From dbkr at freenetproject.org  Thu Mar  1 09:24:25 2007
From: dbkr at freenetproject.org (Dave Baker)
Date: Thu, 1 Mar 2007 09:24:25 +0000
Subject: [freenet-dev] [freenet-cvs] r11936 -
	trunk/freenet/src/freenet/config
In-Reply-To: <20070228225125.E22219BB7E@emu.freenetproject.org>
References: <20070228225125.E22219BB7E@emu.freenetproject.org>
Message-ID: <200703010924.25542.dbkr@freenetproject.org>

On Wednesday 28 February 2007 22:51:25 nextgens at freenetproject.org wrote:
> Author: nextgens
> Date: 2007-02-28 22:51:25 +0000 (Wed, 28 Feb 2007)
> New Revision: 11936
>
> Modified:
>    trunk/freenet/src/freenet/config/PersistentConfig.java
> Log:
> Doh!

Hang on - could somebody explain what's going on with these three commits? As 
I see it, we start by catching an exception that's only caused by class 
changes and lack of a rebuild from clean (and by looking at the code it looks 
like it would wipe the config file). Then after that, it gets 'fixed', but 
rather than jnust reverting the commit, we iterate a different way (well, 
okay, whatever), but what worries me is that it has all the synchronisation 
removed. Was the synchronised {} block unnecessary? If so, why?

>
> Modified: trunk/freenet/src/freenet/config/PersistentConfig.java
> ===================================================================
> --- trunk/freenet/src/freenet/config/PersistentConfig.java	2007-02-28
> 22:41:01 UTC (rev 11935) +++
> trunk/freenet/src/freenet/config/PersistentConfig.java	2007-02-28 22:51:25
> UTC (rev 11936) @@ -34,11 +34,11 @@
>  	public synchronized SimpleFieldSet exportFieldSet(boolean withDefaults) {
>  		SimpleFieldSet fs = new SimpleFieldSet(true);
>  		Iterator configsIterator = configsByPrefix.keySet().iterator();
> -		SubConfig currentSubConfig;
> +		SubConfig current;
>  		while (configsIterator.hasNext()) {
> -			currentSubConfig = (SubConfig) configsIterator.next();
> -			SimpleFieldSet scfs = currentSubConfig.exportFieldSet(withDefaults);
> -			fs.tput(currentSubConfig.prefix, scfs);
> +			current = (SubConfig) configsByPrefix.get(configsIterator.next());
> +			SimpleFieldSet scfs = current.exportFieldSet(withDefaults);
> +			fs.tput(current.prefix, scfs);
>  		}
>  		return fs;
>  	}
>
> _______________________________________________
> cvs mailing list
> cvs at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/cvs




From nextgens at freenetproject.org  Thu Mar  1 12:48:44 2007
From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re_=28NextGen$=29?=)
Date: Thu, 1 Mar 2007 13:48:44 +0100
Subject: [freenet-dev] [freenet-cvs] r11936 -
	trunk/freenet/src/freenet/config
In-Reply-To: <200703010924.25542.dbkr@freenetproject.org>
References: <20070228225125.E22219BB7E@emu.freenetproject.org>
	<200703010924.25542.dbkr@freenetproject.org>
Message-ID: <20070301124844.GA4181@freenetproject.org>

* Dave Baker <dbkr at freenetproject.org> [2007-03-01 09:24:25]:

> On Wednesday 28 February 2007 22:51:25 nextgens at freenetproject.org wrote:
> > Author: nextgens
> > Date: 2007-02-28 22:51:25 +0000 (Wed, 28 Feb 2007)
> > New Revision: 11936
> >
> > Modified:
> >    trunk/freenet/src/freenet/config/PersistentConfig.java
> > Log:
> > Doh!
> 
> Hang on - could somebody explain what's going on with these three commits? As 
> I see it, we start by catching an exception that's only caused by class 
> changes and lack of a rebuild from clean (and by looking at the code it looks 
> like it would wipe the config file).

Indeed :  not a real problem.

> Then after that, it gets 'fixed', but rather than jnust reverting the commit, we
> iterate a different way (well, okay, whatever),

Yep, we iterate more efficiently now.

> but what worries me is that it has all the synchronisation 
> removed. Was the synchronised {} block unnecessary? If so, why?
> 

Cause the whole method is declared as synchronized so getting a
lock on (this) is useless :)

NextGen$

> >
> > Modified: trunk/freenet/src/freenet/config/PersistentConfig.java
> > ===================================================================
> > --- trunk/freenet/src/freenet/config/PersistentConfig.java	2007-02-28
> > 22:41:01 UTC (rev 11935) +++
> > trunk/freenet/src/freenet/config/PersistentConfig.java	2007-02-28 22:51:25
> > UTC (rev 11936) @@ -34,11 +34,11 @@
> >  	public synchronized SimpleFieldSet exportFieldSet(boolean withDefaults) {
> >  		SimpleFieldSet fs = new SimpleFieldSet(true);
> >  		Iterator configsIterator = configsByPrefix.keySet().iterator();
> > -		SubConfig currentSubConfig;
> > +		SubConfig current;
> >  		while (configsIterator.hasNext()) {
> > -			currentSubConfig = (SubConfig) configsIterator.next();
> > -			SimpleFieldSet scfs = currentSubConfig.exportFieldSet(withDefaults);
> > -			fs.tput(currentSubConfig.prefix, scfs);
> > +			current = (SubConfig) configsByPrefix.get(configsIterator.next());
> > +			SimpleFieldSet scfs = current.exportFieldSet(withDefaults);
> > +			fs.tput(current.prefix, scfs);
> >  		}
> >  		return fs;
> >  	}
> >
> > _______________________________________________
> > cvs mailing list
> > cvs at freenetproject.org
> > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/cvs
> 
> 
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl


From dbkr at freenetproject.org  Thu Mar  1 13:41:52 2007
From: dbkr at freenetproject.org (Dave Baker)
Date: Thu, 1 Mar 2007 13:41:52 +0000
Subject: [freenet-dev] [freenet-cvs] r11936 -
	trunk/freenet/src/freenet/config
In-Reply-To: <20070301124844.GA4181@freenetproject.org>
References: <20070228225125.E22219BB7E@emu.freenetproject.org>
	<200703010924.25542.dbkr@freenetproject.org>
	<20070301124844.GA4181@freenetproject.org>
Message-ID: <200703011341.52729.dbkr@freenetproject.org>

On Thursday 01 March 2007 12:48:44 Florent Daigni?re wrote:
> * Dave Baker <dbkr at freenetproject.org> [2007-03-01 09:24:25]:
> > On Wednesday 28 February 2007 22:51:25 nextgens at freenetproject.org wrote:
> > > Author: nextgens
> > > Date: 2007-02-28 22:51:25 +0000 (Wed, 28 Feb 2007)
> > > New Revision: 11936
> > >
> > > Modified:
> > >    trunk/freenet/src/freenet/config/PersistentConfig.java
> > > Log:
> > > Doh!
> >
> > Hang on - could somebody explain what's going on with these three
> > commits? As I see it, we start by catching an exception that's only
> > caused by class changes and lack of a rebuild from clean (and by looking
> > at the code it looks like it would wipe the config file).
>
> Indeed :  not a real problem.
>
> > Then after that, it gets 'fixed', but rather than jnust reverting the
> > commit, we iterate a different way (well, okay, whatever),
>
> Yep, we iterate more efficiently now.
>
> > but what worries me is that it has all the synchronisation
> > removed. Was the synchronised {} block unnecessary? If so, why?
>
> Cause the whole method is declared as synchronized so getting a
> lock on (this) is useless :)

Ah, I see. Fair enough. Can I persuade you to put things like this in the 
commit log message so people like me don't get confused? :)


Dave


>
> NextGen$
>
> > > Modified: trunk/freenet/src/freenet/config/PersistentConfig.java
> > > ===================================================================
> > > --- trunk/freenet/src/freenet/config/PersistentConfig.java	2007-02-28
> > > 22:41:01 UTC (rev 11935) +++
> > > trunk/freenet/src/freenet/config/PersistentConfig.java	2007-02-28
> > > 22:51:25 UTC (rev 11936) @@ -34,11 +34,11 @@
> > >  	public synchronized SimpleFieldSet exportFieldSet(boolean
> > > withDefaults) { SimpleFieldSet fs = new SimpleFieldSet(true);
> > >  		Iterator configsIterator = configsByPrefix.keySet().iterator();
> > > -		SubConfig currentSubConfig;
> > > +		SubConfig current;
> > >  		while (configsIterator.hasNext()) {
> > > -			currentSubConfig = (SubConfig) configsIterator.next();
> > > -			SimpleFieldSet scfs =
> > > currentSubConfig.exportFieldSet(withDefaults);
> > > -			fs.tput(currentSubConfig.prefix, scfs);
> > > +			current = (SubConfig) configsByPrefix.get(configsIterator.next());
> > > +			SimpleFieldSet scfs = current.exportFieldSet(withDefaults);
> > > +			fs.tput(current.prefix, scfs);
> > >  		}
> > >  		return fs;
> > >  	}
> > >
> > > _______________________________________________
> > > cvs mailing list
> > > cvs at freenetproject.org
> > > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/cvs
> >
> > _______________________________________________
> > Devl mailing list
> > Devl at freenetproject.org
> > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
>
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl




From nextgens at freenetproject.org  Thu Mar  1 13:45:54 2007
From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re_=28NextGen$=29?=)
Date: Thu, 1 Mar 2007 14:45:54 +0100
Subject: [freenet-dev] [freenet-cvs] r11936 -
	trunk/freenet/src/freenet/config
In-Reply-To: <200703011341.52729.dbkr@freenetproject.org>
References: <20070228225125.E22219BB7E@emu.freenetproject.org>
	<200703010924.25542.dbkr@freenetproject.org>
	<20070301124844.GA4181@freenetproject.org>
	<200703011341.52729.dbkr@freenetproject.org>
Message-ID: <20070301134553.GB4181@freenetproject.org>

* Dave Baker <dbkr at freenetproject.org> [2007-03-01 13:41:52]:

> On Thursday 01 March 2007 12:48:44 Florent Daigni?re wrote:
> > * Dave Baker <dbkr at freenetproject.org> [2007-03-01 09:24:25]:
> > > On Wednesday 28 February 2007 22:51:25 nextgens at freenetproject.org wrote:
> > > > Author: nextgens
> > > > Date: 2007-02-28 22:51:25 +0000 (Wed, 28 Feb 2007)
> > > > New Revision: 11936
> > > >
> > > > Modified:
> > > >    trunk/freenet/src/freenet/config/PersistentConfig.java
> > > > Log:
> > > > Doh!
> > >
> > > Hang on - could somebody explain what's going on with these three
> > > commits? As I see it, we start by catching an exception that's only
> > > caused by class changes and lack of a rebuild from clean (and by looking
> > > at the code it looks like it would wipe the config file).
> >
> > Indeed :  not a real problem.
> >
> > > Then after that, it gets 'fixed', but rather than jnust reverting the
> > > commit, we iterate a different way (well, okay, whatever),
> >
> > Yep, we iterate more efficiently now.
> >
> > > but what worries me is that it has all the synchronisation
> > > removed. Was the synchronised {} block unnecessary? If so, why?
> >
> > Cause the whole method is declared as synchronized so getting a
> > lock on (this) is useless :)
> 
> Ah, I see. Fair enough. Can I persuade you to put things like this in the 
> commit log message so people like me don't get confused? :)
> 
> 
> Dave

Sure, next time I will do :)

NextGen$


From toad at amphibian.dyndns.org  Thu Mar  1 14:08:57 2007
From: toad at amphibian.dyndns.org (Matthew Toseland)
Date: Thu, 1 Mar 2007 14:08:57 +0000
Subject: [freenet-dev] Fwd: [Content Restrictions and Freenet]
Message-ID: <20070301140857.GC21603@amphibian.dyndns.org>

In reference to http://www.gerv.net/security/content-restrictions/
Which may provide a way for us to provide full web functionality without
having to write whitelist filters for everything including javascript !

----- Forwarded message from me -----

Subject: Content Restrictions and Freenet

Hi, I believe we met at FOSDEM?

I work on Freenet, http://freenetproject.org/ , a censorship resistant
peer to peer system designed for use in hostile regimes. This includes
a web front-end: Files and web sites are published internally and
accessed through localhost. In order to prevent web-bugs (img src's
pointing to files on the real web giving away your IP address, for
example), at present we implement a whitelist filter, which only
supports HTML 4.01 and no scripting. Your Content Restrictions proposal
may give us a way to support more or less full web functionality.

Specific issues:
- We would probably use request=nopost domain=127.0.0.1
- For our purposes, going beyond scripting to restrict all content on
  the page is a good thing. It means we don't have to filter the content
  we pass along to the browser at all, once we know the browser supports
  the content restrictions. This will be the case with any system where
  you can upload HTML including arbitrary tags, although I can see that
  most e.g. forums may want to impose a strictly limited subset.
- We obviously need some way to detect whether this is in operation in a
  script in a filtered page; if it is, we redirect to an unfiltered one.
- Obviously e.g. included CSS files would need to be interpreted with
  the same policy. At present we have to put in an explicit MIME type
  specification in <link rel=stylesheet> to ensure that CSS is filtered
  correctly.
- Because we will want some forms (e.g. a search form within a portal
  site), we will need to be able to have an unfiltered page inline a
  filtered stylesheet.

Thanks!

----- End forwarded message -----
-------------- 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/devl/attachments/20070301/31b6565f/attachment.pgp 

From nextgens at freenetproject.org  Sat Mar  3 15:10:33 2007
From: nextgens at freenetproject.org (NextGen$)
Date: Sat, 3 Mar 2007 15:10:33 +0000
Subject: [freenet-dev] Gentoo, freenet 0.5 ebuild
Message-ID: <20070303151033.GA937@freenetproject.org>

Hi,

	I got in touch with the gentoo guys and convinced them to fix|remove
	their ebuild \o/

	https://bugs.gentoo.org/show_bug.cgi?id=72996#c8
	hopefully we won't have support requests from gentoo users anymore.

