[mnet-devel] Freenet fork appears likely (was Re: Gmane -- Re: Why is Freenet so sick at the moment?) (fwd from s.schear@comcast.net)

[Zooko O'Whielacronx]

I've been paying a bit of attention to Freenet development since Ian Clarke 
asked me to review their new Next-Generation Routing.  (Here is the resulting 
discussion: [1].  I also tossed in some bonus advice about their new erasure 
coding.  Theirs looks like Mojo Nation's, and I advised them to forswear 
chunking the way ZNFF does, but they didn't take that advice.)

I would say Freenet is in a very similar situation as Mnet is in.  They're 
trying to develop in a disruptive new direction (NGR), they're having problems 
with the stability and performance of their current network in the wild, etc.

Glancing at their mailing list, I see a bunch of threads with the subject of 
"prod vs devl networks?", so I expect they'll come up with a solution that is 
similar to the one we've agreed on:

   "The Mnet Philosophy of Development"

   At any given time there are both evolutionary and revolutionary branches 
   under active development.  The former does not slow down the latter, and 
   the latter does not destabilize the former.


I would really *like* to see how Mnet v0.6.2 would do in terms of network 
stability and performance given a Freenet-sized user base.  The code is all 
finished -- we just need server ops, MetaTracker ops, and fresh packages.

Of course, Mnet v0.6 doesn't offer the sort of anonymity features that Freenet 
does.  (Perhaps some revolutionary experiments along those lines e.g.  Achord 
could be conducted on an UNSTABLE branch.) 

Regards,

Zooko

[1] http://sourceforge.net/mailarchive/message.php?msg_id=5802106



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
mnet-devel mailing list
mnet-devel at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mnet-devel