Yeah, if it was realistic it wouldn’t be fun

It honestly (usually) does lead to a fantastic product! I maintain my own significantly used tools independently and completely agree. I also have seen locally (as a corporate pawn and life-long software engineer) what happens once somebody quits and no longer maintains their beautiful project(s).

You work so much FASTER alone than you do in a group. You also can NEVER get as far with your tool when you work alone. I think the best FOSS tools are born from independent savant developers, but for them to reliably be carried on, they have to be passed down to SOMEONE. It’s not your job to foster the entire next generation of tablet-children to be able to push golden commits to your curl 2.0 repo; it is pretty worthwhile to foster at least a handful of interested and headstrong people to understand your work in its entirety, and carry on its progress. And then they can do the same, and FOSS will live on forever (as it SHOULD be).

You probably spent an obscene amount of time developing an S+ tier piece of tooling, it would be pertinent to spend another marginal month or so to raise some lil star to be able to mimic your work once you tap out.

Having more free time is cool, but there’s more things in life. As MC Ride said: “LIKE GETTING YOUR DICK. RODE ALL FUCKING NIGHT.” Find some lil dick rider to carry on your shit.

if a maintainer doesn’t want your sticky grubby toddler hands fuckin in their cookie jar, it’s their fucking cookie jar and they can tell you to fuck off.

Yeah, that is exactly how it works. And doing that leads to your tool dying since you have no clue how to foster a community to take care of it.

Original creators and maintainers are hitting retirement age.

And not many good younger people are available to take the mantle.

This is the long-term cost of how persnickety FOSS maintainers are when it comes to accepting outside contributions to their work.

DeepSeek claimed the model training took 2,788 thousand H800 GPU hours, which, at a cost of $2/GPU hour, comes out to a mere $5.576 million.

That seems impossibly low.

DeepSeek is clear that these costs are only for the final training run, and exclude all other expenses

Took me like 60 seconds to get it. I play poker and rogue likes already though

I thought so too, and maybe they are using a different metric in this article, but I couldn’t tell you since their source URL is a deadline…

https://www.surfrider.org/programs/beach-cleanups

Just cuz Nintendo is a company run by bitch-made cum guzzlers doesn’t mean its fans should bend over like that. It used to be that kids loved Mario and sonic, so they made a shit ton of Mario and sonic flash games. That’s good, creative, harmless, and gets attention because of the brand fandom.

Work arounds are for devs to solve, not schmucks on lemmy. CS:GO used to use something similar to occlusion culling to prevent this exact problem. Don’t send the client all of the enemy locations/data unless the client is within roughly the right distance/sight to see that enemy. This is not a full fix, but dramatically nerfs wall hacking.

I’ve seen community plugins for TF2 and other source engine games that will add “ghost” players. Generate ai characters, turn them invisible, and send their data to clients. If someone keeps shooting at the ghosts, they can easily get caught and banned.

There are entire industries dedicated to finding solutions to this problem, check out this research paper about this exact subject if you want.

Okay, now the player data is encrypted and unreadable by clients.

How will the client display where the players are without data…

Why are you bothering to spend cycles sending this useless, encrypted data…

If you mean to decrypt the player data once it reaches the client, then you have solved no issues…

Brain rot

Yeah, so pick a cool one

https://github.com/pzoj/pzoj-contest/pull/1#issuecomment-2283067326

It doesn’t serve my IEMs

Homies throwing a tantrum like a three year old, what prompted this?

Neither are good, stop using discount brainrot shopping sites

I am a tech consumer and enthusiast first. I am a corporate shill sellout second. I wish for bad practices in the tech community to die, even if it’s my own company doing it.

My concern as an engineer is that the product gets made well. I have no say or control over how the business cretins and marketing scumbags decide to destroy the company through terrible unethical practices like charging SaaS for completely self-contained software.

The short term view is that you need to keep a company afloat. Businesses should fail if they deliver products in awful ways. Yes, if the company fails, I will lose my job, and that is okay. It would be through no fault of my own, or really even the customers who wouldn’t pay for my company’s product. It would be the fault of the business decisions that were made. And the product landscape would then open up after my company’s failure. For example, if Adobe would finally fucking die then we may actually see better products on the PDF, and photo/video editing market. No more monopoly on sub-par creative cloud products.

The more realistic long term view is that software engineers will be okay if their company fails. The overwhelming majority are smart, get paid extremely well, and exist in a field that needs their manpower. They will be able to find a new job much easier than other fields. The tech community will not be okay long-term if bad companies cannot fail.

I am a programmer, and I get paid whether or not the product is bought. Shovel your dogshit somewhere else.

For every hour of work/coding I do, there is probably 4 to 5 hours of waiting for shit to automatically compile, fetch, build, release, apply, get reviewed, approved, and deployed. The downtime is immense, I spend it helping other people with shit or planning company potlucks (I don’t work for Microsoft).

It might not be generative AI, but it is still primitive AI that drives the mapping of the physical streamer’s body to the animated cartoon body