I am archiving older pieces I have written on other sites, making this the definitive home for all my work. This is one of several I am porting over from my GameDev.Net user journal. Enjoy!
I've long contemplated satisfying my technolust and industrial design envy and plunking down the cold, hard cash for an Apple Macintosh personal computer. As I've used the machines (in school) and learned more about them, I've grown fonder and fonder of them and the way things are quite intelligently laid out and implemented. I've spoken about their approach to MDI previously, and I may have mentioned that the OS X memory manager and process scheduler is, as far as user experience is concerned, far superior to any other desktop operating system out there.
But it was reading about Spotlight just a few minutes ago that sealed the deal. See, Spotlight is a good 65% of what I had hoped to accomplish with ReComputing: as a pervasive, extensible indexing technology with an expressive front end (from what I've read), it makes the distinctions of where documents are stored a non-issue. With Smart Folders, it allows you to think in terms of logical collections (related materials) rather than physical ones (directories), although you have to manually have create them.
"Big deal," you might say. "It's only been announced, and you haven't even tried it out yet!" True. However, have you ever known Apple to put up a product announcement/preview on their site and then not deliver? Apple already has a cult following; it doesn't need to put hype on its site as the cultmembers all trek to Cupertino, CA to hang on every word Jobs speaks each year at MacWorld.
Now a lot of this is similar to what was originally announced/planned for Windows Longhorn, with WinFS supposed to be built on Yukon and NTFS, but it was all very confusing, and Longhorn isn't expected until 2006, and WinFS is not expected to debut in the 2006 release anyway. Spotlight is not as completely pervasive as my suggestion, which did away with notions like having to save files (you could identify a file or group of files if you wanted to, but it really wasn't necessary), directories and what not. I still think there's a good argument to be made for such a system, and I still think that a Windows NT 5+ kernel makes for the best underpinnings for such a system.
Apple seems to be taking hold of the industrial design, ergonomics and usability gauntlet(s). I wonder if anyone will take that as a challenge?
Yes, yes, Microsoft owns part of Apple, blah blah blah...