[freenet-chat] Re: [freenet-support] Freenet 0,5 and 0,7

Matthew Toseland toad at amphibian.dyndns.org
Tue Aug 29 20:53:28 UTC 2006


On Sun, Aug 27, 2006 at 12:51:32PM -0400, urza9814 at gmail.com wrote:
> Exactly. The theory of a darknet is you connect to people that you
> already know and trust. Now, there's a good chance of getting a
> worldwide-net because someone in group A may know and trust someone in
> group B, but chances are that not all of group A knows all of group B.

So what? Not everyone in group A knows everyone in group A. It doesn't
prevent routing from working. A node knows only a small number of other
nodes, but if you travel a few hops you can get to a large number of
other nodes.

> For a real-world analogy...I don't have a problem hanging out with my
> girlfriend and her friends...she has no problem being with me and my
> friends...but my friends and her friends would never meet
> independently. 

If one of her friends has a message for one of your friends, she gives
it to your gf, who gives it to you, who give it to that friend of yours.
Yes, you become a bottleneck here. But there is every reason from the
research to expect that there are other possible intermediaries.

> Perhaps they would become friends with time...and
> perhaps people in group A of the darknet would get to know and trust
> people in group B of the darknet....but that would take time. I mean,
> I know that personally it's gonna take a few years of knowing someone
> before I would trust them well enough to talk about the kinda stuff
> some people do on freenet. I mean, yea, that time might be lowered by
> someone else you trust saying 'they're cool, don't worry about
> it'...but still, by the time you have a global network, freenet 1.0 is
> gonna be out.

You don't have to connect to your girlfriend's friends.
> 
> Plus it makes freenet a much better target for government agencies.
> Chances are the people you are connected directly to in freenet you
> know very well. Chances are the people you know very well live in the
> same country as you, if for no other reason than a shared language. So
> chances are, if they bust one freenet node, they can bust all
> connected nodes.

So? If they bust one opennet node, they can bust every opennet node.
Whereas if they bust one darknet node, they can only (easily) bust the
darknet nodes directly connected to it. Spidering outward that way is
far more expensive than harvesting the entire opennet, busting a few
randomly chosen "examples", and cutting off internet access for the
rest.
> 
> And that actually made me think of one other thing. If you have a
> darknet in, say, Germany, they will most likely all speak German and
> upload German files. So how would they get joined to a darknet that
> mostly spoke English and uploads English files? Only people who speak
> both languages relatively well will bother to connect to both
> networks. 

And there are literally millions of such people across europe.

> But they have to not only speak both languages but also know
> and trust someone else who speaks the other language. Which seems to
> point back to smaller networks connected in few places.