	I think it's a good news ; Not that we shouldn't encourage packaging,
	but their package was really old and hasn't been maintained for to
	long.

NextGen$
PS: I've created a new related ticket on mantis :
https://bugs.freenetproject.org/view.php?id=1174


From nextgens at freenetproject.org  Mon Mar  5 15:02:57 2007
From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re_=28NextGen$=29?=)
Date: Mon, 5 Mar 2007 16:02:57 +0100
Subject: [freenet-dev] The new plugin : MDNSDiscovery
Message-ID: <20070305150257.GA4497@freenetproject.org>

Hey!

	This WE I got around coding a new plugin for freenet :
	MDNSDiscovery. It advertises fred's services on the local network.
	Namely _freenet._udp (the node), _fcp._tcp (FCP), _http._tcp (Froxy)
	and _telnet._tcp (TCMI). MDNS is also known as ZeroConf (see [1])
	or Bonjour/Rendez-Vous in apple's world (see [2]).

	So far no FCP client is known to be listening for zeroconf
	advertisement ; maybe their authors should consider doing it so
	that they could "automagicaly" discover FCP servers.
	I have filled in several tickets on mantis (see [3]).

	MacOSx users should now see a bookmark for fproxy in Finder,
	under Network Servers.

	We could imagine lots of applications for that plugin including
	"easy peer swapping capabilities with neighbors" ;) I suggest to
	people interested in such functionality to express their
	interests in a new thread on @darknet-tools.

	The plugin runs fine for me with the testing version of freenet
	and follows 1.4 strict java compatibility. Would someone object
	to me bundling it within the installer when 1017 is out ? Of
	course as it "advertises" freenet on the local subnet, there would
	be an "adequate" warning ;)

	I have also created a wiki page dedicated to its installation
	[6]

NextGen$

PS: I have used jmdns [4] to do it, and would recommend python addicts
to use avahi [5]. I know that having several MDNS stacks on the same
host isn't a good solution but I don't regard using JNI to access DBUS
to talk to avahi as a realistic option.

[1] http://www.zeroconf.org/
[2] http://www.apple.com/macosx/features/bonjour/
[3] https://bugs.freenetproject.org/view.php?id=1179
[4] http://jmdns.sourceforge.net/
[5] http://avahi.org/
[6] http://wiki.freenetproject.org/MDNSDiscoveryPlugin


From toad at amphibian.dyndns.org  Mon Mar  5 16:35:48 2007
From: toad at amphibian.dyndns.org (Matthew Toseland)
Date: Mon, 5 Mar 2007 16:35:48 +0000
Subject: [freenet-dev] The new plugin : MDNSDiscovery
In-Reply-To: <20070305150257.GA4497@freenetproject.org>
References: <20070305150257.GA4497@freenetproject.org>
Message-ID: <20070305163548.GF29425@amphibian.dyndns.org>

Nice! I thought it was possible to detect IP address over Bonjour?

One issue here is that ISP NATs often look like LANs; how do we tell the
difference? We can certainly have an option to make Freenet accessible
from the LAN and to advertise it on the LAN, but first we have to figure
out whether it IS a LAN. I suppose 99% of the time it is because people
are behind a router even if their ISP NATs them as well. :|

On Mon, Mar 05, 2007 at 04:02:57PM +0100, Florent Daigni?re (NextGen$) wrote:
> Hey!
> 
> 	This WE I got around coding a new plugin for freenet :
> 	MDNSDiscovery. It advertises fred's services on the local network.
> 	Namely _freenet._udp (the node), _fcp._tcp (FCP), _http._tcp (Froxy)
> 	and _telnet._tcp (TCMI). MDNS is also known as ZeroConf (see [1])
> 	or Bonjour/Rendez-Vous in apple's world (see [2]).
> 
> 	So far no FCP client is known to be listening for zeroconf
> 	advertisement ; maybe their authors should consider doing it so
> 	that they could "automagicaly" discover FCP servers.
> 	I have filled in several tickets on mantis (see [3]).
> 
> 	MacOSx users should now see a bookmark for fproxy in Finder,
> 	under Network Servers.
> 
> 	We could imagine lots of applications for that plugin including
> 	"easy peer swapping capabilities with neighbors" ;) I suggest to
> 	people interested in such functionality to express their
> 	interests in a new thread on @darknet-tools.
> 
> 	The plugin runs fine for me with the testing version of freenet
> 	and follows 1.4 strict java compatibility. Would someone object
> 	to me bundling it within the installer when 1017 is out ? Of
> 	course as it "advertises" freenet on the local subnet, there would
> 	be an "adequate" warning ;)
> 
> 	I have also created a wiki page dedicated to its installation
> 	[6]
> 
> NextGen$
> 
> PS: I have used jmdns [4] to do it, and would recommend python addicts
> to use avahi [5]. I know that having several MDNS stacks on the same
> host isn't a good solution but I don't regard using JNI to access DBUS
> to talk to avahi as a realistic option.
> 
> [1] http://www.zeroconf.org/
> [2] http://www.apple.com/macosx/features/bonjour/
> [3] https://bugs.freenetproject.org/view.php?id=1179
> [4] http://jmdns.sourceforge.net/
> [5] http://avahi.org/
> [6] http://wiki.freenetproject.org/MDNSDiscoveryPlugin
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> 
-------------- 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/devl/attachments/20070305/1ea820a0/attachment.pgp 

From nextgens at freenetproject.org  Mon Mar  5 16:44:41 2007
From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re_=28NextGen$=29?=)
Date: Mon, 5 Mar 2007 17:44:41 +0100
Subject: [freenet-dev] The new plugin : MDNSDiscovery
In-Reply-To: <20070305163548.GF29425@amphibian.dyndns.org>
References: <20070305150257.GA4497@freenetproject.org>
	<20070305163548.GF29425@amphibian.dyndns.org>
Message-ID: <20070305164441.GF4497@freenetproject.org>

* Matthew Toseland <toad at amphibian.dyndns.org> [2007-03-05 16:35:48]:

> Nice! I thought it was possible to detect IP address over Bonjour?

Well, I haven't implemented that yet: it's called NAT-PMP
http://files.dns-sd.org/draft-cheshire-nat-pmp.txt ... the underlying
library (jmdns) doesn't support it... and only a few hardware device do
atm (like apple's airports).

> 
> One issue here is that ISP NATs often look like LANs; how do we tell the
> difference? We can certainly have an option to make Freenet accessible
> from the LAN and to advertise it on the LAN, but first we have to figure
> out whether it IS a LAN. I suppose 99% of the time it is because people
> are behind a router even if their ISP NATs them as well. :|

Asking the user is the only viable option.

> On Mon, Mar 05, 2007 at 04:02:57PM +0100, Florent Daigni?re (NextGen$) wrote:
> > Hey!
> > 
> > 	This WE I got around coding a new plugin for freenet :
> > 	MDNSDiscovery. It advertises fred's services on the local network.
> > 	Namely _freenet._udp (the node), _fcp._tcp (FCP), _http._tcp (Froxy)
> > 	and _telnet._tcp (TCMI). MDNS is also known as ZeroConf (see [1])
> > 	or Bonjour/Rendez-Vous in apple's world (see [2]).
> > 
> > 	So far no FCP client is known to be listening for zeroconf
> > 	advertisement ; maybe their authors should consider doing it so
> > 	that they could "automagicaly" discover FCP servers.
> > 	I have filled in several tickets on mantis (see [3]).
> > 
> > 	MacOSx users should now see a bookmark for fproxy in Finder,
> > 	under Network Servers.
> > 
> > 	We could imagine lots of applications for that plugin including
> > 	"easy peer swapping capabilities with neighbors" ;) I suggest to
> > 	people interested in such functionality to express their
> > 	interests in a new thread on @darknet-tools.
> > 
> > 	The plugin runs fine for me with the testing version of freenet
> > 	and follows 1.4 strict java compatibility. Would someone object
> > 	to me bundling it within the installer when 1017 is out ? Of
> > 	course as it "advertises" freenet on the local subnet, there would
> > 	be an "adequate" warning ;)
> > 
> > 	I have also created a wiki page dedicated to its installation
> > 	[6]
> > 
> > NextGen$
> > 
> > PS: I have used jmdns [4] to do it, and would recommend python addicts
> > to use avahi [5]. I know that having several MDNS stacks on the same
> > host isn't a good solution but I don't regard using JNI to access DBUS
> > to talk to avahi as a realistic option.
> > 
> > [1] http://www.zeroconf.org/
> > [2] http://www.apple.com/macosx/features/bonjour/
> > [3] https://bugs.freenetproject.org/view.php?id=1179
> > [4] http://jmdns.sourceforge.net/
> > [5] http://avahi.org/
> > [6] http://wiki.freenetproject.org/MDNSDiscoveryPlugin
> > _______________________________________________
> > Devl mailing list
> > Devl at freenetproject.org
> > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> > 




From toad at amphibian.dyndns.org  Mon Mar  5 17:45:45 2007
From: toad at amphibian.dyndns.org (Matthew Toseland)
Date: Mon, 5 Mar 2007 17:45:45 +0000
Subject: [freenet-dev] Easier reference swapping
Message-ID: <20070305174545.GA29567@amphibian.dyndns.org>

We will only get a darknet if it is really easy to swap references with
your friends - opennet or no opennet.

The original idea for Freenet 0.7 reference swapping was that you:
- Go to your node, and ask it to create a bundle.
- Send the bundle to your friends.
- They unzip it and run it to install Freenet.
- The bundle includes your noderef.
- It also includes a one-time key that allows the node to automatically
  connect to yours despite yours not having their noderef yet.

There are two big problems with this:

1) Everyone and his dog is behind a NAT. This means in order to connect
you must have already exchanged references, full stop. THIS SUCKS. It
also affects connectivity for newbies in a bad way (which is important
IMHO).

2) Generally people shouldn't run programs that they receive in emails!

Solution to the first one - and to newbie connectivity issues - is to
implement UP&P and hope that routers implement it properly in future -
is this a realistic hope?

Solution to the second one is to just send the noderef and a link to the
website, and only use full bundles when e.g. giving somebody a CD-R (which
we should make really easy).

Plugins for e.g. IRC clients, IM clients, have been suggested but I'm
not sure how well this would work for newbies, and in any case I set up
a darknet-tools list for people to talk about this and nobody has even
talked about it since a few days after it was set up, let alone done
anything.

<_ph00> so the basic problem is "how to safely exchage refs", and the
solution "eliminate ref exchanging by implementing opennet"?!? Am I the
only one to think that's very stupid?
-------------- 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/devl/attachments/20070305/1fe1de12/attachment.pgp 

From Colin at sq7.org  Mon Mar  5 18:02:42 2007
From: Colin at sq7.org (Colin Davis)
Date: Mon, 05 Mar 2007 13:02:42 -0500
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <20070305174545.GA29567@amphibian.dyndns.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
Message-ID: <45EC5B42.1010701@sq7.org>

I know it's less secure, but what about simply allowing people to 
connect to your machine if they know a passphrase? The passphrase would 
take the place of the Key, but be user-settable, and short.



If that were in place, you could send an e-mail saying:

Hey Jon, I just found this cool new thing called freenet, which lets you 
get to all sorts of sites which aren't on the normal web! It's 
anonymous, and free, you should check it out. It works by connecting 
through each other's computers, but I'll let you connect to me to get 
started.

Go to FreenetProject.org and download it, then give it my hostname, 
which is XXXXXXX and give it the connection passphrase "IamNotEvil".

Don't give anyone else that information, or it won't work. It'll only 
allow one connection.. After your up, you can connect to other friends, 
and everyone's connection gets faster.

I'm on IM if you want to talk about it.
-Person you Know.








Matthew Toseland wrote:
> We will only get a darknet if it is really easy to swap references with
> your friends - opennet or no opennet.
> 
> The original idea for Freenet 0.7 reference swapping was that you:
> - Go to your node, and ask it to create a bundle.
> - Send the bundle to your friends.
> - They unzip it and run it to install Freenet.
> - The bundle includes your noderef.
> - It also includes a one-time key that allows the node to automatically
>   connect to yours despite yours not having their noderef yet.
> 
> There are two big problems with this:
> 
> 1) Everyone and his dog is behind a NAT. This means in order to connect
> you must have already exchanged references, full stop. THIS SUCKS. It
> also affects connectivity for newbies in a bad way (which is important
> IMHO).
> 
> 2) Generally people shouldn't run programs that they receive in emails!
> 
> Solution to the first one - and to newbie connectivity issues - is to
> implement UP&P and hope that routers implement it properly in future -
> is this a realistic hope?
> 
> Solution to the second one is to just send the noderef and a link to the
> website, and only use full bundles when e.g. giving somebody a CD-R (which
> we should make really easy).
> 
> Plugins for e.g. IRC clients, IM clients, have been suggested but I'm
> not sure how well this would work for newbies, and in any case I set up
> a darknet-tools list for people to talk about this and nobody has even
> talked about it since a few days after it was set up, let alone done
> anything.
> 
> <_ph00> so the basic problem is "how to safely exchage refs", and the
> solution "eliminate ref exchanging by implementing opennet"?!? Am I the
> only one to think that's very stupid?
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl



From dbkr at freenetproject.org  Mon Mar  5 18:09:41 2007
From: dbkr at freenetproject.org (Dave Baker)
Date: Mon, 5 Mar 2007 18:09:41 +0000
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <45EC5B42.1010701@sq7.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<45EC5B42.1010701@sq7.org>
Message-ID: <200703051809.41642.dbkr@freenetproject.org>

On Monday 05 March 2007 18:02:42 Colin Davis wrote:
> I know it's less secure, but what about simply allowing people to
> connect to your machine if they know a passphrase? The passphrase would
> take the place of the Key, but be user-settable, and short.

That doesn't solve either problem though, surely?

my 2p on #freenet:

[17:48] <dbkr> as far as both-way-adding goes, I think that's where we reach a 
tradeoff with security, which is one of the main challanges for Freenet.
[17:49] <dbkr> I'm not convinced the whole difficulty of exchanging refs isn't 
a red herring - everyone can handle emailing a file.

I'm definately in favour of the ability to burn a CD with an installer on it 
that installs a node with your reference pre-bundled, although I think 
leaving the installer out for an emailed-version means it's nothing the user 
couldn't do themselves.


Dave


>
>
>
> If that were in place, you could send an e-mail saying:
>
> Hey Jon, I just found this cool new thing called freenet, which lets you
> get to all sorts of sites which aren't on the normal web! It's
> anonymous, and free, you should check it out. It works by connecting
> through each other's computers, but I'll let you connect to me to get
> started.
>
> Go to FreenetProject.org and download it, then give it my hostname,
> which is XXXXXXX and give it the connection passphrase "IamNotEvil".
>
> Don't give anyone else that information, or it won't work. It'll only
> allow one connection.. After your up, you can connect to other friends,
> and everyone's connection gets faster.
>
> I'm on IM if you want to talk about it.
> -Person you Know.
>
> Matthew Toseland wrote:
> > We will only get a darknet if it is really easy to swap references with
> > your friends - opennet or no opennet.
> >
> > The original idea for Freenet 0.7 reference swapping was that you:
> > - Go to your node, and ask it to create a bundle.
> > - Send the bundle to your friends.
> > - They unzip it and run it to install Freenet.
> > - The bundle includes your noderef.
> > - It also includes a one-time key that allows the node to automatically
> >   connect to yours despite yours not having their noderef yet.
> >
> > There are two big problems with this:
> >
> > 1) Everyone and his dog is behind a NAT. This means in order to connect
> > you must have already exchanged references, full stop. THIS SUCKS. It
> > also affects connectivity for newbies in a bad way (which is important
> > IMHO).
> >
> > 2) Generally people shouldn't run programs that they receive in emails!
> >
> > Solution to the first one - and to newbie connectivity issues - is to
> > implement UP&P and hope that routers implement it properly in future -
> > is this a realistic hope?
> >
> > Solution to the second one is to just send the noderef and a link to the
> > website, and only use full bundles when e.g. giving somebody a CD-R
> > (which we should make really easy).
> >
> > Plugins for e.g. IRC clients, IM clients, have been suggested but I'm
> > not sure how well this would work for newbies, and in any case I set up
> > a darknet-tools list for people to talk about this and nobody has even
> > talked about it since a few days after it was set up, let alone done
> > anything.
> >
> > <_ph00> so the basic problem is "how to safely exchage refs", and the
> > solution "eliminate ref exchanging by implementing opennet"?!? Am I the
> > only one to think that's very stupid?
> >
> >
> > ------------------------------------------------------------------------
> >
> > _______________________________________________
> > Devl mailing list
> > Devl at freenetproject.org
> > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
>
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl




From Colin at sq7.org  Mon Mar  5 18:13:43 2007
From: Colin at sq7.org (Colin Davis)
Date: Mon, 05 Mar 2007 13:13:43 -0500
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <200703051809.41642.dbkr@freenetproject.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>	<45EC5B42.1010701@sq7.org>
	<200703051809.41642.dbkr@freenetproject.org>
Message-ID: <45EC5DD7.6080302@sq7.org>

It solves #2- Don't run things you get in e-mail..

Instead of requiring a Noderef, allow someone to connect with just a 
password, and the IP address. This is something you can TELL someone, or 
say in an IM, no file transfer required.




Dave Baker wrote:
> On Monday 05 March 2007 18:02:42 Colin Davis wrote:
>> I know it's less secure, but what about simply allowing people to
>> connect to your machine if they know a passphrase? The passphrase would
>> take the place of the Key, but be user-settable, and short.
> 
> That doesn't solve either problem though, surely?
> 
> my 2p on #freenet:
> 
> [17:48] <dbkr> as far as both-way-adding goes, I think that's where we reach a 
> tradeoff with security, which is one of the main challanges for Freenet.
> [17:49] <dbkr> I'm not convinced the whole difficulty of exchanging refs isn't 
> a red herring - everyone can handle emailing a file.
> 
> I'm definately in favour of the ability to burn a CD with an installer on it 
> that installs a node with your reference pre-bundled, although I think 
> leaving the installer out for an emailed-version means it's nothing the user 
> couldn't do themselves.
> 
> 
> Dave
> 
> 
>>
>>
>> If that were in place, you could send an e-mail saying:
>>
>> Hey Jon, I just found this cool new thing called freenet, which lets you
>> get to all sorts of sites which aren't on the normal web! It's
>> anonymous, and free, you should check it out. It works by connecting
>> through each other's computers, but I'll let you connect to me to get
>> started.
>>
>> Go to FreenetProject.org and download it, then give it my hostname,
>> which is XXXXXXX and give it the connection passphrase "IamNotEvil".
>>
>> Don't give anyone else that information, or it won't work. It'll only
>> allow one connection.. After your up, you can connect to other friends,
>> and everyone's connection gets faster.
>>
>> I'm on IM if you want to talk about it.
>> -Person you Know.
>>
>> Matthew Toseland wrote:
>>> We will only get a darknet if it is really easy to swap references with
>>> your friends - opennet or no opennet.
>>>
>>> The original idea for Freenet 0.7 reference swapping was that you:
>>> - Go to your node, and ask it to create a bundle.
>>> - Send the bundle to your friends.
>>> - They unzip it and run it to install Freenet.
>>> - The bundle includes your noderef.
>>> - It also includes a one-time key that allows the node to automatically
>>>   connect to yours despite yours not having their noderef yet.
>>>
>>> There are two big problems with this:
>>>
>>> 1) Everyone and his dog is behind a NAT. This means in order to connect
>>> you must have already exchanged references, full stop. THIS SUCKS. It
>>> also affects connectivity for newbies in a bad way (which is important
>>> IMHO).
>>>
>>> 2) Generally people shouldn't run programs that they receive in emails!
>>>
>>> Solution to the first one - and to newbie connectivity issues - is to
>>> implement UP&P and hope that routers implement it properly in future -
>>> is this a realistic hope?
>>>
>>> Solution to the second one is to just send the noderef and a link to the
>>> website, and only use full bundles when e.g. giving somebody a CD-R
>>> (which we should make really easy).
>>>
>>> Plugins for e.g. IRC clients, IM clients, have been suggested but I'm
>>> not sure how well this would work for newbies, and in any case I set up
>>> a darknet-tools list for people to talk about this and nobody has even
>>> talked about it since a few days after it was set up, let alone done
>>> anything.
>>>
>>> <_ph00> so the basic problem is "how to safely exchage refs", and the
>>> solution "eliminate ref exchanging by implementing opennet"?!? Am I the
>>> only one to think that's very stupid?
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> _______________________________________________
>>> Devl mailing list
>>> Devl at freenetproject.org
>>> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
>> _______________________________________________
>> Devl mailing list
>> Devl at freenetproject.org
>> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> 
> 
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl



From freenet-devl at david.sowder.com  Mon Mar  5 19:52:33 2007
From: freenet-devl at david.sowder.com (David Sowder (Zothar))
Date: Mon, 05 Mar 2007 13:52:33 -0600
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <45EC5DD7.6080302@sq7.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>	<45EC5B42.1010701@sq7.org>	<200703051809.41642.dbkr@freenetproject.org>
	<45EC5DD7.6080302@sq7.org>
Message-ID: <45EC7501.80509@david.sowder.com>

Colin Davis wrote:
> It solves #2- Don't run things you get in e-mail..
>
> Instead of requiring a Noderef, allow someone to connect with just a 
> password, and the IP address. This is something you can TELL someone, or 
> say in an IM, no file transfer required.
>   
I don't think we should necessarily categorically discount a fred-based
installer distribution servlet because of NAT problems.  What if we
allow decoupling the installer from the noderef?  Then those that can
overcome or don't have the NAT problem can host the installer.  If I,
for some reason, cannot host the installer, perhaps one of my peers is
willing to share their installer hosting such that I could get "access
keys" from his node to pass out to my friends and include my noderef
from the distribution area of FProxy in the email I send to my friend
with the "access key".  My friend saves the attached noderef (which was
attached as a file with a .fref extension) to a file, connects to the my
peer's installer host, uses the "access key", which will probably be
part of the URL, and downloads and installs the node.  Then my friend
can double click on the .fref file he downloaded, which includes a
one-time code generated by the distribution area of my FProxy (or FCP
server) that my node uses to authenticate the addition of a node I
didn't already have the noderef of.  Node installed in a decentralized
way.  Peer connection created.

Pass phrases could work, but I think they should have something like a
32 character minimum length.  (I wonder if there are passphrase
dictionaries yet.)
> Dave Baker wrote:
>   
>> On Monday 05 March 2007 18:02:42 Colin Davis wrote:
>>     
>>> I know it's less secure, but what about simply allowing people to
>>> connect to your machine if they know a passphrase? The passphrase would
>>> take the place of the Key, but be user-settable, and short.
>>>       
>> That doesn't solve either problem though, surely?
>>
>> my 2p on #freenet:
>>
>> [17:48] <dbkr> as far as both-way-adding goes, I think that's where we reach a 
>> tradeoff with security, which is one of the main challanges for Freenet.
>> [17:49] <dbkr> I'm not convinced the whole difficulty of exchanging refs isn't 
>> a red herring - everyone can handle emailing a file.
>>
>> I'm definately in favour of the ability to burn a CD with an installer on it 
>> that installs a node with your reference pre-bundled, although I think 
>> leaving the installer out for an emailed-version means it's nothing the user 
>> couldn't do themselves.
>>
>>
>> Dave
>>
>>
>>     
>>> If that were in place, you could send an e-mail saying:
>>>
>>> Hey Jon, I just found this cool new thing called freenet, which lets you
>>> get to all sorts of sites which aren't on the normal web! It's
>>> anonymous, and free, you should check it out. It works by connecting
>>> through each other's computers, but I'll let you connect to me to get
>>> started.
>>>
>>> Go to FreenetProject.org and download it, then give it my hostname,
>>> which is XXXXXXX and give it the connection passphrase "IamNotEvil".
>>>
>>> Don't give anyone else that information, or it won't work. It'll only
>>> allow one connection.. After your up, you can connect to other friends,
>>> and everyone's connection gets faster.
>>>
>>> I'm on IM if you want to talk about it.
>>> -Person you Know.
>>>
>>> Matthew Toseland wrote:
>>>       
>>>> We will only get a darknet if it is really easy to swap references with
>>>> your friends - opennet or no opennet.
>>>>
>>>> The original idea for Freenet 0.7 reference swapping was that you:
>>>> - Go to your node, and ask it to create a bundle.
>>>> - Send the bundle to your friends.
>>>> - They unzip it and run it to install Freenet.
>>>> - The bundle includes your noderef.
>>>> - It also includes a one-time key that allows the node to automatically
>>>>   connect to yours despite yours not having their noderef yet.
>>>>
>>>> There are two big problems with this:
>>>>
>>>> 1) Everyone and his dog is behind a NAT. This means in order to connect
>>>> you must have already exchanged references, full stop. THIS SUCKS. It
>>>> also affects connectivity for newbies in a bad way (which is important
>>>> IMHO).
>>>>
>>>> 2) Generally people shouldn't run programs that they receive in emails!
>>>>
>>>> Solution to the first one - and to newbie connectivity issues - is to
>>>> implement UP&P and hope that routers implement it properly in future -
>>>> is this a realistic hope?
>>>>
>>>> Solution to the second one is to just send the noderef and a link to the
>>>> website, and only use full bundles when e.g. giving somebody a CD-R
>>>> (which we should make really easy).
>>>>
>>>> Plugins for e.g. IRC clients, IM clients, have been suggested but I'm
>>>> not sure how well this would work for newbies, and in any case I set up
>>>> a darknet-tools list for people to talk about this and nobody has even
>>>> talked about it since a few days after it was set up, let alone done
>>>> anything.
>>>>
>>>> <_ph00> so the basic problem is "how to safely exchage refs", and the
>>>> solution "eliminate ref exchanging by implementing opennet"?!? Am I the
>>>> only one to think that's very stupid?
>>>>         


From Colin at sq7.org  Mon Mar  5 20:06:35 2007
From: Colin at sq7.org (Colin Davis)
Date: Mon, 05 Mar 2007 15:06:35 -0500
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <45EC7501.80509@david.sowder.com>
References: <20070305174545.GA29567@amphibian.dyndns.org>	<45EC5B42.1010701@sq7.org>	<200703051809.41642.dbkr@freenetproject.org>	<45EC5DD7.6080302@sq7.org>
	<45EC7501.80509@david.sowder.com>
Message-ID: <45EC784B.3020908@sq7.org>

I have no problem with an installer, that include your noderef- In an 
ideal world, that'd be the primary way people Get freenet. If I recall, 
.5 created a custom install, but almost no one ever used it.

I think the passphrase solution works well enough, if it's locked down..


As for NAT issues, forgive my being out of touch, but I thought that was 
solved a year ago.. If I recall, I thought that even if Side-A and 
Side-B were both firewalled, they could connect..

When Side-A added the noderef of B, it would start sending out packets 
to B, knowing that they won't get returned.. But the Sending of these 
packets would open a NAT-hole..
Then, Side-B would get around to adding the noderef of A, and start 
sending packets to A, opening a NAT-hole on their own side.. Side A's 
packets then get in through this hole, just as B's now arrive at A.

What am I missing?

-Colin


That said,
David Sowder (Zothar) wrote:
> Colin Davis wrote:
>> It solves #2- Don't run things you get in e-mail..
>>
>> Instead of requiring a Noderef, allow someone to connect with just a 
>> password, and the IP address. This is something you can TELL someone, or 
>> say in an IM, no file transfer required.
>>   
> I don't think we should necessarily categorically discount a fred-based
> installer distribution servlet because of NAT problems.  What if we
> allow decoupling the installer from the noderef?  Then those that can
> overcome or don't have the NAT problem can host the installer.  If I,
> for some reason, cannot host the installer, perhaps one of my peers is
> willing to share their installer hosting such that I could get "access
> keys" from his node to pass out to my friends and include my noderef
> from the distribution area of FProxy in the email I send to my friend
> with the "access key".  My friend saves the attached noderef (which was
> attached as a file with a .fref extension) to a file, connects to the my
> peer's installer host, uses the "access key", which will probably be
> part of the URL, and downloads and installs the node.  Then my friend
> can double click on the .fref file he downloaded, which includes a
> one-time code generated by the distribution area of my FProxy (or FCP
> server) that my node uses to authenticate the addition of a node I
> didn't already have the noderef of.  Node installed in a decentralized
> way.  Peer connection created.
> 
> Pass phrases could work, but I think they should have something like a
> 32 character minimum length.  (I wonder if there are passphrase
> dictionaries yet.)
>> Dave Baker wrote:
>>   
>>> On Monday 05 March 2007 18:02:42 Colin Davis wrote:
>>>     
>>>> I know it's less secure, but what about simply allowing people to
>>>> connect to your machine if they know a passphrase? The passphrase would
>>>> take the place of the Key, but be user-settable, and short.
>>>>       
>>> That doesn't solve either problem though, surely?
>>>
>>> my 2p on #freenet:
>>>
>>> [17:48] <dbkr> as far as both-way-adding goes, I think that's where we reach a 
>>> tradeoff with security, which is one of the main challanges for Freenet.
>>> [17:49] <dbkr> I'm not convinced the whole difficulty of exchanging refs isn't 
>>> a red herring - everyone can handle emailing a file.
>>>
>>> I'm definately in favour of the ability to burn a CD with an installer on it 
>>> that installs a node with your reference pre-bundled, although I think 
>>> leaving the installer out for an emailed-version means it's nothing the user 
>>> couldn't do themselves.
>>>
>>>
>>> Dave
>>>
>>>
>>>     
>>>> If that were in place, you could send an e-mail saying:
>>>>
>>>> Hey Jon, I just found this cool new thing called freenet, which lets you
>>>> get to all sorts of sites which aren't on the normal web! It's
>>>> anonymous, and free, you should check it out. It works by connecting
>>>> through each other's computers, but I'll let you connect to me to get
>>>> started.
>>>>
>>>> Go to FreenetProject.org and download it, then give it my hostname,
>>>> which is XXXXXXX and give it the connection passphrase "IamNotEvil".
>>>>
>>>> Don't give anyone else that information, or it won't work. It'll only
>>>> allow one connection.. After your up, you can connect to other friends,
>>>> and everyone's connection gets faster.
>>>>
>>>> I'm on IM if you want to talk about it.
>>>> -Person you Know.
>>>>
>>>> Matthew Toseland wrote:
>>>>       
>>>>> We will only get a darknet if it is really easy to swap references with
>>>>> your friends - opennet or no opennet.
>>>>>
>>>>> The original idea for Freenet 0.7 reference swapping was that you:
>>>>> - Go to your node, and ask it to create a bundle.
>>>>> - Send the bundle to your friends.
>>>>> - They unzip it and run it to install Freenet.
>>>>> - The bundle includes your noderef.
>>>>> - It also includes a one-time key that allows the node to automatically
>>>>>   connect to yours despite yours not having their noderef yet.
>>>>>
>>>>> There are two big problems with this:
>>>>>
>>>>> 1) Everyone and his dog is behind a NAT. This means in order to connect
>>>>> you must have already exchanged references, full stop. THIS SUCKS. It
>>>>> also affects connectivity for newbies in a bad way (which is important
>>>>> IMHO).
>>>>>
>>>>> 2) Generally people shouldn't run programs that they receive in emails!
>>>>>
>>>>> Solution to the first one - and to newbie connectivity issues - is to
>>>>> implement UP&P and hope that routers implement it properly in future -
>>>>> is this a realistic hope?
>>>>>
>>>>> Solution to the second one is to just send the noderef and a link to the
>>>>> website, and only use full bundles when e.g. giving somebody a CD-R
>>>>> (which we should make really easy).
>>>>>
>>>>> Plugins for e.g. IRC clients, IM clients, have been suggested but I'm
>>>>> not sure how well this would work for newbies, and in any case I set up
>>>>> a darknet-tools list for people to talk about this and nobody has even
>>>>> talked about it since a few days after it was set up, let alone done
>>>>> anything.
>>>>>
>>>>> <_ph00> so the basic problem is "how to safely exchage refs", and the
>>>>> solution "eliminate ref exchanging by implementing opennet"?!? Am I the
>>>>> only one to think that's very stupid?
>>>>>         
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl



From Colin at sq7.org  Mon Mar  5 20:06:35 2007
From: Colin at sq7.org (Colin Davis)
Date: Mon, 05 Mar 2007 15:06:35 -0500
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <45EC7501.80509@david.sowder.com>
References: <20070305174545.GA29567@amphibian.dyndns.org>	<45EC5B42.1010701@sq7.org>	<200703051809.41642.dbkr@freenetproject.org>	<45EC5DD7.6080302@sq7.org>
	<45EC7501.80509@david.sowder.com>
Message-ID: <45EC784B.3020908@sq7.org>

I have no problem with an installer, that include your noderef- In an 
ideal world, that'd be the primary way people Get freenet. If I recall, 
.5 created a custom install, but almost no one ever used it.

I think the passphrase solution works well enough, if it's locked down..


As for NAT issues, forgive my being out of touch, but I thought that was 
solved a year ago.. If I recall, I thought that even if Side-A and 
Side-B were both firewalled, they could connect..

When Side-A added the noderef of B, it would start sending out packets 
to B, knowing that they won't get returned.. But the Sending of these 
packets would open a NAT-hole..
Then, Side-B would get around to adding the noderef of A, and start 
sending packets to A, opening a NAT-hole on their own side.. Side A's 
packets then get in through this hole, just as B's now arrive at A.

What am I missing?

-Colin


That said,
David Sowder (Zothar) wrote:
> Colin Davis wrote:
>> It solves #2- Don't run things you get in e-mail..
>>
>> Instead of requiring a Noderef, allow someone to connect with just a 
>> password, and the IP address. This is something you can TELL someone, or 
>> say in an IM, no file transfer required.
>>   
> I don't think we should necessarily categorically discount a fred-based
> installer distribution servlet because of NAT problems.  What if we
> allow decoupling the installer from the noderef?  Then those that can
> overcome or don't have the NAT problem can host the installer.  If I,
> for some reason, cannot host the installer, perhaps one of my peers is
> willing to share their installer hosting such that I could get "access
> keys" from his node to pass out to my friends and include my noderef
> from the distribution area of FProxy in the email I send to my friend
> with the "access key".  My friend saves the attached noderef (which was
> attached as a file with a .fref extension) to a file, connects to the my
> peer's installer host, uses the "access key", which will probably be
> part of the URL, and downloads and installs the node.  Then my friend
> can double click on the .fref file he downloaded, which includes a
> one-time code generated by the distribution area of my FProxy (or FCP
> server) that my node uses to authenticate the addition of a node I
> didn't already have the noderef of.  Node installed in a decentralized
> way.  Peer connection created.
> 
> Pass phrases could work, but I think they should have something like a
> 32 character minimum length.  (I wonder if there are passphrase
> dictionaries yet.)
>> Dave Baker wrote:
>>   
>>> On Monday 05 March 2007 18:02:42 Colin Davis wrote:
>>>     
>>>> I know it's less secure, but what about simply allowing people to
>>>> connect to your machine if they know a passphrase? The passphrase would
>>>> take the place of the Key, but be user-settable, and short.
>>>>       
>>> That doesn't solve either problem though, surely?
>>>
>>> my 2p on #freenet:
>>>
>>> [17:48] <dbkr> as far as both-way-adding goes, I think that's where we reach a 
>>> tradeoff with security, which is one of the main challanges for Freenet.
>>> [17:49] <dbkr> I'm not convinced the whole difficulty of exchanging refs isn't 
>>> a red herring - everyone can handle emailing a file.
>>>
>>> I'm definately in favour of the ability to burn a CD with an installer on it 
>>> that installs a node with your reference pre-bundled, although I think 
>>> leaving the installer out for an emailed-version means it's nothing the user 
>>> couldn't do themselves.
>>>
>>>
>>> Dave
>>>
>>>
>>>     
>>>> If that were in place, you could send an e-mail saying:
>>>>
>>>> Hey Jon, I just found this cool new thing called freenet, which lets you
>>>> get to all sorts of sites which aren't on the normal web! It's
>>>> anonymous, and free, you should check it out. It works by connecting
>>>> through each other's computers, but I'll let you connect to me to get
>>>> started.
>>>>
>>>> Go to FreenetProject.org and download it, then give it my hostname,
>>>> which is XXXXXXX and give it the connection passphrase "IamNotEvil".
>>>>
>>>> Don't give anyone else that information, or it won't work. It'll only
>>>> allow one connection.. After your up, you can connect to other friends,
>>>> and everyone's connection gets faster.
>>>>
>>>> I'm on IM if you want to talk about it.
>>>> -Person you Know.
>>>>
>>>> Matthew Toseland wrote:
>>>>       
>>>>> We will only get a darknet if it is really easy to swap references with
>>>>> your friends - opennet or no opennet.
>>>>>
>>>>> The original idea for Freenet 0.7 reference swapping was that you:
>>>>> - Go to your node, and ask it to create a bundle.
>>>>> - Send the bundle to your friends.
>>>>> - They unzip it and run it to install Freenet.
>>>>> - The bundle includes your noderef.
>>>>> - It also includes a one-time key that allows the node to automatically
>>>>>   connect to yours despite yours not having their noderef yet.
>>>>>
>>>>> There are two big problems with this:
>>>>>
>>>>> 1) Everyone and his dog is behind a NAT. This means in order to connect
>>>>> you must have already exchanged references, full stop. THIS SUCKS. It
>>>>> also affects connectivity for newbies in a bad way (which is important
>>>>> IMHO).
>>>>>
>>>>> 2) Generally people shouldn't run programs that they receive in emails!
>>>>>
>>>>> Solution to the first one - and to newbie connectivity issues - is to
>>>>> implement UP&P and hope that routers implement it properly in future -
>>>>> is this a realistic hope?
>>>>>
>>>>> Solution to the second one is to just send the noderef and a link to the
>>>>> website, and only use full bundles when e.g. giving somebody a CD-R
>>>>> (which we should make really easy).
>>>>>
>>>>> Plugins for e.g. IRC clients, IM clients, have been suggested but I'm
>>>>> not sure how well this would work for newbies, and in any case I set up
>>>>> a darknet-tools list for people to talk about this and nobody has even
>>>>> talked about it since a few days after it was set up, let alone done
>>>>> anything.
>>>>>
>>>>> <_ph00> so the basic problem is "how to safely exchage refs", and the
>>>>> solution "eliminate ref exchanging by implementing opennet"?!? Am I the
>>>>> only one to think that's very stupid?
>>>>>         
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl



From nextgens at freenetproject.org  Mon Mar  5 20:31:31 2007
From: nextgens at freenetproject.org (NextGen$)
Date: Mon, 5 Mar 2007 20:31:31 +0000
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <45EC784B.3020908@sq7.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<45EC5B42.1010701@sq7.org>
	<200703051809.41642.dbkr@freenetproject.org>
	<45EC5DD7.6080302@sq7.org> <45EC7501.80509@david.sowder.com>
	<45EC784B.3020908@sq7.org>
Message-ID: <20070305203131.GF937@freenetproject.org>

* Colin Davis <Colin at sq7.org> [2007-03-05 15:06:35]:

> I have no problem with an installer, that include your noderef- In an 
> ideal world, that'd be the primary way people Get freenet. If I recall, 
> .5 created a custom install, but almost no one ever used it.
> 
> I think the passphrase solution works well enough, if it's locked down..
> 
> 
> As for NAT issues, forgive my being out of touch, but I thought that was 
> solved a year ago.. If I recall, I thought that even if Side-A and 
> Side-B were both firewalled, they could connect..
> 
> When Side-A added the noderef of B, it would start sending out packets 
> to B, knowing that they won't get returned.. But the Sending of these 
> packets would open a NAT-hole..
> Then, Side-B would get around to adding the noderef of A, and start 
> sending packets to A, opening a NAT-hole on their own side.. Side A's 
> packets then get in through this hole, just as B's now arrive at A.
> 
> What am I missing?
> 
> -Colin


Both of you are missing the important point : on darknet we don't want to
reveal the network topology : we can't rely on a third party to do the job.
As far as I know all the NAT circumventing methods are involving a third
party.


Please move the thread to @tech :)

NextGen$
> 
> 
> That said,
> David Sowder (Zothar) wrote:
> > Colin Davis wrote:
> >> It solves #2- Don't run things you get in e-mail..
> >>
> >> Instead of requiring a Noderef, allow someone to connect with just a 
> >> password, and the IP address. This is something you can TELL someone, or 
> >> say in an IM, no file transfer required.
> >>   
> > I don't think we should necessarily categorically discount a fred-based
> > installer distribution servlet because of NAT problems.  What if we
> > allow decoupling the installer from the noderef?  Then those that can
> > overcome or don't have the NAT problem can host the installer.  If I,
> > for some reason, cannot host the installer, perhaps one of my peers is
> > willing to share their installer hosting such that I could get "access
> > keys" from his node to pass out to my friends and include my noderef
> > from the distribution area of FProxy in the email I send to my friend
> > with the "access key".  My friend saves the attached noderef (which was
> > attached as a file with a .fref extension) to a file, connects to the my
> > peer's installer host, uses the "access key", which will probably be
> > part of the URL, and downloads and installs the node.  Then my friend
> > can double click on the .fref file he downloaded, which includes a
> > one-time code generated by the distribution area of my FProxy (or FCP
> > server) that my node uses to authenticate the addition of a node I
> > didn't already have the noderef of.  Node installed in a decentralized
> > way.  Peer connection created.
> > 
> > Pass phrases could work, but I think they should have something like a
> > 32 character minimum length.  (I wonder if there are passphrase
> > dictionaries yet.)
> >> Dave Baker wrote:
> >>   
> >>> On Monday 05 March 2007 18:02:42 Colin Davis wrote:
> >>>     
> >>>> I know it's less secure, but what about simply allowing people to
> >>>> connect to your machine if they know a passphrase? The passphrase would
> >>>> take the place of the Key, but be user-settable, and short.
> >>>>       
> >>> That doesn't solve either problem though, surely?
> >>>
> >>> my 2p on #freenet:
> >>>
> >>> [17:48] <dbkr> as far as both-way-adding goes, I think that's where we reach a 
> >>> tradeoff with security, which is one of the main challanges for Freenet.
> >>> [17:49] <dbkr> I'm not convinced the whole difficulty of exchanging refs isn't 
> >>> a red herring - everyone can handle emailing a file.
> >>>
> >>> I'm definately in favour of the ability to burn a CD with an installer on it 
> >>> that installs a node with your reference pre-bundled, although I think 
> >>> leaving the installer out for an emailed-version means it's nothing the user 
> >>> couldn't do themselves.
> >>>
> >>>
> >>> Dave
> >>>
> >>>
> >>>     
> >>>> If that were in place, you could send an e-mail saying:
> >>>>
> >>>> Hey Jon, I just found this cool new thing called freenet, which lets you
> >>>> get to all sorts of sites which aren't on the normal web! It's
> >>>> anonymous, and free, you should check it out. It works by connecting
> >>>> through each other's computers, but I'll let you connect to me to get
> >>>> started.
> >>>>
> >>>> Go to FreenetProject.org and download it, then give it my hostname,
> >>>> which is XXXXXXX and give it the connection passphrase "IamNotEvil".
> >>>>
> >>>> Don't give anyone else that information, or it won't work. It'll only
> >>>> allow one connection.. After your up, you can connect to other friends,
> >>>> and everyone's connection gets faster.
> >>>>
> >>>> I'm on IM if you want to talk about it.
> >>>> -Person you Know.
> >>>>
> >>>> Matthew Toseland wrote:
> >>>>       
> >>>>> We will only get a darknet if it is really easy to swap references with
> >>>>> your friends - opennet or no opennet.
> >>>>>
> >>>>> The original idea for Freenet 0.7 reference swapping was that you:
> >>>>> - Go to your node, and ask it to create a bundle.
> >>>>> - Send the bundle to your friends.
> >>>>> - They unzip it and run it to install Freenet.
> >>>>> - The bundle includes your noderef.
> >>>>> - It also includes a one-time key that allows the node to automatically
> >>>>>   connect to yours despite yours not having their noderef yet.
> >>>>>
> >>>>> There are two big problems with this:
> >>>>>
> >>>>> 1) Everyone and his dog is behind a NAT. This means in order to connect
> >>>>> you must have already exchanged references, full stop. THIS SUCKS. It
> >>>>> also affects connectivity for newbies in a bad way (which is important
> >>>>> IMHO).
> >>>>>
> >>>>> 2) Generally people shouldn't run programs that they receive in emails!
> >>>>>
> >>>>> Solution to the first one - and to newbie connectivity issues - is to
> >>>>> implement UP&P and hope that routers implement it properly in future -
> >>>>> is this a realistic hope?
> >>>>>
> >>>>> Solution to the second one is to just send the noderef and a link to the
> >>>>> website, and only use full bundles when e.g. giving somebody a CD-R
> >>>>> (which we should make really easy).
> >>>>>
> >>>>> Plugins for e.g. IRC clients, IM clients, have been suggested but I'm
> >>>>> not sure how well this would work for newbies, and in any case I set up
> >>>>> a darknet-tools list for people to talk about this and nobody has even
> >>>>> talked about it since a few days after it was set up, let alone done
> >>>>> anything.
> >>>>>
> >>>>> <_ph00> so the basic problem is "how to safely exchage refs", and the
> >>>>> solution "eliminate ref exchanging by implementing opennet"?!? Am I the
> >>>>> only one to think that's very stupid?
> >>>>>         
> > _______________________________________________
> > Devl mailing list
> > Devl at freenetproject.org
> > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> 
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl


From toad at amphibian.dyndns.org  Mon Mar  5 22:36:43 2007
From: toad at amphibian.dyndns.org (Matthew Toseland)
Date: Mon, 5 Mar 2007 22:36:43 +0000
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <45EC5DD7.6080302@sq7.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<45EC5B42.1010701@sq7.org>
	<200703051809.41642.dbkr@freenetproject.org>
	<45EC5DD7.6080302@sq7.org>
Message-ID: <20070305223643.GA12373@amphibian.dyndns.org>

I don't understand why a password and IP address is easier than a
one-time reference. I suppose it has the advantage of being able to
write it down - but for it to be secure it would need to be a one-time
password; you'd need to generate a new one every time ...

Hmmm. Maybe we should provide both mechanisms?

On Mon, Mar 05, 2007 at 01:13:43PM -0500, Colin Davis wrote:
> It solves #2- Don't run things you get in e-mail..
> 
> Instead of requiring a Noderef, allow someone to connect with just a 
> password, and the IP address. This is something you can TELL someone, or 
> say in an IM, no file transfer required.
> 
> Dave Baker wrote:
> > On Monday 05 March 2007 18:02:42 Colin Davis wrote:
> >> I know it's less secure, but what about simply allowing people to
> >> connect to your machine if they know a passphrase? The passphrase would
> >> take the place of the Key, but be user-settable, and short.
> > 
> > That doesn't solve either problem though, surely?
> > 
> > my 2p on #freenet:
> > 
> > [17:48] <dbkr> as far as both-way-adding goes, I think that's where we reach a 
> > tradeoff with security, which is one of the main challanges for Freenet.
> > [17:49] <dbkr> I'm not convinced the whole difficulty of exchanging refs isn't 
> > a red herring - everyone can handle emailing a file.
> > 
> > I'm definately in favour of the ability to burn a CD with an installer on it 
> > that installs a node with your reference pre-bundled, although I think 
> > leaving the installer out for an emailed-version means it's nothing the user 
> > couldn't do themselves.
> > 
> > 
> > Dave
> > 
> > 
> >>
> >>
> >> If that were in place, you could send an e-mail saying:
> >>
> >> Hey Jon, I just found this cool new thing called freenet, which lets you
> >> get to all sorts of sites which aren't on the normal web! It's
> >> anonymous, and free, you should check it out. It works by connecting
> >> through each other's computers, but I'll let you connect to me to get
> >> started.
> >>
> >> Go to FreenetProject.org and download it, then give it my hostname,
> >> which is XXXXXXX and give it the connection passphrase "IamNotEvil".
> >>
> >> Don't give anyone else that information, or it won't work. It'll only
> >> allow one connection.. After your up, you can connect to other friends,
> >> and everyone's connection gets faster.
> >>
> >> I'm on IM if you want to talk about it.
> >> -Person you Know.
> >>
> >> Matthew Toseland wrote:
> >>> We will only get a darknet if it is really easy to swap references with
> >>> your friends - opennet or no opennet.
> >>>
> >>> The original idea for Freenet 0.7 reference swapping was that you:
> >>> - Go to your node, and ask it to create a bundle.
> >>> - Send the bundle to your friends.
> >>> - They unzip it and run it to install Freenet.
> >>> - The bundle includes your noderef.
> >>> - It also includes a one-time key that allows the node to automatically
> >>>   connect to yours despite yours not having their noderef yet.
> >>>
> >>> There are two big problems with this:
> >>>
> >>> 1) Everyone and his dog is behind a NAT. This means in order to connect
> >>> you must have already exchanged references, full stop. THIS SUCKS. It
> >>> also affects connectivity for newbies in a bad way (which is important
> >>> IMHO).
> >>>
> >>> 2) Generally people shouldn't run programs that they receive in emails!
> >>>
> >>> Solution to the first one - and to newbie connectivity issues - is to
> >>> implement UP&P and hope that routers implement it properly in future -
> >>> is this a realistic hope?
> >>>
> >>> Solution to the second one is to just send the noderef and a link to the
> >>> website, and only use full bundles when e.g. giving somebody a CD-R
> >>> (which we should make really easy).
> >>>
> >>> Plugins for e.g. IRC clients, IM clients, have been suggested but I'm
> >>> not sure how well this would work for newbies, and in any case I set up
> >>> a darknet-tools list for people to talk about this and nobody has even
> >>> talked about it since a few days after it was set up, let alone done
> >>> anything.
> >>>
> >>> <_ph00> so the basic problem is "how to safely exchage refs", and the
> >>> solution "eliminate ref exchanging by implementing opennet"?!? Am I the
> >>> only one to think that's very stupid?
> >>>
> >>>
> >>> ------------------------------------------------------------------------
> >>>
> >>> _______________________________________________
> >>> Devl mailing list
> >>> Devl at freenetproject.org
> >>> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> >> _______________________________________________
> >> Devl mailing list
> >> Devl at freenetproject.org
> >> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> > 
> > 
> > _______________________________________________
> > Devl mailing list
> > Devl at freenetproject.org
> > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> 
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> 
-------------- 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/devl/attachments/20070305/f6fc4ba8/attachment.pgp 

From toad at amphibian.dyndns.org  Mon Mar  5 22:41:37 2007
From: toad at amphibian.dyndns.org (Matthew Toseland)
Date: Mon, 5 Mar 2007 22:41:37 +0000
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <45EC7501.80509@david.sowder.com>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<45EC5B42.1010701@sq7.org>
	<200703051809.41642.dbkr@freenetproject.org>
	<45EC5DD7.6080302@sq7.org> <45EC7501.80509@david.sowder.com>
Message-ID: <20070305224137.GB12373@amphibian.dyndns.org>

On Mon, Mar 05, 2007 at 01:52:33PM -0600, David Sowder (Zothar) wrote:
> Colin Davis wrote:
> > It solves #2- Don't run things you get in e-mail..
> >
> > Instead of requiring a Noderef, allow someone to connect with just a 
> > password, and the IP address. This is something you can TELL someone, or 
> > say in an IM, no file transfer required.
> >   
> I don't think we should necessarily categorically discount a fred-based
> installer distribution servlet because of NAT problems.  What if we
> allow decoupling the installer from the noderef?  Then those that can
> overcome or don't have the NAT problem can host the installer. 

They are a sufficiently small fraction of the overall target audience as
to be uninteresting. UP&P would make them a large enough group to be
interesting, although still only something like 50% because UP&P is
extremely unreliable from what I have heard.

> If I,
> for some reason, cannot host the installer, perhaps one of my peers is
> willing to share their installer hosting such that I could get "access
> keys" from his node to pass out to my friends and include my noderef
> from the distribution area of FProxy in the email I send to my friend
> with the "access key".  My friend saves the attached noderef (which was
> attached as a file with a .fref extension) to a file, connects to the my
> peer's installer host, uses the "access key", which will probably be
> part of the URL, and downloads and installs the node.  Then my friend
> can double click on the .fref file he downloaded, which includes a
> one-time code generated by the distribution area of my FProxy (or FCP
> server) that my node uses to authenticate the addition of a node I
> didn't already have the noderef of.  Node installed in a decentralized
> way.  Peer connection created.

This is not very clear. Please divide into what friend A does and what
friend B does.
> 
> Pass phrases could work, but I think they should have something like a
> 32 character minimum length.  (I wonder if there are passphrase
> dictionaries yet.)

Passphrases from movies or well-known-people are probably in a big list
somewhere.
-------------- 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/devl/attachments/20070305/f17542c5/attachment.pgp 

From toad at amphibian.dyndns.org  Mon Mar  5 22:47:00 2007
From: toad at amphibian.dyndns.org (Matthew Toseland)
Date: Mon, 5 Mar 2007 22:47:00 +0000
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <45EC784B.3020908@sq7.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<45EC5B42.1010701@sq7.org>
	<200703051809.41642.dbkr@freenetproject.org>
	<45EC5DD7.6080302@sq7.org> <45EC7501.80509@david.sowder.com>
	<45EC784B.3020908@sq7.org>
Message-ID: <20070305224700.GC12373@amphibian.dyndns.org>

On Mon, Mar 05, 2007 at 03:06:35PM -0500, Colin Davis wrote:
> As for NAT issues, forgive my being out of touch, but I thought that was 
> solved a year ago.. If I recall, I thought that even if Side-A and 
> Side-B were both firewalled, they could connect..
> 
> When Side-A added the noderef of B, it would start sending out packets 
> to B, knowing that they won't get returned.. But the Sending of these 
> packets would open a NAT-hole..
> Then, Side-B would get around to adding the noderef of A, and start 
> sending packets to A, opening a NAT-hole on their own side.. Side A's 
> packets then get in through this hole, just as B's now arrive at A.

Only if both A and B know the other's IP address.

That's not easy. Most people don't know their IP address.

There are several issues here:
1. How to deal with NATs. I'm leaning strongly towards implementing
UP&P. This might even remove the need for using central STUN servers -
at least most of the time after installation.
2. How to easily give somebody a node reference that Just Works, without
them having to do anything beyond double clicking on the file. Two
options: 1) IP:port + password (password may be auto-generated, and
probably should be one-time), 2) one-time references.
3. How to get the node source code in the first place, and whether it's
a good idea for this to be distributed; if it is, we can include the
initial noderef. (It probably is, but we need #2 as well for e.g. if the
user already has a node or has other safe means of obtaining fred).
> 
> What am I missing?
-------------- 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/devl/attachments/20070305/710deead/attachment.pgp 

From toad at amphibian.dyndns.org  Mon Mar  5 22:48:17 2007
From: toad at amphibian.dyndns.org (Matthew Toseland)
Date: Mon, 5 Mar 2007 22:48:17 +0000
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <20070305203131.GF937@freenetproject.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<45EC5B42.1010701@sq7.org>
	<200703051809.41642.dbkr@freenetproject.org>
	<45EC5DD7.6080302@sq7.org> <45EC7501.80509@david.sowder.com>
	<45EC784B.3020908@sq7.org>
	<20070305203131.GF937@freenetproject.org>
Message-ID: <20070305224817.GD12373@amphibian.dyndns.org>

On Mon, Mar 05, 2007 at 08:31:31PM +0000, NextGen$ wrote:
> 
> Both of you are missing the important point : on darknet we don't want to
> reveal the network topology : we can't rely on a third party to do the job.
> As far as I know all the NAT circumventing methods are involving a third
> party.

UP&P doesn't. As long as we are sure we are behind a router, we can be
sure that we are on our own LAN, unless Mallory controls the router.
> 
> Please move the thread to @tech :)

IMHO this is relevant in the relatively near future.
> 
> NextGen$
-------------- 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/devl/attachments/20070305/2c6cab5a/attachment.pgp 

From carlin at jlab.org  Mon Mar  5 18:02:24 2007
From: carlin at jlab.org (Chris Carlin)
Date: Mon, 05 Mar 2007 13:02:24 -0500
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <20070305174545.GA29567@amphibian.dyndns.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
Message-ID: <45EC5B30.8090508@jlab.org>

Matthew Toseland wrote:

> There are two big problems with this:
> 
> 1) Everyone and his dog is behind a NAT. This means in order to connect
> you must have already exchanged references, full stop. THIS SUCKS. It
> also affects connectivity for newbies in a bad way (which is important
> IMHO).

Will the IPv6 stuff in Vista will affect this? I'm under the impression 
that Vista has some sort of IPv6 tunneling enabled by default; does it 
defeat the evils of NAT?

Just an idle curiosity...

~Chris


From toad at amphibian.dyndns.org  Tue Mar  6 03:03:01 2007
From: toad at amphibian.dyndns.org (Matthew Toseland)
Date: Tue, 6 Mar 2007 03:03:01 +0000
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <45EC5B30.8090508@jlab.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<45EC5B30.8090508@jlab.org>
Message-ID: <20070306030301.GA5773@amphibian.dyndns.org>

On Mon, Mar 05, 2007 at 01:02:24PM -0500, Chris Carlin wrote:
> Matthew Toseland wrote:
> 
> > There are two big problems with this:
> > 
> > 1) Everyone and his dog is behind a NAT. This means in order to connect
> > you must have already exchanged references, full stop. THIS SUCKS. It
> > also affects connectivity for newbies in a bad way (which is important
> > IMHO).
> 
> Will the IPv6 stuff in Vista will affect this? I'm under the impression 
> that Vista has some sort of IPv6 tunneling enabled by default; does it 
> defeat the evils of NAT?

I'm not capable of providing a civil answer to this question right now.
> 
> Just an idle curiosity...
> 
> ~Chris
-------------- 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/devl/attachments/20070306/e138bc5e/attachment.pgp 

From toad at amphibian.dyndns.org  Tue Mar  6 03:11:37 2007
From: toad at amphibian.dyndns.org (Matthew Toseland)
Date: Tue, 6 Mar 2007 03:11:37 +0000
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <20070306030301.GA5773@amphibian.dyndns.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<45EC5B30.8090508@jlab.org>
	<20070306030301.GA5773@amphibian.dyndns.org>
Message-ID: <20070306031137.GA1004@amphibian.dyndns.org>

On Tue, Mar 06, 2007 at 03:03:01AM +0000, Matthew Toseland wrote:
> On Mon, Mar 05, 2007 at 01:02:24PM -0500, Chris Carlin wrote:
> > 
> > Will the IPv6 stuff in Vista will affect this? I'm under the impression 
> > that Vista has some sort of IPv6 tunneling enabled by default; does it 
> > defeat the evils of NAT?
> 
> I'm not capable of providing a civil answer to this question right now.

Sorry, I would like to know; tell me if you find out.
> > 
> > Just an idle curiosity...
-------------- 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/devl/attachments/20070306/1d7517cb/attachment.pgp 

From Volodya at WhenGendarmeSleeps.org  Tue Mar  6 05:44:35 2007
From: Volodya at WhenGendarmeSleeps.org (Volodya)
Date: Tue, 06 Mar 2007 05:44:35 +0000
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <20070305223643.GA12373@amphibian.dyndns.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>	<45EC5B42.1010701@sq7.org>	<200703051809.41642.dbkr@freenetproject.org>	<45EC5DD7.6080302@sq7.org>
	<20070305223643.GA12373@amphibian.dyndns.org>
Message-ID: <45ECFFC3.7030007@WhenGendarmeSleeps.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Matthew Toseland wrote:
> I don't understand why a password and IP address is easier than a
> one-time reference. I suppose it has the advantage of being able to
> write it down - but for it to be secure it would need to be a one-time
> password; you'd need to generate a new one every time ...
> 
> Hmmm. Maybe we should provide both mechanisms?

One thing that might be done is not having an increadibly secure password protection (just
secure enough), but when somebody adds themselves via password they get added in the
disabled mode, then the person tells you "It asks me to tell you to enable me" and you do
so. If somebody intersepts the password in between and uses it, the second person will get
a request to inform you that password has been used already, so you just go and delete the
bugger who used it.

In other words: Bring security away from the machine and to the person.

                              - Volodya

- --
http://freedom.libsyn.com/       Voice of Freedom, Radical Podcast
http://freeselfdefence.info/     Self-defence wiki
http://www.kingstonstudents.org/ Kingston University students' forum

 "None of us are free until all of us are free."    ~ Mihail Bakunin
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)

iD8DBQFF7P/DuWy2EFICg+0RAtblAJ4vSyPViz4+lbcLO6n3FPA6b5deIwCguCcR
O8vOSZ8JrCwTaq565LAZfP0=
=5jwq
-----END PGP SIGNATURE-----


From the_bishop at web.de  Tue Mar  6 06:08:51 2007
From: the_bishop at web.de (Martin Scheffler)
Date: Tue, 6 Mar 2007 07:08:51 +0100
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <45ECFFC3.7030007@WhenGendarmeSleeps.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<20070305223643.GA12373@amphibian.dyndns.org>
	<45ECFFC3.7030007@WhenGendarmeSleeps.org>
Message-ID: <200703060708.56888.the_bishop@web.de>

Am Dienstag, 6. M?rz 2007 06:44 schrieb Volodya:
> One thing that might be done is not having an increadibly secure
> password protection (just secure enough), but when somebody adds
> themselves via password they get added in the disabled mode, then the
> person tells you "It asks me to tell you to enable me" and you do so.
> If somebody intersepts the password in between and uses it, the second
> person will get a request to inform you that password has been used
> already, so you just go and delete the bugger who used it.
>
> In other words: Bring security away from the machine and to the person.

Yes, sounds reasonable at first.
But how to distinguish MITM and the authorized peer? Maybe by being able 
to exchange p2p-messages (while in restricted mode) to verify responses 
in-band which can't be used by the eavesdropper. Is this possible?

if you are still talking to the peer person out-of-band after untrusted 
initial connect, you could verify the node key fingerprints to be sure, 
but everything in-band shall be distrusted at this point.

good byte
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
Url : http://emu.freenetproject.org/pipermail/devl/attachments/20070306/6af3d040/attachment.pgp 

From the_bishop at web.de  Tue Mar  6 06:12:48 2007
From: the_bishop at web.de (Martin Scheffler)
Date: Tue, 6 Mar 2007 07:12:48 +0100
Subject: [freenet-dev] what about the automatic updater?
Message-ID: <200703060712.50393.the_bishop@web.de>

does the updated code gets uploaded to freenet this time?
i had no luck updating the node by itself so far...

good byte
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
Url : http://emu.freenetproject.org/pipermail/devl/attachments/20070306/ba84b248/attachment.pgp 

From Volodya at WhenGendarmeSleeps.org  Tue Mar  6 06:18:40 2007
From: Volodya at WhenGendarmeSleeps.org (Volodya)
Date: Tue, 06 Mar 2007 06:18:40 +0000
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <200703060708.56888.the_bishop@web.de>
References: <20070305174545.GA29567@amphibian.dyndns.org>	<20070305223643.GA12373@amphibian.dyndns.org>	<45ECFFC3.7030007@WhenGendarmeSleeps.org>
	<200703060708.56888.the_bishop@web.de>
Message-ID: <45ED07C0.2050901@WhenGendarmeSleeps.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Martin Scheffler wrote:
> Am Dienstag, 6. M?rz 2007 06:44 schrieb Volodya:
>> One thing that might be done is not having an increadibly secure
>> password protection (just secure enough), but when somebody adds
>> themselves via password they get added in the disabled mode, then the
>> person tells you "It asks me to tell you to enable me" and you do so.
>> If somebody intersepts the password in between and uses it, the second
>> person will get a request to inform you that password has been used
>> already, so you just go and delete the bugger who used it.
>>
>> In other words: Bring security away from the machine and to the person.
> 
> Yes, sounds reasonable at first.
> But how to distinguish MITM and the authorized peer?

'Authorised peer' will tell you that an was unable to connect, and then you know that
somebody intercepted the password.

Like i said it is *still* a 1 time password, meaning that if real user typed it the
intruder won't be able to use it, so intruder must do it before the real peer does, which
will raise the alarm since that peer is your friend and you will be immediately informed
that 'pass doesn't work, mate'.

                       - Volodya

- --
http://freedom.libsyn.com/       Voice of Freedom, Radical Podcast
http://freeselfdefence.info/     Self-defence wiki
http://www.kingstonstudents.org/ Kingston University students' forum

 "None of us are free until all of us are free."    ~ Mihail Bakunin
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)

iD8DBQFF7Qe/uWy2EFICg+0RAlFjAKDkypEG+pzTe+4ag4xxyqD+oRhjzACfZT5S
e7tga3oXIvonTaOT0hZb/L8=
=8zKA
-----END PGP SIGNATURE-----


From the_bishop at web.de  Tue Mar  6 06:39:51 2007
From: the_bishop at web.de (Martin Scheffler)
Date: Tue, 6 Mar 2007 07:39:51 +0100
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <45ED07C0.2050901@WhenGendarmeSleeps.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<200703060708.56888.the_bishop@web.de>
	<45ED07C0.2050901@WhenGendarmeSleeps.org>
Message-ID: <200703060739.52676.the_bishop@web.de>

Am Dienstag, 6. M?rz 2007 07:18 schrieb Volodya:
> 'Authorised peer' will tell you that an was unable to connect, and then
> you know that somebody intercepted the password.
If the MITM has the ability not only to read the IP+OTP messages, but to 
redirect the traffic from IP-A to IP-B through his fake node, then you 
can not distinguish MITM and peer.
out of band by phone, PGP or handwritten papermail would be the best proof 
in that case.

> Like i said it is *still* a 1 time password, meaning that if real user
> typed it the intruder won't be able to use it, so intruder must do it
> before the real peer does, which will raise the alarm since that peer
> is your friend and you will be immediately informed that 'pass doesn't
> work, mate'.
not really... the MITM could try to use the both passwords immediately and 
fake the both other sides.
If you dont have out-of-band means to verify the information from the peer 
node, you have a connection to the wrong node and only see the forged 
verification.

good byte
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
Url : http://emu.freenetproject.org/pipermail/devl/attachments/20070306/0c59cce4/attachment.pgp 

From nextgens at freenetproject.org  Tue Mar  6 10:30:51 2007
From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re_=28NextGen$=29?=)
Date: Tue, 6 Mar 2007 11:30:51 +0100
Subject: [freenet-dev] what about the automatic updater?
In-Reply-To: <200703060712.50393.the_bishop@web.de>
References: <200703060712.50393.the_bishop@web.de>
Message-ID: <20070306103051.GB4145@freenetproject.org>

* Martin Scheffler <the_bishop at web.de> [2007-03-06 07:12:48]:

> does the updated code gets uploaded to freenet this time?
> i had no luck updating the node by itself so far...
> 
> good byte

Toad hasn't announced the new build yet: it's likely he hasn't inserted
it.

NextGen$


From bbackde at googlemail.com  Tue Mar  6 10:35:17 2007
From: bbackde at googlemail.com (bbackde at googlemail.com)
Date: Tue, 6 Mar 2007 11:35:17 +0100
Subject: [freenet-dev] what about the automatic updater?
In-Reply-To: <20070306103051.GB4145@freenetproject.org>
References: <200703060712.50393.the_bishop@web.de>
	<20070306103051.GB4145@freenetproject.org>
Message-ID: <d6149a790703060235w5ce64b21udfa5e90c26e77813@mail.gmail.com>

But my node updated to 1018???!!!

