• 13 Posts
  • 36 Comments
Joined 2 years ago
cake
Cake day: July 10th, 2023

help-circle

  • Maybe an outside one here, but when Dragon Age: Origins was still new, I was much younger and hadn’t had a serious girlfriend yet. And in general, I got quite a bit more attached to characters in games, including the romantic interests especially.

    Spoilers ahead if you haven’t beaten the game

    spoiler

    At the end of the game, my character died and I had romanced the character Leliana. The end screen shows a somber note about Leliana being heartbroken and wandering the world missing the character.

    Something about that really struck something in me and made me extremely sad. I think it’s probably the only time I’ve ever really cried because of a video game.

    I can’t actually find a reference to this ending online. The wiki isn’t 100% complete. Otherwise, I would have linked the blurb here.

    Now a days I struggle to connect with characters in games. They will never be as real as my friends or my partner, so it’s hard to form such an emotional connection as I did when I was a teenager.


  • I think the voting ratio pretty well demonstrates one obvious hurdle: people don’t like it.

    Even people who are happy with AI seem repulsed by the idea of it replacing human intimacy. So unless you have some real ride or die friend and family, you’ll have to risk judgment from them or keep it a secret

    There’s also the fact that right now most AI bots are toadies by nature - meaning they will rarely disagree with you and will generally capitulate to your will. That would be off-putting for some, including myself.

    They also (currently) don’t have great memory and will not always recall things that are important to you. No matter hpw any times you tell them.

    Obviously, there are larger issues to consider, but if you’re already considering the leap then you may not necessarily be convinced by arguments about humanity and what’s natural.




  • Research paper here: https://metr.org/Early_2025_AI_Experienced_OS_Devs_Study.pdf

    I think these are interesting findings, they definitely conform to my experience, but I’d want to always have more data and larger sample sizes.

    One point in particular stood out to me - the lack of LLM context. I have a new person on my team (junior level) that uses AI for everything, and it’s so obvious that the LLM is getting tripped up with the lack of context. Not all the required information is in a single repository, and it needs additional info like the documentation and spoken architectures that aren’t explicitly documented, historical choices as to why we make decisions in a certain way, or even just style guides. The LLM context window just isn’t large enough right now to be a truly effective programmer for large, complex projects



  • While insulation helps, it’s fundamentally less efficient because the heat sync is inside the building and insulation isn’t foolproof. And often these units are made of plastic, so they leak more heat into the house from the chassis than you’d expect. Even the best hose is less efficient than a window unit where the entire heat sync is outside the building

    So while $20 and some duct tape helps a great deal, it definitely won’t be as efficient as a window unit







  • My dad did so much right, but his one failing was financial. He was an insurance salesman and had plenty of money when I was very young, but at some point it all dried up and he seemed unable to make more. He didn’t starve or anything, but at a certain point my brother had to step in and buy his house or he was going to lose it.

    So now, I’m very cognizant of my spending and always having a good cash reserve.

    But, he was also extremely generous when he did have money. His favorite way to spend money was on the people he loved and to make them happy.

    So now, I also give freely. If it makes someone I love happy, and I can afford it, I’ll give them whatever I think might make them smile, if even for a day