Tuesday, May 31, 2005


Beagle search software programmer heads to Google | News.blog | CNET News.com: "Jon Trowbridge, the Novell programmer who led the Beagle desktop search tool Linux, has left to become an open-source developer at Google. Trowbridge announced the move in December 2006 on his blog, saying

Novell's Joe Shaw (one of the Ximian monkey) has taken over as maintainer of the Beagle project

listen to Joe (juli 2006)

Beagle is a search tool that ransacks your personal information space to find whatever you're looking for. Beagle can search in many different domains.

Helpful links
Why "Beagle"?

As local storages (hard disks) becoming bigger and faster, users end
up storing enormous amount of data. Local storages are mostly used to
store personal information: work documents, personal contacts, mails,
chat-logs etc. There are tools to search each of these entities
independantly, however, the tools either lack better usability or
requires some sort of expertise to customize them. In a way, all
these entities are related to each other. For example, a mail could
have come from a person in your contact-list with whom you might have
had a chat couple of days back and would have sent some documents as

These entities belong to a single group of information called
"Personal data space". Users have always been in a constant look-out
for a unified tool to search his "Personal data space".

What made it difficult in developing such a tool?

* The system should be able to extract and index the content of
various entities and independent of the format.

* The system should be able to search such contents efficiently.

* The system should be able to update itself on changes to these entities.

* The system should be capable enough to do a meta-data oriented search.

What is "Beagle"?

"Beagle" is a framework that can provide a single, unified way to
index and search your "personal data".

Beagle, unlike the existing search tools, does a "meta-data" centric
search. When we say "meta-data", it refers to useful information
about the entity which may not be part of the content but can be used
to identify the entity.

For example, a file "xyz.tgz" is being downloaded from
sourceforge.net. Once the download is over, it will be stored in the
FileSystem as xyz.tgz and no other information about that will be
available. Imagine a situation wherein we can attach some "useful"
information to the file, say, "downloaded from sf.net" and a search on
"downloaded from sf.net" fetches these file. How cool will it be???

This sort of "useful" information is called a "meta-data" and Beagle
is such a tool that can fetch you "details" on basis of such

Not, just the FileSystem, it can search your mails, contact-list, chat
logs, web blogs and contents of various work/personal documents
independant of formats. Given a search "phrase"/"word", it can list
all of the matching items in a user-friendly and relevant order.

Testing Beagle

in user mode: (not su)
beagled --replace --debug --fg

if the client is not running start
best --no-tray --show-window

add other directories

We are pleased to announce the release of Beagle 0.1.3. Among other things, this release contains Python bindings to libbeagle (courtesy of Raphael Slinckx), so you can now easily search from any Python program. So what are you waiting for? Go write some code!

No comments: