On Thursday 18 January 2007 02:28, Jan Engelhardt wrote: > ... > > Over the years, people have managed to keep their files > together. UNIX: ~, as always. DOS: Dedicated folder on C:, > or another drive. Win 4.x-6.x: My Documents. Why would we > suddenly need search engines?
Because now we have multi-hundred-gigabyte mass storage and that was not true even five years ago. > People unable to label their > files properly (putting them into the right-named directory > as a bonus) are a lost case like their files. If you have a large library of publications, media, etc. then manual organization is impractical at best and impossible in general (owing to the fact that no single hierarchy adequately captures all required organizational schemes). File names are often poor indicators of what a file contains. For example, files retrieved from the ACM digital library have names that start with the primary author's surname followed by the starting page number of the article within the print publication in which it originally appeared. (With a ".pdf" suffix, of course.) Files retrieved from CiteSeer follow no naming convention at all and simply reflect what their author called them on the computer where the paper was prepared, and often the name does not hint at either the author or the subject of the paper. Content-based access is sorely needed in today's computers. > -`J' Randall Schulz -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
