I was reading a verge article [1], and this stood out to me.<p>> <i>But in STEM fields, directory structure remains crucially important. Astronomers, for example, may work with hundreds of thousands of files in the same format — which can be unwieldy to scale to a searchable system, Plavchan says.</i><p>Anecdotally, I've found searchable systems to be lukewarm. Search engines like Google are good for out-of-the-blue questions, but when looking within local files or data inside programs, search systems struggle to get what I need quickly. If I remember a specific file name, Spotlight search on macOS usually tells me what I need, while the default search on Windows does more or less the same. But these search systems never really help me find things in-depth, perhaps some mathematical formula embedded somewhere in a PDF file compiled from LaTeX. Or these search systems are not as forgiving when I'm trying to look for something that isn't text-based, like pictures or 3D assets.<p>Search is complex, so what can I do to depend less on these built-in systems? How can I make things more searchable in general? What methodological approach can be adopted to make searching for something, regardless of context, better?<p>Mindful folder structures, tagging files, consistent naming, and associating content to regular text as much as I can is a start, but I wonder if there's more that I'm failing to do.<p>[1] https://www.theverge.com/22684730/students-file-folder-directory-structure-education-gen-z