> From: "Scott G. Miller" <scgmille at freenetproject.org>
> 
> >> What??!??!
> >> User-friendliness is "not really appropriate for our goals of freedom of
> >> information" ??!?
> >Of course user friendliness is.  Being exactly like
> >Joe-Poorly-Designed-Filesharing-Network isn't.
> 
> Point taken. Sorry for the misunderstanding.
> 
> Anyway, Sebastian's new installer, and the new gui configurator announced
> earlier today, should hopefully create a brain-dead-easy foundation for a
> truly user-friendly freenet on windows.
> 
> All we need now is an arsenal of kickass freenet client apps.
> 
> Status of traditional internet applications, and their implementation (or
> lack thereof) over Freenet:
> 
> 1) Web browsing - now working well :)
> 
> 2) Email - nothing yet - apps urgently needed - either dedicated freenet
> f-mail application, or (preferably) POP/SMTP servers to support any email
> client in the way that fproxy/fcpproxy supports any web browser. This would
> be best if it uses openPGP libraries, and supports automatic key exchange,
> even runs a central in-freenet PK service.
> 
> 3) Chat - nothing yet - value of this could be questionable because of rate
> of key insertions required, and their effect on Freenet
> 
> 4) Usenet - the FMB beta client is an excellent start - it needs to support
> n 'newsgroups' instead of just one.

These last three things were being done by the EOF project, but I never much 
liked the implementation of them. The e-mail and usenet were implemented using 
servlets (ick), and chat--as you said--is of questionable value (but EOF has a 
history of doing crazy stuff, so we'll probably do it anyway).  Of the three, 
usenet is probably the best idea.

BTW--EOF is now becoming a standards body for implementing other Internet 
protocols over Freenet. Code coming out of the project, if any, will consist of 
referance distributions.

> 
> 5) File-sharing - while the Napster and Gnutella protocols as they stand are
> totally impossible within Freenet, many concepts can be borrowed and adapted
> into a scheme well-suited to Freenet, and front-ended with a specialised
> Freenet client. Search functionality is also very possible - some of FMB's
> architectural concepts could be borrowed for this (provided that FMB isn't
> filed under a software patent (...just kidding!)).

While no centralization should be put into Freenet itself, no one can stop a 
client writer from putting centralization in their client if they so choose. I 
would probably build a Freenet file sharing system using Freegle as a front-end.

> 
> 6) FTP - not supported yet - repositories of downloadable media files are
> presently implemented as a loose network of inter-linked freesites - scope
> exists for a freenet client to be implemented as a localhost FTP server,
> which inserts files that are picked up by an anonymous master server, which
> re-inserts these files under a master tree. Enumerated keys instead of
> date-based SSKs. Through using SSK public or private keys under the root,
> the FTP server can support the FTP client in inserting and requesting keys.

Some months ago, someone on the EOF lists mentioned that they were working on a 
FTP-Freenet client.  I think he got it working, too.

> 
> 7) CVS - nothing done yet, however the idea aroused interesting discussion
> on #freenet recently, where developers rated the idea as viable.
> 
> Cheers
> David
> 
> 
> 
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://lists.freenetproject.org/mailman/listinfo/devl
> 





-----------------------------------------------
Runbox Mail Manager - www.runbox.com
Online email application

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

Reply via email to