koromak

I desperately want this to be true, but at least in my sector, it isn't. You still need talented and knowledgable programmers, but they don't do very much programming. Its all code review, infrastructure, devops.

At least for a small business, users are catching on that they can build a dirty app that gets them what they specifically want, instead of relying on some paid software to give everyone a little bit of what they want. Partially this suggests I'm just in the wrong sector, but it is absolutely happening.

I dont' think this matters to Google or Amazon, they can't be replaced. But small businesses are a different story.

And the result of all this? We need to heavily rely on AI, so that we can outpace individual users in delivering what they want. I hate it, I didn't give the order, but I do see the writing on the wall. This workflow is miserable, it sucks the fun out of the job, but unfortunately it really is faster. And small businesses rely on the income coming in next year, not in 5 years.

As a side note, I also think users are becoming extremely used to having a chatbot do everything for them. Every site is going to have one, and apps that don't will fall behind.

I'd like to be on a different multiverse timeline honestly

Does anyone truly understand all the little edge cases with CSS?

I've write tons and tons of CSS, have done for a decade. I don't sit and think about the exact interactions, I just know a couple things that might work if I'm getting something unexpected.

I don't really see it possible to commit that to memory, unless I literally start working on an interpreter myself.

EDIT: Of course people are already doing this. D2NN seems to be the keyword, Diffractive Deep Neural Networks

I suppose if your lens is actually a vertical stack of lenses, each with hundreds of inputs and outputs, then why wouldn't this work? Although I cannot fathom fabricating it. Maybe start by finding the absolute simplest/smallest image classifier you can

Yeah I'm confused as to what its actually doing

Cognito feels like its bones are grinding together

Most indie devs don't have time and money to optimize. They will make the game primarily for the biggest audience, and then make it somewhat playable for everyone else.

The closer Switch is to the Steam Deck, the more likely both will be targeted.

Yeah my experience has been thats its basically unplayable. I'm the kind of person who refunds when a game is <60fps though.

I just hope its powerful enough that Indies can target it along with the Steam Deck, rather than just hope an pray like they did for Switch 1's late lifecycle. The amount of <30fps indie titles on there was sad.

I get sleep paralysis (less so these days) and I absolutely get the instant REM thing. I'm dreaming before I'm asleep every single night, and I'm often dreaming for 5-10 minutes after I wake up. Its just a stream of audiovisual nonsense that doesn't shut off until I'm properly awake. Always figured it was normal.

And then imagine how each separate converstation can affect another. What a tangled mess this would be to work on.

I follow some mountain bike youtube channels, not professional riders but content creators, and literally every one of them has a yearly major injury. Broken bones are just normal.

118$ for a sweatshirt. Guess those rights were expensive.

Its so dependent on font size (or more accurately PPI) that its hard to pick. On my current monitor my favorite Berkely Mono looks thin and hard to read unless I bump up the size higher than I'd like. But drag it over to a Retina screen and it looks fantastic.

Anyone know of a tool like zizmor for GitLab CI/CD? Pretty confident my setup is unsafe after reading through this.

Honestly safety in CI/CD seems near impossible anyways.

Oh boy something new to try in KSP. Funny I've never seen anyone do this, maybe the imperfect simulation means its never more efficient (in-game).

The last few intel generations were soldiered too. I'm still using one.

I've held onto a 2011 car waiting for this swing. Maybe in 2026 I can get a modern car with genuine tactile controls

I'm not convinced this is a good idea either, but abstractions and syntax are what makes people love or hate a language. I enjoy writing python over JS because of ergonomics. Its not really about productivity, its about being happy when writing and reading code.

That being said, stapling on a compile step maybe isn't the answer.

This isn't a billion dollar company trading on the NYSE. Its a free website to play chess.

Asking for advice: I do not have a multiplayer app, but I have some large, interconnected, denormalized trees on my frontend as user profiles. Think like a tiled layout, where a user can add/remove/resize tiles, and then add a number of components into each tiled slot, each of those having their own profiles too. Multiple "layouts" can exist with different arrangements of tiles, and theres some other complexity with individual tiles referencing and sharing other pieces of state globally.

Making safe updates via regular REST is difficult, as you have to make sure someone with two tabs open isn't going to make an update on tab 1, then another on tab 2 which puts the overall profile into an invalid state. And in general, ordering matters. Skipping an update serverside that was properly applied clientside could break things.

The dumb-as-rocks solution I came up with is to just send the minimal amount of data over that can completely overwrite a particular chunk of state, and place it behind a queue. Usually thats fine, but sometimes thats a lot of wasted data, 50KB when the actual change was only a couple bytes.

I don't need CRDTs for any of the regular reasons, but it seems like it would make state management a million times easier, even for a single user. For one, I'd get syncing between a user's browser tabs, which is good. But moreover, I can just make simple changes to frontend state, and trust that the CRDT is going to negotiate it properly with the server. I no longer have to deal with it myself.

Does this make sense? Or is the overhead required to make something like Yjs work not worth it when I don't even need multiplayer and local-first.

The best part of any new stack: Spending 4 days repeatedly destroying and creating it while you figure out a single god damn policy to add to the project at a time

This is a wild take. Software developers do the dirty work. We're one step below wall street.

To note, we migrated from Redshfit, which had 7-30 second performance. Our current managed solution is something like 1 - 5. Duckdb just smashes everything else, at least on our data.

I very, very nearly migrated to a full Duckdb solution for customer-facing historical stock data. It would have been magical, and ridiculously, absurdly, ungodly fast. But the cloud costs ended up being close to a managed analytics solution, with significantly more moving parts (on our end). But I think thats just our use case, going forward I'd look at duckdb as an option for any large-scale datasets.

Using ECS/EKS containers reading from a segmented dataset in EFS is a really solid solution, you can get sub second performance over 6 billion rows / 10000 columns with proper management and reasonably restrictive queries.

Another option is to just deploy a couple huge EC2 instances that can fully fit the dataset. Costs here were about the same, but with a little more pain in server management. But the speed man, its just unbelievable.

Oroboros eating its tail

Another idea for something in this vein is caching. For small jobs, S3 and Dynamo are my typical go-to, and while they're both "simple" it still takes some effort to get all the code in place, more effort than you usually want to dedicate to a simple cache. The most frustrating thing that can happen is an endpoint or lambda goes down because you made a mistake in the damn cache layer, I'm a big believer in the cache having near-zero mental overhead.

Something to simplify a cached piece of data into 1 or 2 lines of code would be nice.

Yeah I completely missed this as a security flaw. Time to go deploy a fix...

I was immediately thinking of ADOM and NetHack. Would have been so cool.

Can someone give a quick description of why this one is special? I'm familiar with turing machines (sort of), but I can't deduce why this specific instruction set is impressive. Whats an Ackerman-level function? Whats it actually computing?