Well google seems to work out well enuff for people to find stuff
You'd be surprised. I can't count the number of times someone complains
"I can't find X on google" and I get it on the first page. And those
are simple, unstructured queries.
No I wouldn't be surprised at all ;)
The namesys.com doc that John pointed to gives a query example like :
type=email from=me to=him contains=[this that other]
When the "file" is saved to the "filesystem" the FS automatically
generates this information in the DB.
You might get some false positives, but nother more than with find
over a if you use maildir format
And for data that doesn't have as simple a format as email? Say a
random word (or even ooo) document? How about a jpeg? Requiring the
user to write some sort of tag or description isn't legal for this
challenge- if they can't remember a filename they won't remember the
tags. Plus it would make things worlds harder for machine generated files.
You give "properties" to a JPEG when you store it in a heirarchy.
The filename at the very least, it's path as well maybe
On top of this- you say this is no worse than maildir, but its no
better either. So the system is no better, more complicated, and really
doesn't fit many types of data. I don't see an advantage in it.
That may be true, I suppose we'll have to see what they come out with...
--
Michael O'Keefe | [EMAIL PROTECTED]
Live on and Ride a 03 BMW F650GSDakar| [EMAIL PROTECTED] / |
I like less more or less less than |Work:+1 858 845 3514 / |
more. UNIX-live it,love it,fork() it |Fax :+1 858 845 2652 /_p_|
My views are MINE ALONE, blah, blah, |Home:+1 760 788 1296 \`O'|
blah, yackety yack - don't come back |Fax :+1 858 _/_\|_,
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
