Ask HN：有哪些有趣的冷门爱好？
4 分•作者: e-topy•4 个月前
我正在寻找一些新颖有趣、人不太多的领域，能够让我做出有意义的贡献。
2022年，我曾尝试使用 OpenAI 的 RL Gym，当时第一个非指令型 GPT3 模型刚问世。我曾想更深入地研究机器学习，但犹豫了。
在那之前，我玩过3D打印机、机械键盘、无人机等等。
这些东西都火起来了，虽然它们仍然很有趣，而且我喜欢我的茶轴键盘，也负责管理本地黑客空间的 Prusa 打印机，但它们已经，用一个不太恰当的词来说，已经“工业化”了。
现在我也有时间和金钱来投入这些了，不像我15岁时，评估着 Ender 打印机的主板升级，但我知道我永远不会买。
目前，我正在制作一个国际象棋引擎，但这已经是一个被解决的问题了。
还有生物黑客技术，虽然设计芯片植入体内确实很有趣，但我只有一个身体，不想冒险尝试。
一个有前景的想法是“个人电脑2.0”，人们尝试创新人机交互（HCI），虽然我真的很喜欢这个想法，并且有一些研究想法，但我希望在深入研究之前能多探索一下。
查看原文
I'm looking for something novel and interesting, that isn't absolutely crowded that I could meaningfully contribute to.<p>In 2022 I was toying around with OpenAI's RL Gym, right when the first non-instruct GPT3 model came out. I was thinking about getting into ML a lot more, but hesitated.
Before that it was 3D printers, mechanical keyboards, drones, etc.
All of these have exploded, and while they are still very interesting, I do love my Browns and manage Prusas for my local hackerspace, they have just, for the lack of a better term, industrialized.
I'm also now in a position where I have time and money for it, not like when I was 15 and rating Ender motherboard upgrades I knew I'd never buy.<p>Right now I'm making a chess engine, but that's already a solved problem.
There's also biohacking, and while designing chips to go into my body is really interesting, I only have one, and don't want to push it too far.
One promising idea is a kind of 'Personal Computer 2', where people try to innovate HCI, and while I really like that and do have some research ideas, I'd like to explore a bit more before delving deep into it.