|
by
Michelle Delio
The
Gnutella Next Generation development team announced on Friday that
they are developing a new open source technology for search engines.
The
group believes that "gPulp" (general Purpose Location
Protocol) will eventually become the standard search tool on every
network and computing device. "GPulp will be a ubiquitous, open,
free, and powerful tool that lets users find anything - anything! --
on any network," promised Gnutella Next Generation (Gnutella NG)
team manager Sebastien Lambla.
GPulp
is different from existing search technologies in a number of ways.
It is
based on the Gnutella structure, an open source application
originally created by Nullsoft. Gnutella enables users to exchange
any type of file without having to go through a central server, as
users must do with most of the 'sharing' programs, like Napster, that
are currently in use.
The
first generation of Gnutella
Using
the basic protocols originally developed for Gnutella, gPulp will
search for information across a network in real time. The returned
search results will be based on current information, not on records
stored in a database of indexed Web pages.
And
since gPulp will search an entire network, any and all data that is
available on that network can be located through a single interface.
Currently search engines return only the data stored on their own
servers, so searchers often need to query several search engines
before they find the information they are hunting for.
Lambla
offers this example. "Say that you are looking for someone's
telephone number. To do this you would need to access a directory
server, a LDAP server. The problem is, there are several LDAP servers
on the Internet, so you have to find each of them, and query them for
the phone number."
"With
gPulp, each LDAP server would be searched and you would find out
which one has the number that you are looking for. Then, you can look
on that specific LDAP server, which can also be accessed from the
gPulp network, to get the phone number."
GPulp
will also add a powerful location capability to networks.
"The
biggest problem today is Domain Names," said Lambla. "There
is a limited pool of names that can be used, but there are more and
more devices connected to the Internet. Today it's the desktop
computer, a handheld device and a phone. Tomorrow it will be your
lamps, your bath, etc. All these devices will be connected to the
Internet, and gPulp will be able to locate each single device on the
network, without you having to give a Domain Name to each device."
Open
source program developer Chris Childress thinks that gPulp could be
the "über-app" that will completely change the Internet.
"It
used to be that the Web really was a web. You could literally click
your way around the world just following the links on people's pages,
and eventually you'd end up back where you started out. Now, the Web
is becoming segmented and static. But with a single and ubiquitous
search and location system that is totally open source and available
to everyone, the web can become dynamic again."
Of
course, gPulp won't achieve its goals if it isn't widely deployed
across the entire Internet. But Lambla doesn't think that will be a problem.
"More
and more, the big names of the computer industry, like Intel,
recognize that peer-to-peer technology has a huge potential, and that
it will change the landscape of the Internet industry. They will be
interested in gPulp."
One
other sticking point might be ease of use. Gnutella has the
reputation of being somewhat inaccessible to an inexperienced user.
But Lambla said this will not be an issue either, since gPulp will be
a framework over which developers will build their applications.
"The
'easiness' of a program will be up to the developer, and the
protocol will be completely hidden from the end user. The end user
won't see the underlying network. GPulp is a location protocol, so
the software will use it, not the users."
The
Gnutella NG group hopes to release a first working draft of the gPulp
code in early 2001, Lambla said, quickly adding that the group will
not commit to a firm deadline.
"We
won't hurry to release this protocol, because it's such a big step
in the computing industry. We need people from Intel, Cisco, IBM,
Sun, even Microsoft to join the gPulp effort so that everyone can
agree on this new search platform. And we need to rally all the
developers out there interested in Gnutella and in new technologies
in general.
"There's
quite a lot of work to do."
What do
you think? Post your comments on our Forums.
|
