The New Windows And Organising Your Stuff

This month’s PCWorld gets hold of an early prototype of the next Windows, which, apart from the usual ‘interface enhancements’ illustrates what I think is going to be the most important change in how we store and retrieve files.

The magazine says that the new ‘Longhorn’ version Windows Explorer — the program which lists what files you have, and which folder there are in — “routinely displayed much more information about files and computer resources than it does in Windows XP”. There’s a panel in the program that “let users and/or applications associate search keywords, comments, and categories with files, data within files, or objects stored on other devices, computers, or networks.”

This basically means that, instead of lumping all your files in a specific directory, or folder, where they languish, you can give your files dynamic order depending on what keywords you assign them. Say you assign the keywords ‘home’ and ‘flubber’ to a file: you can then create ‘virtual folders’ using either, or both, those keywords which will turn up all files of whatever kind which contain those same keywords. This is called WinFS and in theory will allow you to find related resources regardless of their physical location or object type. If you’re interested, there’s more here on the Microsoft website.

I think this is a great innovation and one that is long overdue. The whole folder metaphor is tired and irrelevant to how we use data these days. However I have some worries: Given that most folk today still give their files less than helpful file names, and have yet to discover the joys of creating subfolders to give order to their hard drive, isn’t the ‘dynamic approach’ going to just make things messier? It will largely hinge, as far as I can see, on folk spending an extra few minutes entering keywords into each file’s properties box. Given we’re able to do that now in programs like Microsoft Word, but rarely actually do, what are the chances of that happening? Great in theory, I just worry about the implementation.

In the meantime, I use dtSearch to find stuff, and it works like a charm. It ain’t pretty but it’s sturdy and very configurable. Otherwise, check out X1, which is on the cusp of releasing a new version. Other good search programs: 80-20 Retriever and Enfish Find. Personally I couldn’t live without one of ’em.