Recoll: A Linux Desktop Search Engine That Works
- Feb 29, 2008
Most of us are to the point where we've got tens or hundreds of gigabytes of data on our hard drives and tens or hundreds of thousands of files. What good is all that information if you can't lay hands on what you want when you need it?
Anything is easy to find if you know the directory it's in and the filename, regardless of OS. But when you need 10-year-old information, it's unlikely that you'll know either. The fix is a desktop search engine. A good one will reduce to just minutes the time taken in those document searches that used to take hours.
I've found a desktop search engine that works, and its name is Recoll.
Recoll vs. Beagle
Although Beagle is the default search engine on several distros, including Novell SLED, Xandros, and Fedora, that doesn't make Beagle the best choice.
Beagle runs in continuous background reindexing mode, which is a great way to keep the hard drive light on all the time, translating to extra and unnecessary wear on the HD. While this is going on, all the other applications on your HD slow down. That reindexing program is a disk/CPU resource hog.
Beagle also works with a single search term/string, which makes it a very blunt instrument for searching through tens of gigabytes of files for anything in particular. Recoll, on the other hand, provides full Boolean search options on multiple terms.
And I found Beagle to be slow. It also went into the bit bucket within 10 minutes after I installed Recoll, and Recoll finished building the indexes so I could try it. Recoll is fast. Searches of my home filesystem rarely take more than a minute and usually take just seconds.