DrSiemer

It's not the most energy efficient workflow, but I work on relatively small codebases and I made a tool that let's me dump all of it in an LLM with a single copy/paste. This works surprisingly well with Gemini 2.5 Pro (1.000.000 ctx).

The only real mistakes it makes are some model specific quirks, like occasionally stripping out certain array index operators. Other than that, it works fine with 150.000 token size conversations. I've gone up to 500.000 with no real issues besides a bit of a slowdown. It's also great for log analysis, which I have maximized to 900.000 tokens.

A year ago I built a React Native Android app for my wife called "Pimp daddy", which she uses to track her earnings as an independent contractor.

The whole concept is meant to poke fun at the idea of me "checking up on her" (I file her tax returns) and the entire theme is 80s pimp styled.

Every time she submits something, she'll get a random pimp remark, like "Go get that money for me, girl!". She just rolls her eyes and ignores it, but it's what made it fun for me to work on it.

Edgy stuff like that could jeopardize my account in the near future. It might just be security now, but an automated "naughty words detector" will be an obvious next step.

I doubt I will invest any more time in hobby app development if I have to deal with some humorless overbearing watchdog telling me what I can and cannot install on my own device. Very sad to see Android following Microsofts anti power user direction.

That's actually not tinnitus but (from what I've been told) cochloreal hyperacusis, another form of hearing damage.

I always have that, but I only hear a random high pitched tinnitus noise in one ear, rising and falling in volume for max 10 seconds, about once every few months.

I can still hear old CRTs in my forties, although it's less maddening now. They had those mosquito devices, that are intended to repel kids, for a while at a shopping mall near me. They repelled me very effectively as well.

A friend once thought it was funny to try the 15.000Hz silent ringtone on me, although I had told him not to. It made me react without conscious input and I nearly broke his phone.

In my experience, a core skill of programmers is that they are people who are able to deal with setbacks and frustration in a good way. Those come with the territory.

I've always wondered if there is a correlation between developers and people that can enjoy tough but fair games like Super Meat Boy because of this.

When using a reasonably smart llm, code moves are usually fine, but you have to pay attention whenever uncommon words (like urls or numbers) are involved.

It kind of forces you to always put such data in external files, which is better for code organization anyway.

If it's not necessary for understanding the code, I'll usually even leave this data out entirely when passing the code over.

In Python code I often see Gemini add a second h to a random header file extension. It always feels like the llm is making sure that I'm still paying attention.

Exactly this. In the early days of image generation I once spent about 9 hours carefully remastering a generated image until I was happy with the result.

Are those hours completely without value, because I used genAI during most of them? It involved a LOT of consideration and a few pretty tough choices. But because I didn't do my 10.000 hours, and it's sad for those that did, I'm not allowed to create this way?

Whatever, I'm still putting it on my wall. It means more to me than most "legit" art does.

So tired of pointing that there is something in between "marketeer story tellers" and "real artists".

Bots in the Hall, Neural Viz, GossipGoblin, even Joel Havers animations. All made with genAI and all undeniably creative works, that could not have been created, at least not in that time frame, by a single person without it.

I love Matt's work and often agree with him, but the "no heart" take is just too harsh.

One of my future hobby projects is developing a low-cost device that can detect pigeon calls and will automatically respond with random audio of whatever local predator they fear the most. I can't stand their monotone calls in the morning.

These days I'll just dump all relevant code into an LLM and have it explained to me instantly.

Being able to ask questions about the parts that are unclear (or just plain wrong) is so much easier than trying to cram the entire thing into my brain RAM.

In my experience it actually helps me learn faster too, since I rarely get stumped on random gotcha's anymore.

You can't stop online stupidity and misinformation with censorship. It would at best create an echo chamber of government supported online stupidity and misinformation.

Sounds good!

A German forum I'm on allows members limited invites based on participation. The catch is, you are responsible for the people you invite. If they get in trouble, you will share a part of the punishment.

I'm not a fan of this option, but it seems to me the only way forward for online interaction is very strong identification on any place where you can post anything.

