| I am kind of the opposite. I like retro computing but I never find myself using old software on old hardware. It is basically a museum piece sitting on a shelf at that point. If I actually want to use my hardware, I want new software on it. If I am going to create software for it, I want to use modern languages and modern language versions. I want modern media codecs. I want to be able to read modern media and use modern filesystems. I want Linux on it if I can, including the conveniences of modern package management. I may want to use containers. If it is going to connect to the Internet, I want it to be secure. I do not want to deal with pointless compatibility problems, limited aofrware selection, or decades old bugs. For me, part of the fun is seeing how much you can make old hardware do. And with new software, that can really surprise you sometimes. And in this age of the cloud, the hardware may not be the bottleneck you think it is. I will give two examples. First, I have used OpenCode to write native software on old platforms. Toy compilers can be fun. The locally running software is not that demanding with the hard work being done in the cloud by a frontier AI but with the results tested locally and natively on the old hardware. This is a lot more fun, and makes more sense, than doing it in an emulator. Second, the core of my homelab is Proxmox running on a 2013 Mac Pro. It runs dozens of VMs and containers. It cost me $200. Where else am I getting 64 GB and 12 cores for $200? |