• 0 Posts
  • 120 Comments
Joined 1 year ago
cake
Cake day: October 30th, 2023

help-circle

  • Why don’t these kinda articles include the thing being talked about? I don’t want to know what an American thinks about denouncing Nazis; I want to know what was so egregious in the original resolution that America felt it was better to side with the Nazis.

    Edit: Here’s the damn resolution. there’s no valid reason to oppose it. First of all it’s not legally binding, second it’s pretty much just 75ish different ways of saying “member states should oppose Nazis cause Nazis are bad, 'm-kay.” Honestly just seems like the US opposed it cause it was introduced by Russia and was thus “political.”

    We condemn without reservation all forms of religious and ethnic intolerance or hatred at home and around the world Deputy U.S. Representative to the Economic and Social Council, Stefanie Amadeo

    Obviously not given that you just voted against a resolution that did exactly that.




  • There are local servers all around the world. You generally interact with the one closest to you, but it isn’t the “single source of truth.” At some point changes should be reconciled with the source authority, but all the remote servers are trying to update the source at pretty much the same time. Resolving updates from your global diaspora of servers in an accurate manner is hard and you’re likely to see odd bugs arise, especially when a large number of people take similar actions at once. You can see this in trending viral video likes–toggling your like on and off or viewing the video at the same time from 2 different locations is likely to give you a different like count each time. You can also see this when Trump rebrands the official POTUS account and starts spewing his vitriol through it, large numbers of people will realize they don’t want to deal with his bull shit, and some of their updates will fail to resolve in a timely manner.

    This doesn’t happen to everyone, but if you unfollow someone and it works you aren’t gonna complain.








  • To me this smells of typical subsidizing of a product to capture market share then lock in that market share. Anything I’m missing?

    That’s exactly it.

    From their email:

    What you get:

    2,000 code suggestions a month: Get context-aware suggestions tailored to your VS Code workspace and GitHub projects.

    50 Copilot Chat messages a month: Use Copilot Chat in VS Code and on GitHub to ask questions and refactor, debug, document, and explain code.

    Choose your AI model: You can select between Anthropic’s Claude 3.5 Sonnet or OpenAI’s GPT 4o.

    Render edits across multiple files: Use Copilot Edits to make changes to multiple files you’re working with.

    Access the Copilot Extensions ecosystem: Use third-party agents to conduct web searches via Perplexity, access information from Stack Overflow, and more.

    So it’s just a rate limited thing meant to get you signed up and then cut you off right when you get used to it. I get access through work and well, it just sucks.


  • Different operating systems have their own interfaces to allow user level programs (like games) to communicate with hardware. This is a great-over-simplification, but one OS may understand something like “drawTriangle(x, y, z)” while another may expect “drawPolygon([x, y, z])”.

    There are software projects to attempt to translate commands meant for one OS for a different OS (such as “Wine” or Valve’s “Proton”) and those work fairly well in cases that: 1) there’s an analogous command, 2) the analogous commands have been accurately mapped, and 3) the analogous commands operate in user space.

    That last point is the primary reason why, despite the best efforts of developers, some games still cannot work across OSs. Operating systems are built on top of different levels with the lowest being the “kernel” (of “kernel level anti-cheat” notoriety) and the highest being the user space (where you interact). Both Windows and Linux have these, but the boundaries around them, what they can and cannot do, and how to interact across those boundaries differs between each system.

    So when a Windows game installs a driver to monitor everything that your computer does that driver (kernel level anti-cheat) is tailored very specifically to the extremely powerful, low level, and unique Windows kernel. Linux cannot run that natively. If the game pretends that spying on you is an essential component to launch then the game will not launch. If, however, a game is perfectly happy to just stay in user space where it belongs then it will probably work fine with the available translation layers.






  • It depends on your jurisdiction, distribution platform, and the AI you use.

    In the US and distributing on Steam you can use AI assets, but you do need to indicate your use of AI.

    Keep a couple of things in mind though:

    • Your future customers may not appreciate that you used AI and you will get downvoted reviews solely because you opted for AI
    • You’ll likely have a better result making as much as you can yourself and using AI to fill in the gaps (i.e. if you have an interior scene, make the models and lay stuff out and maybe only use AI for a picture hanging on the wall or a coach fabric material–things that are inconsequential to the actual game play).

    Overall, I’d say if the supplemental use of AI during development is the difference between you finishing and publishing a game and you not, then sure–go ahead and use it, but make clear to prospective buyers how it was used and be prepared for blowback.