Re: what needs to be done for beagle network

["Adam Lofts" <adam lofts gmail com>]

Hi!

On 4/21/06, D Bera <dbera web gmail com > wrote:

2) Many (1)s connected in a distributed way: Similar to the
beagle-network which was implemented and then removed to give way to
webservices. Here one beagle daemon runs on each client and server.
And beagle daemons can query each other. Clients beagle-search only
queries to the local daemon which in turn distributes the query to
other daemons.

  Comments: Different ball game altogether. Implementing a
P2P/distributed system is always tricky and difficult. There are many
issues, authentication, network outages, post-processing results to
remove duplicates, control the amount of flooding, handle live query.

I think this is the most exciting solution. There are clear difficulties in implementation but i'm sure that libraries have been written to handle p2p stuff easily. Also it doesn't require proper p2p, simply the query client connecting to all demons it it aware of would acheive a similar result.

It seems easier to say "A demon handles the stuff close [1] to it, everything else must query the demon" that to have stuff like the demon indexing remote mounted partitions etc. To solve the uri problem the each demon would need to provide a method to locate a query result remotely, maybe just a uri like file://192.168.0.*/path/to/my/file and then beagle client could look for ways to get at it.

One cool immediate solution would be to have iTunes style sharing. In the configuration box have a "Allow searching on the local network [x]" which will turn on an Avahi service thing. Then beagle-search will monitor the LAN for demons appearing and automatically search on them! What a cool SoC project that would be :)

[1] The definition of close remains a problem :)

Adam