P2P Infrastructure for Social Networking and Web 2.0 applications
The landscape of internet usage has changed dramatically in the recent years, both in the way the computers connected to the network interact as well as the way the end-users using these computers interact - with the internet, and with each other.
On the networking plane, infused by the (somewhat infamous) success of P2P file-swapping softwares, last decade has witnessed an increased emphasis of using resources available at the edge to perform tasks which would otherwise have heavily burdened any centralized infrastructure. Thus to say, there is an increased proliferation of peer-to-peer mechanisms to either replace, or more often supplement, the client-server paradigm.
On the application plane, with the
advent of Web 2.0 and social networks, we witness
end-users participating not only as passive consumer of
content from content provided by the web-sites
(client-server), but rather that the content itself is
often contributed by the end-users themselves. Thus, at
a logical level, many of these Web 2.0 applications are
inherently peer-to-peer in nature.
Nevertheless, somewhat ironically,
all current Web 2.0 applications rely on an underlying
infrastructure based on the traditional client-server
model. When the user interactions are peer-to-peer in
nature, and while there is such a proliferation of
unrelated P2P systems and applications, it is but
natural to ask if and how to realize a peer-to-peer
underlying networking infrastructure for Web 2.0
applications?
Perhaps in this irony lies the opportunity for P2P to redeem itself. The almost only well understood popular P2P application (besides Skype) is file-sharing and like many other technologies, it is susceptible to unethical use. The technology of P2P file-sharing, and even more generally the whole P2P paradigm has often been demonized because of that. The baffling question has often been, what is the application? We believe that P2P infrastructure for Web 2.0 applications, and thus in turn the Web 2.0 applications themselves, is the answer. The match could not have been any better or more natural than when both the underlying network resources/infrastructure, as well as the content is provided and consumed by end-users. Specifically for social networking applications, a P2P infrastructure, while challenging, can protect the users from the whims of social networking sites which (can) trample user privacy among other things.
Some of the work here intersects with the mTeam project's research agenda.
Publications & manuscripts
We outline the motivation of a P2P infrastructure for social networking and some of the main technical challenges in realizing such an infrastructure in a manuscript.
Software
PBDMS: A very preliminary version of a social software for bibliographic content management (without almost any security features, yet) can be found here.
PeerSoN: Peer-to-Peer Social Network
For part of this work, I also collaborate with researchers from Deutsche Telekom Laboratories (Berlin, Germany) and EPFL Lausanne (Switzerland), including:
-
Sonja Buchegger (Senior research scientist, Deutsche Telekom Laboratories, Berlin)
-
Le-Hung Vu (PhD. student, EPFL Switzerland)
Students
-
Youssef Affify (Student, EPFL Switzerland)
-
Doris Reim (Masters student, TU Berlin & Deutsche Telekom Laboratories, Berlin)
Past members and contributors
-
Do Hoang Hai (NTU Singapore)
- Worked on an early implementation of PBDMS as part of his final year project (FYP 2007/8)