Tavin Cole wrote:
> 
> Freenet doesn't have to do _everything_ ..  Let Freenet excel at being
> one thing: a decentralized, quasi-anonymous, efficient system for inserting
> and retrieving data by key.  Then create another system -- a decentralized,
> quasi-anonymous metadata search network.  Let each network become optimized
> to perform its specific purpose.


This was actually a method suggested in "Peer-to-Peer File Sharing and 
Copyright Law after
Napster" 
( http://www.eff.org/IP/P2P/Napster/20010227_P2P_Copyright_White_Paper.html ) 
to avoid
problems.

But this begs a second question.  Does the current infrastructure provided by
Freenet allow a flexible searching mechanism to be added with little effort?

If such a method can be added simply by taking advantage of the existing 
architecture
then it would actually be better to do both within freenet itself.

However, if the implementation of a flexible searching mechanism requires a
significant effort that utilizes little of what freenet already provides then
they should be separate.



> Incidentally, it would be possible with the 0.4 architecture and perhaps
> some modifications to support running distinct nodes of distinct P2P
> networks within the same Fred.
> 


True, and if these two peices of functionality are to be implemented separately,
it would definately be user friendly to bundle the two together.

Best regards,
    Martin Peck.

_______________________________________________
Devl mailing list
Devl at freenetproject.org
http://lists.freenetproject.org/mailman/listinfo/devl

Reply via email to