What is the best folder hierarchy/structure for a new hard drive?

I have tried to organize my files over decades, but there is no easy way to maintain a hierarchy. I still have a "Misc" and "Unorganized" folder with massive amounts of documents The best thing to do is : 1) Use http://voidtools.com everything search tool - It instantly finds files of a given name or wildcard, does not consume any CPU or disk space - 90% of the time you can find files based on the names you remember. 2) Use windows search and index file content - this takes up space and time, but very easy to find files with specific content. 3) Delete stuff - This is no longer the 2000s, no need to keep old setups and stuff - you can always download it again. Old documents etc are also quite useless, you never realize until you lose data and then you understand you really never needed that or miss it.

See Everything (voidtools) doesn't run on Mac OS or Windows 8 unfortunately.

This is very much a reflection of who you are.  While modern search tools are amazing, good organization also helps.    Think about how you use the data.  Are you the kind of person who is more task-oriented?  Then file things by the activity they relate to.  Do you think time and history is more important?  Organize by date.  And so on.

There is no best, which is why there's no default layout. How you arrange is dependent on how you tend to work with it and may change for different scenarios. That's why you have things like Smart Folders in OS X. If you are not familiar with OS X, the operating system has a service called 'mds' which maintains a metadata database of files on the disk. The data contains basic information about the file: name, size, creation date, last open date, file type. It also stores file type specific for each type of file: bitrate, camera model, tempo, url it was downloaded from, copyright, altitude -- whatever is pertinent to the specific type of data, including text contents. It's very fast and very complete (despite most people just accessing it through Spotlight and making basic searches on name or content). The search is saved and appears as a folder, but the contents of the folder are the result of the search. I might want a view of all my pictures that are associated with a specific country, but I may have pictures stored on multiple disks or in multiple directories, so I create a smart folder that   contains Kind is Image AND Country IS Ireland. The view is autoupdated as I store files anywhere on my attached disks, it's available in file dialogs, etc. Windows 7 (and I presume 8) has a similar, if considerably less convenient analog where you can use Windows Search to specify criteria to select a group of files, then you can turn those files into a Library (a folder full of file links). This differs slightly from OS X in that the metadata extraction is not as detailed,  the process takes more steps, and it doesn't dynamically update, but it's similar in spirit. As early as mid 1990's, Microsoft was considering replacing it's filesystem with a database (essentially BLOBs in an MS SQL database) to address this problem. For a slew of reasons, this never came to fruition, but it's development was to implement a system that works like Apple's mds service, but also storing the file itself in the database. There are analogous features to Windows Search available for Linux desktop environments as well (though less convenient). Generally speaking, the industry is moving to a model where the organization of information is more dynamic. One could layout their directories in any way they like, but access to the files is brokered by searches that present 'views' of sets of files that are not constrained by the hierarchy.