I accept that there may be internal barriers. I don't accept that these
barriers are insurmountable, and I certainly don't expect the constant
implication that the darknets will be really small.
> 
> On 8/27/06, diddler4u at hotmail.com <diddler4u at hotmail.com> wrote:
> >Evan,
> >
> >You are right - there is a lot of data to show that social networks do
> >expand in the method being said here, but that data is based on known,
> >non-anonymous social networks. In an anonymous network the rule of thumb is
> >trust no one.
> >
> >If an openet is not the solution, neither is posting information with an
> >embeded IP number the solution. I don't know how the openet is hackable,
> >especially if node connections pr paths through nodes change randomly
> >(TOR-like), but with a manually established network it only takes capturing
> >1 node and the entire freenet is at risk. I would be more inclined to
> >exchange node information with someone if the information were encrypted -
> >private/public key. In an anonymous social network I would be more inclined
> >to expand that network to others because my node information is encrypted.
> >
> >
> >
> >>From: "Evan Daniel" <evanbd at gmail.com>
> >>Reply-To: evand at pobox.com, support at freenetproject.org
> >>To: "urza9814 at gmail.com" <urza9814 at gmail.com>
> >>CC: support at freenetproject.org
> >>Subject: Re: [freenet-support] Freenet 0,5 and 0,7
> >>Date: Sun, 27 Aug 2006 10:06:37 -0400
> >>
> >>Please justify your assumptions.
> >>
> >>There is a lot of data on social networks that says that is not how
> >>they look.  I see no reason to believe the social networks a freenet
> >>darknet would be built upon would be different.
> >>
> >>Evan
> >>
> >>On 8/26/06, urza9814 at gmail.com <urza9814 at gmail.com> wrote:
> >>>Yea, but you don't know all the nodes in the network, you just know
> >>>the ones your connected to. So if one of those links between the
> >>>networks goes down, half your downloads stall out and die. And
> >>>wouldn't that put a pretty big strain on certain computers? I mean, if
> >>>you get this global network of small networks...90% of the data you
> >>>request will probably be on another 'network'. The number of
> >>>connections between these networks is going to be a lot smaller than
> >>>connections within the network. Therefore the computers that connect
> >>>between them are gonna have a much greater strain on them than the
> >>>ones that are only linked to one 'network'. And if these individual
> >>>networks fully connect and integrate...you have an opennet. Except you
> >>>have to physically get your node connections from someone else. So you
> >>>have an opennet with much fewer connections, which doesn't seem like a
> >>>good thing.
> >>>
> >>>
> >>>On 8/26/06, Evan Daniel <evanbd at gmail.com> wrote:
> >>> > On 8/26/06, diddler4u at hotmail.com <diddler4u at hotmail.com> wrote:
> >>> > > >>Freenet 0.5 is an opennet. You connect to any random node that
> >>>happens
> >>> > > >>to be on. Freenet 0.7 doesn't have this yet. In 0.7, there is no
> >>>main
> >>> > > >>network. There might be now, but the idea of the way it currently
> >>>is
> >>> > > >>setup is to allow small groups to connect without connecting to
> >>> > > >>everyone else.
> >>> > > >
> >>> > > >That is not true.  Freenet 0.7 is designed to form one global
> >>>network, not
> >>> > > >multiple independent networks consisting of small groups.
> >>> > > >
> >>> > > >Ian.
> >>> > >
> >>> > > Ian,
> >>> > >
> >>> > > How can freenet grow to be a global network unless someone in one
> >>>group
> >>> > > trades connection information with someone in another group?
> >>> > >
> >>> > > Hypothetical - A group of people in England, another in France,
> >>>another in
> >>> > > Russia, and another in China have grown individual trusted 0.7
> >>>freenets. No
> >>> > > one in any of these groups knows someone in the other freenet group,
> >>>and
> >>> > > they don't want to just advertise in IRC chat to find someone to
> >>>connect to
> >>> > > because they don't know and trust this as a way to add people to
> >>>their
> >>> > > freenet. How will these freenet groups become a part of a global
> >>>network?
> >>> >
> >>> > They won't.  But your assumptions are off -- there's lots of good
> >>> > reasons to assume that once a small local network passes a handful of
> >>> > connected users it will gain a connection to a different network.  And
> >>> > then you have a global network.  This is what is meant when people say
> >>> > 0.7 is designed to form a global network -- there is no magic, except
> >>> > for the underlying properties of the social connections the network is
> >>> > built upon.
> >>> >
> >>> > Evan
> >>> > _______________________________________________
> >>> > Support mailing list
> >>> > Support at freenetproject.org
> >>> > http://news.gmane.org/gmane.network.freenet.support
> >>> > Unsubscribe at
> >>>http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
> >>> > Or mailto:support-request at freenetproject.org?subject=unsubscribe
> >>> >
> >>>
> >>>
> >>>--
> >>><HTML>
> >>><a
> >>>href="http://www.spreadfirefox.com/?q=affiliates&id=0&t=57"><img
> >>>border="0" alt="Get Firefox!" title="Get Firefox!"
> >>>src="http://sfx-images.mozilla.org/affiliates/Buttons/180x60/blank.gif"/></a>
> >>>
> >>>
> >>_______________________________________________
> >>Support mailing list
> >>Support at freenetproject.org
> >>http://news.gmane.org/gmane.network.freenet.support
> >>Unsubscribe at
> >>http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
> >>Or mailto:support-request at freenetproject.org?subject=unsubscribe
> >
> >_________________________________________________________________
> >Express yourself instantly with MSN Messenger! Download today it's FREE!
> >http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
> >
> >_______________________________________________
> >Support mailing list
> >Support at freenetproject.org
> >http://news.gmane.org/gmane.network.freenet.support
> >Unsubscribe at 
> >http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
> >Or mailto:support-request at freenetproject.org?subject=unsubscribe
> >
> 
> 
> -- 
> <HTML>
> <a href="http://www.spreadfirefox.com/?q=affiliates&amp;id=0&amp;t=57"><img
> border="0" alt="Get Firefox!" title="Get Firefox!"
> src="http://sfx-images.mozilla.org/affiliates/Buttons/180x60/blank.gif"/></a>
> _______________________________________________
> Support mailing list
> Support at freenetproject.org
> http://news.gmane.org/gmane.network.freenet.support
> Unsubscribe at 
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
> Or mailto:support-request at freenetproject.org?subject=unsubscribe
> 

-- 
Matthew J Toseland - toad at amphibian.dyndns.org
Freenet Project Official Codemonkey - http://freenetproject.org/
ICTHUS - Nothing is impossible. Our Boss says so.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
Url : http://emu.freenetproject.org/pipermail/chat/attachments/20060829/c971fc04/attachment.pgp 


More information about the chat mailing list