I recently read an InfoWorld article entitled Gartner: Ignore Longhorn and stick with XP where it states

Microsoft Corp. may choose never to release its vaunted and long-overdue project WinFS, following its removal from the next version of Windows, according to analysts Gartner Inc.
...
Microsoft has said Longhorn will still include local desktop searching as a taste of the power of WinFS' relational database capabilities, but Gartner sees this as a hint that WinFS may never arrive. "Because Microsoft has committed to improving search without WinFS, it may choose never to deliver the delayed WinFS," Gartner said.

The fundamental premise of the above statements is that the purpose of WinFS is to make local desktop search better or to use a cruder term to create "Google for the Desktop". It may be true that when it first started getting pitched one of the scenarios people described was making search better. However as WinFS progressed the primary scenarios its designers focused on enabling didn't have much to do with search. If you read Longhorn evangelist Jeremy Mazner's blog posting entitled What happened to WinFS? posted after the Longhorn changes were announced you'll find the following excerpt

The WinFS team spent a solid couple weeks going through this evaluation.  There are of course plenty of things you could do to increase the confidence level on a project the size of WinFS, since it has so many features, including:

• Built-in schemas for calendar, contacts, documents, media, etc
• Extensibility for adding custom schema or business logic
• File system integration, like promotion/demotion and valid win32 file paths
• A synchronization runtime for keeping content up to date
• Rich API support for app scenarios like grouping and filtering
• A self-tuning management service to keep the system running well
• Tools for deploying schema, data and applications

The above feature list is missing the recent decision to incorporate the features of the object relational mapping technology codenamed ObjectSpaces into WinFS. Taking all these features together none of them is really focused on making it easier for me to find things on my desktop.

At its core, WinFS was about storing strongly typed objects in the file system instead of opaque blobs of bits. The purpose of doing this was to make accessing and manipulating the content and metadata of these files simpler and more consistent. For example, instead of having to know how to manipulate JPEG, TIFF, GIF and BMP files there would just be a Photo item type that applications would have to deal with. Similarly one could imagine just interacting with a built in Music item instead of programming against MP3, WMA, OGG, AAC, and WAV files. In talking to Mike Deem a few months ago and recently seeing Bill Gates discuss his vision for WinFS to folks in our building a few weeks ago it is clear to me that the major benefits of WinFS to end users is the possibilities it creates in user interfaces for data organization.

Recently I switched from using WinAmp to iTunes on the strength of the music organizational capabilities of the iTunes library and "smart playlists". The strength of iTunes is that it provides a consistent interface to interacting with music files regardless of their underlying type (AAC, MP3, etc) and provides ways to add metadata about these music files (ratings, number of times played) then organize these files according to this metadata. Another application that shows the power of the data organization based on rich, structured metadata is Search Folders in Outlook 2003. When I used to think of WinFS I got excited about being able to perform SQL-like queries over items in the file system. Then I heard Bill Gates and Mike Deem speak about WinFS then saw them getting excited about the ability to take the data organizational capabilities of features like the My Pictures and My Music folders in Windows to the next level it all clicked.

Now this isn't to say that there aren't some searches made better by coming up with a consistent way to interact with certain file types and providing structured metadata about these files. For example a search like

Get me all the songs [regardless of file type] either featuring or created by G-Unit or any of its members (Young Buck, 50 Cent, Tony Yayo or Lloyd Banks) between 2002 and 2004 on my hard drive

is made possible with this system. However it is more likely that I want to navigate this in a UI like the iTunes media library than I want to type the equivalent of SQL queries over my file system.

More importantly, this system doesn't make much easier to find stuff I've lost on my file system like Java code I wrote while in college or drafts of articles created several years ago that I never finished. When I think "Google on the Desktop", that's the problem I want to see solved. However MSN just bought LookOut so I have faith that we will be solving this problem in the near future as well.