Friday, July 21, 2006

Help Me Test Kerjodando - A Simple P2P File Sharing and Social Netwrok Website

Hi this is the email that I sent to potential kerjodando testers:

Please help me to test my new software idea - kerjodando p2p and social network.


All I need is your ip address (preferably home one - as going to be used for p2p!).

This can be found here ( ) just reply to this email including your ip address.

ip address is four number separated by dots

e.g. is my ip address.

I will then send you a link to download a test file from.

The first stage of the test is to allow friends to connect (anonymously) to each other and download this file.

If you have any questions please do not hesitate to contact me.


You need to be at the PC where you are going to be testing when you find your ip address. So you may need to forward this email to your home email address.

Not a good idea to test at work as not fair and against normal work rules.

Can test in internet cafe.

Should get through every firewall and filter even "Great Firewall of China".

Blog about development at

Holding page for website at

We aim to be a worldwide social network and p2p site so all citizens can have their view on all events e.g. the world cup:

Zizou - What Did You See?


Friday, July 14, 2006

Alpha Testers Needed

The kerjodando project needs alpha / beta testers to test the latest changes to the client.

Please email me stating what type (nothing specific) of files you are interested in sharing / downloading e.g. MP3, adult, movies, books etc.

And your ip address.

I will then put you in a swarm (group) and send you the address of the swarm (a bit like a torrent) for you to click on.

kerjodando works like a friend to friend p2p file sharing application.

When the kerjodando website is developed this process will be automatic behind the scenes.

email Ezzy Elliott

Tuesday, July 11, 2006

Low Level Marketing

Trying to make at least one post to each of the p2p consortium websites to get some expert input into the development of kerjodando.

kerjodando p2p - Progress Report

  1. Concept - done
  2. Alpha client with core functionality - done
  3. kerjodando p2p holding page ( ) - done
  4. kerjodando development blog ( ) -done
  5. Working on beta client that can join a swarm specified in a url


  • Swarms of files and connections
  • Click to join a swarm
  • Private and secure (anonymous p2p)

Alpha Client

  • Modded latest version of ANts p2p so can join specified swarm
  • Script written to update kerjodando whenever ANts p2p updated

Saturday, July 01, 2006

Only Small World P2P Network Navigation (Search) Scalable

Only small world networks are scalable i.e. can grow in size without becoming very inefficient.

A small world network, where nodes represent people and edges connect people that know each other, captures the small world phenomenon of strangers being linked by a mutual acquaintance.

The small world phenomenon (also known as the small world effect) is the hypothesis that everyone in the world can be reached through a short chain of social acquaintances. The concept gave rise to the famous phrase six degrees of separation after a 1967 small world experiment by social psychologist Stanley Milgram which suggested that two random US citizens were connected on average by a chain of six acquaintances.

This has far reaching consequences for 'anonymous' P2P file sharing. A leading researcher on small world networks, Jon Kleinberg, states the following:

Moreover, α=2 is the only exponent at which any decentralized algorithm can achieve a delivery time bounded by any polynomial in logN: for every other exponent, an asymptotically much larger delivery time is required, regardless of the algorithm employed.

Jon Kleinberg

Lets take this phrase by phrase to see what does this actually mean for anonymous file sharing clients like ANts P2P, MUTE and Freenet.

  • α=2 is the only exponent at which any decentralized algorithm can achieve a delivery time bounded by any polynomial in logN.
  • logN refers to the width of a network.
  • As the number of users (N) increases, the average number of hops between any two users, the width, increases by a factor of logN.
  • However, it is expected delivery time (T) for messages sent between users that is important for file sharing.
  • T increases by a factor of logN only for small world networks (α=2). For other networks T is very much larger.
  • T for a ad-hoc overlay p2p net (like ANts, MUTE or Freenet) is the number of users a file or query has to be forwarded via to reach its destination, the path length (L).
  • For a small network the path length is small so that a workable amount of bandwidth in total is used in uploading and downloading for each hop along path length.
  • For a large network L is far too large unless the network is small world although ironically the width of the network still scales by logN.

  • for every other exponent, an asymptotically much larger delivery time is required, regardless of the algorithm employed.
  • So you can come up with whatever clever message routing method you like, it will not scale unless it is small world.
  • It is the network structure, the topology not the routing algorithm that determines if a network scales.
  • So neither ANts P2P or MUTE will scale by improving routing algorithms while they remain networks with a random topology(structure).
  • However, Iain Clark of Freenet seems to have already made this break through and has redesigned Freenet 0.7 to be small world.

I have decided to do the same for ANts p2p file sharing by making a very simple mod that will force the network to be small world.

kerjodando, the name of my ANts p2p mod, will force users to connect to only friends or others that share a similar interest to themselves, a p2p small world social network.

I call these groups, swarms and they are controlled, like bit torrent, by a simple tracker. Users can be members of many swarms covering all their interests and social networks. Swarms can either be private (invitation only) or public.

In this way kerjodando will be more scalable and be able to grow much bigger.