The folks at DRITTE have released BitMate, a low-bandwidth optimised BitTorent client.

From Dritte – BitMate – BitTorrent for the Less Privileged:

BitTorrent is a hugely popular peer-to-peer file sharing system. In countries where broadband Internet is widespread, BitTorrent accounts for as much as 70% of the overall Internet traffic. In contrast, in developing countries, BitTorrent is almost unusable on the typically low bandwidth dialup connections and accounts for less than 10% of the overall traffic. BitMate is designed to enhance the performance of hosts with low-bandwidth connections. [...]

BitMate outperforms vanilla BitTorrent by as much as 70% in download performance, while at the same time improving upload contribution by as much as 1000%! BitMate also outperforms strategic clients like BitTyrant in low-bandwidth conditions by as much as 60% in download performance (without cheating).

In summary:

• In countries with broadband, 70% of internet traffic is BitTorrent. In countries with no broadband it’s < 10%
• BitMate is free, open source BitTorrent client optimised for users on low bandwidths. Source code here.
• BitMate is up to 70% faster than regular BitTorrent for downloading on low bandwidths
• It can also improve upload contribution by up to 1000% (!)
• There are plans for a lightweight “BitMate 2″ that’s written in Python and is a few hundred KBs big.

This is excellent news – I’d encourage anybody interested to read the information on Dritte’s site.

Low Bandwidth Open Educational Resources

After reading the above, I’m reminded of this discussion on the UNESCO Acess2OER wiki. In particular the scenario:

Suppose you’re in Zambia, and you use an application like Miro to subscribe to a feed, say the Participatory Culture Foundation (PCF) podcast. Normally, the connection would be made straight to the PCF server, and would put immediate strain on your network, stopping others from browsing the web, or doing email. However, with our new and improved download system “super miro”, the subscription doesn’t go straight to the PCF server, but it goes to a local server at the school, then via a national Zambian school gateway (run by the NREN, providing an internet exchange point for Zambian schools and Universities), and only then goes to the PCF server.

What’s the big idea?

Open educational resources (OER) are “digitised materials offered freely and openly for educators, students and self-learners to use and reuse for teaching, learning and research” – think MIT’s Open Courseware or the Khan Academy.

Good bandwidth is key to delivering these resources. Content is typically in the form of audio, video and PDF files so someone on a slow internet connection in a typical developing world higher education institution won’t have a good experience, if any at all.

The idea on the UNESCO wiki is to improve content delivery to users on a slow network by offering an application (for now called “Super Miro” and based on Miro) that manages feeds of content by intelligently acquiring it in the most bandwidth efficient manner.

To put it into pictures,  this is how OER content is normally distributed:

Diagram 1: Traditional method of OER content delivery

The idea on the UNESO wiki looks something like this:

Diagram 2: Deliver OER by local cache/transcoding + bittorrent

What’s the advantage of this approach?

In principle, diagram 2 has the following advantages:

1. Original content is still managed by the publishing organisations and end users can still browse through it there.
2. A cache can be maintained in-country on national ISPs network and primed by:
   • Batch downloads
   • 6-monthly hard drive deliveries
   • Transcoding versions for different bandwidths.
3. Institutional bandwidth can be used very efficiently if users on the (local) network can share content amongst themselves.
4. Low bandwidth optimised bittorent can be used at almost every stage to make this architecture work even better.

Having said that, even adding low bandwidth bittorrent to diagram 1 would be an improvement.

Anyway, I’ll stop rambling. I don’t know what became of the above – perhaps now is a good time to dust it off. Jon Thompson over at Aid Worker Daily also suggests BitMate would also be useful for map sharing in a crisis.

{ 0 comments }