code anyways.

In some cases it's useful - like if I am shaky on the DSA knowledge needed for a specific operation or optimization then Claude can replace Stackoverflow. But, man, I'm so frustrated with it.

> So far Opus 4.5 has been the one LLM that keeps mostly coding in a, how to say, predictable way even with a existing code base.

I find this to be true only if you have very explicit rules in CLAUDE.md and even then it still messes up.

I have "you will use the shared code <here>" twice in my CLAUDE.md as it will constantly write duplicate code.

Something that is also annoying is that if it moves some code somewhere with the intent to slightly modify it I've seen it delete the code, then implement from scratch, and then modify it to what it has been specified to do. This completely breaks tests. I then have to say "look at this earlier commit - you've caused a complete regression."

> AI is pretty bad at Python and Go as well

I disagree with this. At least for Go.

It's what I had in the drawer.

I have something similar that I built a long while ago. It's a large Do Not Disturb sign. There is a raspberry pi mini that controls power going to it. There is a corresponding daemon running on a PC that checks Discord call status. In call? Power to sign. Not in call? No power to sign. The client/server daemons are in Golang and communicate over RPC.

What do you feel is a good approach to licensing user generated content?

I ended up building out a "spec" for Opus 4.5 to consume. I just copy-pasted all of the documentation into a markdown file and added it to the context window. Did fine after that. I also had the LLM write any "gotchas" to the spec file. Works great.

Also using Kustomiza and Argo. It's really good imo.

This makes me think of using org mode to build runbooks.

> The community is extremely polarizing. Not just politically, but even from just being a general contributor. Depending on who reviews your PR, you can have a very pleasant experience to a downright demeaning one. I get that every community isn't perfect, but after years of dealing with generally the same problem people, I just quit contributing.

God, I feel this. I ended up just stop trying to seek help because most of the time I'd come across the MOST self-righteous people I've ever met.

I don't think Nix will ever be easy to use because half the community genuinely does not care about their fellow human beings enough to write software for their fellow human beings.

That simplicity can lead to more complex and hard to read/support code.

For example, to encode a JSON structure with a dynamic top-level key you need to write a custom marshaller OR marshal twice. That's... awful. Like bonkers level insane.

I'm currently an American who might move to the Netherlands if my interview goes well enough. The projected salary that I'll have is 105k euros. Reading around, this seems like a relatively okay rate given cost of living. I'll be leaving a decent amount of money on the table from my current job but I'm relatively young and want to use the opportunity to be in Europe for a two to five years before returning to the states.

Is there anything about your personal experience in the Netherlands as an American engineer that you could share?

I have been running NixOS on my laptop and on my homelab.

I would not use it for a production system. Currently trying to migrate off of it to, I dunno, something else and Ansible. Nix has eaten hours of my life and upstream packages break all of the time. Couple of weekends ago the Mullvad module broke. Before that Virtualbox. Before that ZFS. You can have a perfectly fine configuration that you never touch but upstream instability will prevent a nixos rebuild from actually working.

I've commented on this before in a different k8s thread (one about a k8s outage) but something that bears repeating is that the entire job market is Kubernetes.

My personal experience is that it is very, very hard to find a job right now if your professional experience is primarily non-k8s orchestration systems. Most job positions out there require deep Kubernetes knowledge as well as hands-on experience with different supporting products like ArgoCD and Flux.

I chose Nomad for a large initiative at my current employer and it is honestly pretty amazing given the use case but I regret choosing it because I feel like I'm unhirable now given that every devops/SRE/platform engineering position open on the market (each one with hundreds of applicants) is heavily Kubernetes focused.

What's really unfortunate is that it feels like not choosing k8s is career suicide for SREs/DevOps/platform engineering.

I chose Nomad for my current role and it works fantastic.

I used ECS in my last role and it was also fantastic.

I'm currently gently looking for a job and the market is brutal. I'm not even getting responses back from applications and I have ten years of experience. Every job posting I see requires deep k8s knowledge and hands-on experience with some ancillary k8s project like ArgoCD or Flux. I'm actually thinking about downplaying non-k8s experience on my resume and just floating the k8s that I do have straight to the top because, at the very least, it'll get my resume looked at.

If I was going to go back in time, I'd pick k8s instead of Nomad or ECS even if they weren't the best contextual choices because, hey, I gotta' put food on the table.