On 3/6/07, Florent Daigni?re (NextGen$) <nextgens at freenetproject.org> wrote:
> * Martin Scheffler <the_bishop at web.de> [2007-03-06 07:12:48]:
>
> > does the updated code gets uploaded to freenet this time?
> > i had no luck updating the node by itself so far...
> >
> > good byte
>
> Toad hasn't announced the new build yet: it's likely he hasn't inserted
> it.
>
> NextGen$
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
>


From nextgens at freenetproject.org  Tue Mar  6 10:37:25 2007
From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re_=28NextGen$=29?=)
Date: Tue, 6 Mar 2007 11:37:25 +0100
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <45ECFFC3.7030007@WhenGendarmeSleeps.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<45EC5B42.1010701@sq7.org>
	<200703051809.41642.dbkr@freenetproject.org>
	<45EC5DD7.6080302@sq7.org>
	<20070305223643.GA12373@amphibian.dyndns.org>
	<45ECFFC3.7030007@WhenGendarmeSleeps.org>
Message-ID: <20070306103722.GC4145@freenetproject.org>

* Volodya <Volodya at WhenGendarmeSleeps.org> [2007-03-06 05:44:35]:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Matthew Toseland wrote:
> > I don't understand why a password and IP address is easier than a
> > one-time reference. I suppose it has the advantage of being able to
> > write it down - but for it to be secure it would need to be a one-time
> > password; you'd need to generate a new one every time ...
> > 
> > Hmmm. Maybe we should provide both mechanisms?
> 
> One thing that might be done is not having an increadibly secure password protection (just
> secure enough), but when somebody adds themselves via password they get added in the
> disabled mode, then the person tells you "It asks me to tell you to enable me" and you do
> so. If somebody intersepts the password in between and uses it, the second person will get
> a request to inform you that password has been used already, so you just go and delete the
> bugger who used it.
> 
> In other words: Bring security away from the machine and to the person.
> 
>                               - Volodya
> 

So far a node is *passive* and won't react upon reception of any unknown data.
If we want to tell the user that the password has already been used, we
would need to change that behaviour :/

I'm not sure it's a good idea.

NextGen$


From nextgens at freenetproject.org  Tue Mar  6 10:40:02 2007
From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re_=28NextGen$=29?=)
Date: Tue, 6 Mar 2007 11:40:02 +0100
Subject: [freenet-dev] what about the automatic updater?
In-Reply-To: <d6149a790703060235w5ce64b21udfa5e90c26e77813@mail.gmail.com>
References: <200703060712.50393.the_bishop@web.de>
	<20070306103051.GB4145@freenetproject.org>
	<d6149a790703060235w5ce64b21udfa5e90c26e77813@mail.gmail.com>
Message-ID: <20070306103959.GD4145@freenetproject.org>

* bbackde at googlemail.com <bbackde at googlemail.com> [2007-03-06 11:35:17]:

> But my node updated to 1018???!!!

Ok, then it's probably a timing issue : could you please wait a bit more
Martin and get back to us if it hasn't updated before the end of the day
?

Thanks.
PS: my node hasn't detected the update yet either.

> 
> On 3/6/07, Florent Daigni?re (NextGen$) <nextgens at freenetproject.org> wrote:
> >* Martin Scheffler <the_bishop at web.de> [2007-03-06 07:12:48]:
> >
> >> does the updated code gets uploaded to freenet this time?
> >> i had no luck updating the node by itself so far...
> >>
> >> good byte
> >
> >Toad hasn't announced the new build yet: it's likely he hasn't inserted
> >it.
> >
> >NextGen$
> >_______________________________________________
> >Devl mailing list
> >Devl at freenetproject.org
> >http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> >


From edt at aei.ca  Tue Mar  6 12:57:45 2007
From: edt at aei.ca (Ed Tomlinson)
Date: Tue, 6 Mar 2007 07:57:45 -0500
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <20070306103722.GC4145@freenetproject.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<45ECFFC3.7030007@WhenGendarmeSleeps.org>
	<20070306103722.GC4145@freenetproject.org>
Message-ID: <200703060757.45611.edt@aei.ca>

Hi,

Why not take the simple way?  The last time I had to reset my node I ran
a swapbot that used irc to get some refs.  If I remember correctly there is
a project that creates an anonymous irc network.  Why not have nodes
create or connect to this network, at least some of the time, and use a
standardized swapbot to get references.   If a person running a node is
willing to be less anonymous, we could even use freenode...  

Comments?
Ed


On Tuesday 06 March 2007 05:37, Florent Daigni?re wrote:
> * Volodya <Volodya at WhenGendarmeSleeps.org> [2007-03-06 05:44:35]:
> 
> > -----BEGIN PGP SIGNED MESSAGE-----
> > Hash: SHA1
> > 
> > Matthew Toseland wrote:
> > > I don't understand why a password and IP address is easier than a
> > > one-time reference. I suppose it has the advantage of being able to
> > > write it down - but for it to be secure it would need to be a one-time
> > > password; you'd need to generate a new one every time ...
> > > 
> > > Hmmm. Maybe we should provide both mechanisms?
> > 
> > One thing that might be done is not having an increadibly secure password protection (just
> > secure enough), but when somebody adds themselves via password they get added in the
> > disabled mode, then the person tells you "It asks me to tell you to enable me" and you do
> > so. If somebody intersepts the password in between and uses it, the second person will get
> > a request to inform you that password has been used already, so you just go and delete the
> > bugger who used it.
> > 
> > In other words: Bring security away from the machine and to the person.
> > 
> >                               - Volodya
> > 
> 
> So far a node is *passive* and won't react upon reception of any unknown data.
> If we want to tell the user that the password has already been used, we
> would need to change that behaviour :/
> 
> I'm not sure it's a good idea.
> 
> NextGen$
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> 


From nextgens at freenetproject.org  Tue Mar  6 13:35:14 2007
From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re_=28NextGen$=29?=)
Date: Tue, 6 Mar 2007 14:35:14 +0100
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <200703060757.45611.edt@aei.ca>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<45ECFFC3.7030007@WhenGendarmeSleeps.org>
	<20070306103722.GC4145@freenetproject.org>
	<200703060757.45611.edt@aei.ca>
Message-ID: <20070306133514.GI4145@freenetproject.org>

* Ed Tomlinson <edt at aei.ca> [2007-03-06 07:57:45]:

> Hi,
> 
> Why not take the simple way?  The last time I had to reset my node I ran
> a swapbot that used irc to get some refs.  If I remember correctly there is
> a project that creates an anonymous irc network.  Why not have nodes
> create or connect to this network, at least some of the time, and use a
> standardized swapbot to get references.   If a person running a node is
> willing to be less anonymous, we could even use freenode...  
> 
> Comments?
> Ed

That wouldn't create a small world topology...
FYI, we are not discussing how to implement opennet but what are opennet
alternatives.

NextGen$

> 
> 
> On Tuesday 06 March 2007 05:37, Florent Daigni?re wrote:
> > * Volodya <Volodya at WhenGendarmeSleeps.org> [2007-03-06 05:44:35]:
> > 
> > > -----BEGIN PGP SIGNED MESSAGE-----
> > > Hash: SHA1
> > > 
> > > Matthew Toseland wrote:
> > > > I don't understand why a password and IP address is easier than a
> > > > one-time reference. I suppose it has the advantage of being able to
> > > > write it down - but for it to be secure it would need to be a one-time
> > > > password; you'd need to generate a new one every time ...
> > > > 
> > > > Hmmm. Maybe we should provide both mechanisms?
> > > 
> > > One thing that might be done is not having an increadibly secure password protection (just
> > > secure enough), but when somebody adds themselves via password they get added in the
> > > disabled mode, then the person tells you "It asks me to tell you to enable me" and you do
> > > so. If somebody intersepts the password in between and uses it, the second person will get
> > > a request to inform you that password has been used already, so you just go and delete the
> > > bugger who used it.
> > > 
> > > In other words: Bring security away from the machine and to the person.
> > > 
> > >                               - Volodya
> > > 
> > 
> > So far a node is *passive* and won't react upon reception of any unknown data.
> > If we want to tell the user that the password has already been used, we
> > would need to change that behaviour :/
> > 
> > I'm not sure it's a good idea.
> > 
> > NextGen$
> > _______________________________________________
> > Devl mailing list
> > Devl at freenetproject.org
> > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> > 
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl


From ian at locut.us  Tue Mar  6 15:05:56 2007
From: ian at locut.us (Ian Clarke)
Date: Tue, 6 Mar 2007 09:05:56 -0600
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <20070305223643.GA12373@amphibian.dyndns.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<45EC5B42.1010701@sq7.org>
	<200703051809.41642.dbkr@freenetproject.org>
	<45EC5DD7.6080302@sq7.org>
	<20070305223643.GA12373@amphibian.dyndns.org>
Message-ID: <823242bd0703060705s280801d4o217062ddfcf12817@mail.gmail.com>

You don't understand why:

43.213.43.1:54533:foriemdhs

...is easier than...

identity=MZHvIrRsHRLqqTaEN-S5lLpeD8yk572zFfPFxsCsS-M
location=0.16978598221719554
testnet=false
myName=SanityMBP
lastGoodVersion=Fred,0.7,1.0,1009
sig=1fd90af28fddd96d0b159c7bd93658f9267b84dbacef75bc400074258eabe81e,433cac175114111647b4d1e0af84448c34620adb6accf307e14daf72960a8048
version=Fred,0.7,1.0,1009
dsaPubKey.y=Xw1RbImKwC5EYlfOZtvV0nUz-ujplXfW6IHq125LfajUe2m6sA~IHwvu75wtCvdp~TgjNlgddQMhVBtOkhvSIr1cdiWFYqSaeULGFDeFJdON8PZj1Apt7IN2xIVh17WUA99JoeZhCKUH9d-nKubllovx0qFyVpezkPaluWhQrOx0JCYEYJOmah5xIUlvdPfG7bkSWGnWDq5VfJAQTIW~XW2CBjtXh1jbsfOAE1l9k35JY44jNJ4Ehq-qlPLYO1gqpHE~HsikKxYFGX3FcByxRIYCTxVTSfO9bx6vAxzaBQnDIC6zthhzKJAezRJEM~YZ4zZA0q39~0Y8eKK1yIrSOQ
physical.udp=sanity1.dyndns.org:14505;76.167.194.83:14505
dsaGroup.g=UaRatnDByf0QvTlaaAXTMzn1Z15LDTXe-J~gOqXCv0zpz83CVngSkb--bVRuZ9R65OFg~ATKcuw8VJJwn1~A9p5jRt2NPj2EM7bu72O85-mFdBhcav8WHJtTbXb4cxNzZaQkbPQUv~gEnuEeMTc80KZVjilQ7wlTIM6GIY~ZJVHMKSIkEU87YBRtIt1R~BJcnaDAKBJv~oXv1PS-6iwQRFMynMEmipfpqDXBTkqaQ8ahiGWA41rY8d4jDhrzIgjvkzfxkkcCpFFOldwW8w8MEecUoRLuhKnY1sm8nnTjNlYLtc1Okeq-ba0mvwygSAf4wxovwY6n1Fuqt8yZe1PDVg
dsaGroup.q=ALFDNoq81R9Y1kQNVBc5kzmk0VvvCWosXY5t9E9S1tN5
dsaGroup.p=AIYIrE9VNhM38qPjirGGT-PJjWZBHY0q-JxSYyDFQfZQeOhrx4SUpdc~SppnWD~UHymT7WyX28eV3YjwkVyc~--H5Tc83hPjx8qQc7kQbrMb~CJy7QBX~YSocKGfioO-pwfRZEDDguYtOJBHPqeenVDErGsfHTCxDDKgL2hYM8Ynj8Kes0OcUzOIVhShFSGbOAjJKjeg82XNXmG1hhdh2tnv8M4jJQ9ViEj425Mrh6O9jXovfPmcdYIr3C~3waHXjQvPgUiK4N5Saf~FOri48fK-PmwFZFc-YSgI9o2-70nVybSnBXlM96QkzU6x4CYFUuZ7-B~je0ofeLdX7xhehuk
ark.pubURI=SSK at 1CA0TRYSSEnK~Tj7-OqVajmsqS85iUVhjmuz8EU9HQs,MuvO8pTCCPCTqUIMpkEXUBu55nV2DgiGjhSgr~9FY0I,AQABAAE/ark
ark.number=128
End

I think it speaks for itself.

Ian.

