Friday, September 05, 2003
( 3:04 PM ) Matt
A new file search engine
For a long time I've wanted to make a query based file system. It would be ideal if metadata extracted from the files could be used to generate searchable attributes. Someone at Stanford has come up with a project along these lines. He calls it Storage. It looks like a good start in what I consider to be the right direction. It's too bad that it is really nothing more than a spiffy file system search frontend. Check it out.
Eventually, I'd really like to see the whole notion of files as descrete objects go away. The boundaries between one file and the next should be more fuzzy. This might be especially useful for revision control. Say you are editing a document, but it's a big document, and you want to have a bunch of people read different sections of it and then integrate their comments. You don't want to have to merge n copies of the document back together with their changes. So, you could just start out by using seperate files for each chapter of your work, but maybe it's hard to decide that when you set out, and it's a pain to do it after you've written a bunch. No more files, I say. Just data.# -
Comments: Post a Comment