Ha, I'm working on a similar tool: https://github.com/DrSiemer/codemerger

Glad to see I'm not the only one who prefers to work like that. I don't need many different models though, the free version of Gemini 2.5 Pro is usually enough for me. Especially the 1.000.000 token context length is really useful. I can just keep dumping full code merges in.

I'll have a look at the alternatives mentioned though. Some questions just seem to throw certain models into logic loops.

Wish they would stop mentioning AGI. It's like the creator of a new car claiming it's a step closer to teleportation.

Switch space? On any Windows OS ctrl-arrows (left and right) move the cursor by word. I don't exactly understand how people work with code or even do text editing without it.

I don't think anybody in our team uses tabs, but afaik you still call it tab width

Meanwhile, the war on tab width marches on. Back when I got tired of inconsistent tab widths, I simply based my preference on the silly fact that some research showed how developers using two spaced tabs make slightly more money.

Among our team we do not enforce a specific width, but respecting the choice of the original author of a project is required.

Kind of annoying how LLMs will sometimes randomly decide to change tab width.

Not if you learn to use the right shortcuts for it (ctrl+arrows, mostly).

An expiration date could be hard to estimate, but a minimum level of support based on sales might be worth thinking about.

As I understand it, that's not the main focus of this initiative though. It's mostly about opening up dead games for stuff like private servers, so we are not forced to lose access to all the art within when the developer pulls the plug.

It seems like a relatively small ask to at least _allow_ hardcore fans to keep something alive. I can imagine publishers that have a great connection with their community (like Coffeestain) actually benefitting from handling stuff like this in a decent way.

A clear example of whataboutism.

It's not like we are donating anything besides a minute or two of our time... It's just a signature.

If that was all that was needed to force law makers to fix human rights violations, I'll sign that as well, but most of those problems are not as easy to fix.

We're not that far off from generated pixel art looking indistinguishable from the real deal.

Friends working together generally only happens when they are still in school and have a surplus of time for projects that do not generate a consistent paycheck. As somebody with a lot of ideas and plenty of capable friends, my experience is that none of them are interested in doing unpaid work once they get on the job market.

For most people, especially younger, pixel art is just a retro esthetic. Most indie developers do not even have the money to pay themselves, let alone hire an artist.

Settings page does not work. Profile page does not work. I've been waiting for my first test for 30 minutes now. Not a great first impression...

Why would you ask a 1B model anything? Those are only useful for rephrasing output at best.

It has gotten to the point where I now almost exclusively associate the use of emojis outside of social chat with LLM output. They are on the same level as all the "useful comments" added to my code: trash that the prompter was too lazy to remove.

Would this also work for adding effects to existing audio? A simple reverb and pitch bend on a recorded vocal would make me a lot more excited than experimental synth effects.

I think articles like this don't garner a lot of attention, because why would anybody even bother to engage? If you don't see this unstoppable steam train for what it is, well, enjoy your screaming in the void.

We can all agree the current generation of AI lacks humanity, but we are not going to stop using it because of that. Would be nice if we can get the energy cost down though.

It's a Python script that exports all the playlists, including liked songs.

As somebody who has listened to hundreds of audiobooks, I can tell you authors are generally not the best choice to voice their own work. They may know every intent, but they are writers, not actors.

The most skilled readers will make you want to read books _just because they narrated them_. They add a unique quality to the story, that you do not get from reading yourself or from watching a video adaptation.

Currently I'm in The Age of Madness, read by Steven Pacey. He's fantastic. The late Roy Dotrice is worth a mention as well, for voicing Game of Thrones and claiming the Guinness world record for most distinct voices (224) in one series.

It will be awesome if we can create readings automatically, but it will be a while before TTS can compete with the best readers out there.

A valid point of criticism, but I do wonder if this only applies to those who can (still) easily spot LLM assisted output.

Many people lack the time or writing skills to produce something elegant by themselves, so for them it's like fake breasts: an upgrade for the less discerning larger part of the target audience, as long as they don't look to closely.