On 3/5/07, Matthew Toseland <toad at amphibian.dyndns.org> wrote:
> I don't understand why a password and IP address is easier than a
> one-time reference. I suppose it has the advantage of being able to
> write it down - but for it to be secure it would need to be a one-time
> password; you'd need to generate a new one every time ...
>
> Hmmm. Maybe we should provide both mechanisms?
>
> On Mon, Mar 05, 2007 at 01:13:43PM -0500, Colin Davis wrote:
> > It solves #2- Don't run things you get in e-mail..
> >
> > Instead of requiring a Noderef, allow someone to connect with just a
> > password, and the IP address. This is something you can TELL someone, or
> > say in an IM, no file transfer required.
> >
> > Dave Baker wrote:
> > > On Monday 05 March 2007 18:02:42 Colin Davis wrote:
> > >> I know it's less secure, but what about simply allowing people to
> > >> connect to your machine if they know a passphrase? The passphrase would
> > >> take the place of the Key, but be user-settable, and short.
> > >
> > > That doesn't solve either problem though, surely?
> > >
> > > my 2p on #freenet:
> > >
> > > [17:48] <dbkr> as far as both-way-adding goes, I think that's where we reach a
> > > tradeoff with security, which is one of the main challanges for Freenet.
> > > [17:49] <dbkr> I'm not convinced the whole difficulty of exchanging refs isn't
> > > a red herring - everyone can handle emailing a file.
> > >
> > > I'm definately in favour of the ability to burn a CD with an installer on it
> > > that installs a node with your reference pre-bundled, although I think
> > > leaving the installer out for an emailed-version means it's nothing the user
> > > couldn't do themselves.
> > >
> > >
> > > Dave
> > >
> > >
> > >>
> > >>
> > >> If that were in place, you could send an e-mail saying:
> > >>
> > >> Hey Jon, I just found this cool new thing called freenet, which lets you
> > >> get to all sorts of sites which aren't on the normal web! It's
> > >> anonymous, and free, you should check it out. It works by connecting
> > >> through each other's computers, but I'll let you connect to me to get
> > >> started.
> > >>
> > >> Go to FreenetProject.org and download it, then give it my hostname,
> > >> which is XXXXXXX and give it the connection passphrase "IamNotEvil".
> > >>
> > >> Don't give anyone else that information, or it won't work. It'll only
> > >> allow one connection.. After your up, you can connect to other friends,
> > >> and everyone's connection gets faster.
> > >>
> > >> I'm on IM if you want to talk about it.
> > >> -Person you Know.
> > >>
> > >> Matthew Toseland wrote:
> > >>> We will only get a darknet if it is really easy to swap references with
> > >>> your friends - opennet or no opennet.
> > >>>
> > >>> The original idea for Freenet 0.7 reference swapping was that you:
> > >>> - Go to your node, and ask it to create a bundle.
> > >>> - Send the bundle to your friends.
> > >>> - They unzip it and run it to install Freenet.
> > >>> - The bundle includes your noderef.
> > >>> - It also includes a one-time key that allows the node to automatically
> > >>>   connect to yours despite yours not having their noderef yet.
> > >>>
> > >>> There are two big problems with this:
> > >>>
> > >>> 1) Everyone and his dog is behind a NAT. This means in order to connect
> > >>> you must have already exchanged references, full stop. THIS SUCKS. It
> > >>> also affects connectivity for newbies in a bad way (which is important
> > >>> IMHO).
> > >>>
> > >>> 2) Generally people shouldn't run programs that they receive in emails!
> > >>>
> > >>> Solution to the first one - and to newbie connectivity issues - is to
> > >>> implement UP&P and hope that routers implement it properly in future -
> > >>> is this a realistic hope?
> > >>>
> > >>> Solution to the second one is to just send the noderef and a link to the
> > >>> website, and only use full bundles when e.g. giving somebody a CD-R
> > >>> (which we should make really easy).
> > >>>
> > >>> Plugins for e.g. IRC clients, IM clients, have been suggested but I'm
> > >>> not sure how well this would work for newbies, and in any case I set up
> > >>> a darknet-tools list for people to talk about this and nobody has even
> > >>> talked about it since a few days after it was set up, let alone done
> > >>> anything.
> > >>>
> > >>> <_ph00> so the basic problem is "how to safely exchage refs", and the
> > >>> solution "eliminate ref exchanging by implementing opennet"?!? Am I the
> > >>> only one to think that's very stupid?
> > >>>
> > >>>
> > >>> ------------------------------------------------------------------------
> > >>>
> > >>> _______________________________________________
> > >>> Devl mailing list
> > >>> Devl at freenetproject.org
> > >>> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> > >> _______________________________________________
> > >> Devl mailing list
> > >> Devl at freenetproject.org
> > >> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> > >
> > >
> > > _______________________________________________
> > > Devl mailing list
> > > Devl at freenetproject.org
> > > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> >
> > _______________________________________________
> > Devl mailing list
> > Devl at freenetproject.org
> > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> >
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.6 (GNU/Linux)
>
> iD8DBQFF7Jt7A9rUluQ9pFARAr7gAJ9iJHAur3dIsrClY+AmFSQgOy6bXgCfY2Po
> kqDOiGlYjY/lZ0v66rlsip0=
> =NcEY
> -----END PGP SIGNATURE-----
>
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
>


-- 
Founder and CEO, Thoof Inc
Email: ian at thoof.com
Office: +1 512 485 1970
Cell: +1 310 593 3724
AIM: ian.clarke at mac.com
Skype: sanity


From toad at amphibian.dyndns.org  Tue Mar  6 16:42:22 2007
From: toad at amphibian.dyndns.org (Matthew Toseland)
Date: Tue, 6 Mar 2007 16:42:22 +0000
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <823242bd0703060705s280801d4o217062ddfcf12817@mail.gmail.com>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<45EC5B42.1010701@sq7.org>
	<200703051809.41642.dbkr@freenetproject.org>
	<45EC5DD7.6080302@sq7.org>
	<20070305223643.GA12373@amphibian.dyndns.org>
	<823242bd0703060705s280801d4o217062ddfcf12817@mail.gmail.com>
Message-ID: <20070306164222.GA3992@amphibian.dyndns.org>

Well, if you are exchanging electronically, you are probably able to
move the latter as easily as the former, no?

Sadly no, because of the [ insert favourite expletive ] NATs which kill
DCC. But you can move a small file with IM clients, surely? It's moving
a file and then double clicking on it or even executing it in place,
versus copying a line of text manually.

So it would be easier for IRC and other stuff crippled by NATs. Would it
be easier for anything practically used by semi-geek (rather than
ubergeek) users?

Also you can write it down. However, because it would probably have to
be a one-time password (for security when dealing with newbies; if it's
a two way exchange, you could maybe have more permanent passwords, at a
cost in security), you can't remember it and write it down. And
even if it wasn't one-time, you probably can't write it down anyway if
you don't have a semistatic IP - but a lot of people *do* have
semistatic IPs...

Anyway, it would help in some cases, sure. So we should probably have
both.

On Tue, Mar 06, 2007 at 09:05:56AM -0600, Ian Clarke wrote:
> You don't understand why:
> 
> 43.213.43.1:54533:foriemdhs
> 
> ...is easier than...
> 
> identity=MZHvIrRsHRLqqTaEN-S5lLpeD8yk572zFfPFxsCsS-M
> location=0.16978598221719554
> testnet=false
> myName=SanityMBP
> lastGoodVersion=Fred,0.7,1.0,1009
> sig=1fd90af28fddd96d0b159c7bd93658f9267b84dbacef75bc400074258eabe81e,433cac175114111647b4d1e0af84448c34620adb6accf307e14daf72960a8048
> version=Fred,0.7,1.0,1009
> dsaPubKey.y=Xw1RbImKwC5EYlfOZtvV0nUz-ujplXfW6IHq125LfajUe2m6sA~IHwvu75wtCvdp~TgjNlgddQMhVBtOkhvSIr1cdiWFYqSaeULGFDeFJdON8PZj1Apt7IN2xIVh17WUA99JoeZhCKUH9d-nKubllovx0qFyVpezkPaluWhQrOx0JCYEYJOmah5xIUlvdPfG7bkSWGnWDq5VfJAQTIW~XW2CBjtXh1jbsfOAE1l9k35JY44jNJ4Ehq-qlPLYO1gqpHE~HsikKxYFGX3FcByxRIYCTxVTSfO9bx6vAxzaBQnDIC6zthhzKJAezRJEM~YZ4zZA0q39~0Y8eKK1yIrSOQ
> physical.udp=sanity1.dyndns.org:14505;76.167.194.83:14505
> dsaGroup.g=UaRatnDByf0QvTlaaAXTMzn1Z15LDTXe-J~gOqXCv0zpz83CVngSkb--bVRuZ9R65OFg~ATKcuw8VJJwn1~A9p5jRt2NPj2EM7bu72O85-mFdBhcav8WHJtTbXb4cxNzZaQkbPQUv~gEnuEeMTc80KZVjilQ7wlTIM6GIY~ZJVHMKSIkEU87YBRtIt1R~BJcnaDAKBJv~oXv1PS-6iwQRFMynMEmipfpqDXBTkqaQ8ahiGWA41rY8d4jDhrzIgjvkzfxkkcCpFFOldwW8w8MEecUoRLuhKnY1sm8nnTjNlYLtc1Okeq-ba0mvwygSAf4wxovwY6n1Fuqt8yZe1PDVg
> dsaGroup.q=ALFDNoq81R9Y1kQNVBc5kzmk0VvvCWosXY5t9E9S1tN5
> dsaGroup.p=AIYIrE9VNhM38qPjirGGT-PJjWZBHY0q-JxSYyDFQfZQeOhrx4SUpdc~SppnWD~UHymT7WyX28eV3YjwkVyc~--H5Tc83hPjx8qQc7kQbrMb~CJy7QBX~YSocKGfioO-pwfRZEDDguYtOJBHPqeenVDErGsfHTCxDDKgL2hYM8Ynj8Kes0OcUzOIVhShFSGbOAjJKjeg82XNXmG1hhdh2tnv8M4jJQ9ViEj425Mrh6O9jXovfPmcdYIr3C~3waHXjQvPgUiK4N5Saf~FOri48fK-PmwFZFc-YSgI9o2-70nVybSnBXlM96QkzU6x4CYFUuZ7-B~je0ofeLdX7xhehuk
> ark.pubURI=SSK at 1CA0TRYSSEnK~Tj7-OqVajmsqS85iUVhjmuz8EU9HQs,MuvO8pTCCPCTqUIMpkEXUBu55nV2DgiGjhSgr~9FY0I,AQABAAE/ark
> ark.number=128
> End
> 
> I think it speaks for itself.
> 
> Ian.
> 
> On 3/5/07, Matthew Toseland <toad at amphibian.dyndns.org> wrote:
> > I don't understand why a password and IP address is easier than a
> > one-time reference. I suppose it has the advantage of being able to
> > write it down - but for it to be secure it would need to be a one-time
> > password; you'd need to generate a new one every time ...
> >
> > Hmmm. Maybe we should provide both mechanisms?
> >
> > On Mon, Mar 05, 2007 at 01:13:43PM -0500, Colin Davis wrote:
> > > It solves #2- Don't run things you get in e-mail..
> > >
> > > Instead of requiring a Noderef, allow someone to connect with just a
> > > password, and the IP address. This is something you can TELL someone, or
> > > say in an IM, no file transfer required.
> > >
> > > Dave Baker wrote:
> > > > On Monday 05 March 2007 18:02:42 Colin Davis wrote:
> > > >> I know it's less secure, but what about simply allowing people to
> > > >> connect to your machine if they know a passphrase? The passphrase would
> > > >> take the place of the Key, but be user-settable, and short.
> > > >
> > > > That doesn't solve either problem though, surely?
> > > >
> > > > my 2p on #freenet:
> > > >
> > > > [17:48] <dbkr> as far as both-way-adding goes, I think that's where we reach a
> > > > tradeoff with security, which is one of the main challanges for Freenet.
> > > > [17:49] <dbkr> I'm not convinced the whole difficulty of exchanging refs isn't
> > > > a red herring - everyone can handle emailing a file.
> > > >
> > > > I'm definately in favour of the ability to burn a CD with an installer on it
> > > > that installs a node with your reference pre-bundled, although I think
> > > > leaving the installer out for an emailed-version means it's nothing the user
> > > > couldn't do themselves.
> > > >
> > > >
> > > > Dave
> > > >
> > > >
> > > >>
> > > >>
> > > >> If that were in place, you could send an e-mail saying:
> > > >>
> > > >> Hey Jon, I just found this cool new thing called freenet, which lets you
> > > >> get to all sorts of sites which aren't on the normal web! It's
> > > >> anonymous, and free, you should check it out. It works by connecting
> > > >> through each other's computers, but I'll let you connect to me to get
> > > >> started.
> > > >>
> > > >> Go to FreenetProject.org and download it, then give it my hostname,
> > > >> which is XXXXXXX and give it the connection passphrase "IamNotEvil".
> > > >>
> > > >> Don't give anyone else that information, or it won't work. It'll only
> > > >> allow one connection.. After your up, you can connect to other friends,
> > > >> and everyone's connection gets faster.
> > > >>
> > > >> I'm on IM if you want to talk about it.
> > > >> -Person you Know.
> > > >>
> > > >> Matthew Toseland wrote:
> > > >>> We will only get a darknet if it is really easy to swap references with
> > > >>> your friends - opennet or no opennet.
> > > >>>
> > > >>> The original idea for Freenet 0.7 reference swapping was that you:
> > > >>> - Go to your node, and ask it to create a bundle.
> > > >>> - Send the bundle to your friends.
> > > >>> - They unzip it and run it to install Freenet.
> > > >>> - The bundle includes your noderef.
> > > >>> - It also includes a one-time key that allows the node to automatically
> > > >>>   connect to yours despite yours not having their noderef yet.
> > > >>>
> > > >>> There are two big problems with this:
> > > >>>
> > > >>> 1) Everyone and his dog is behind a NAT. This means in order to connect
> > > >>> you must have already exchanged references, full stop. THIS SUCKS. It
> > > >>> also affects connectivity for newbies in a bad way (which is important
> > > >>> IMHO).
> > > >>>
> > > >>> 2) Generally people shouldn't run programs that they receive in emails!
> > > >>>
> > > >>> Solution to the first one - and to newbie connectivity issues - is to
> > > >>> implement UP&P and hope that routers implement it properly in future -
> > > >>> is this a realistic hope?
> > > >>>
> > > >>> Solution to the second one is to just send the noderef and a link to the
> > > >>> website, and only use full bundles when e.g. giving somebody a CD-R
> > > >>> (which we should make really easy).
> > > >>>
> > > >>> Plugins for e.g. IRC clients, IM clients, have been suggested but I'm
> > > >>> not sure how well this would work for newbies, and in any case I set up
> > > >>> a darknet-tools list for people to talk about this and nobody has even
> > > >>> talked about it since a few days after it was set up, let alone done
> > > >>> anything.
> > > >>>
> > > >>> <_ph00> so the basic problem is "how to safely exchage refs", and the
> > > >>> solution "eliminate ref exchanging by implementing opennet"?!? Am I the
> > > >>> only one to think that's very stupid?
> > > >>>
> > > >>>
> > > >>> ------------------------------------------------------------------------
> > > >>>
> > > >>> _______________________________________________
> > > >>> Devl mailing list
> > > >>> Devl at freenetproject.org
> > > >>> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> > > >> _______________________________________________
> > > >> Devl mailing list
> > > >> Devl at freenetproject.org
> > > >> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> > > >
> > > >
> > > > _______________________________________________
> > > > Devl mailing list
> > > > Devl at freenetproject.org
> > > > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> > >
> > > _______________________________________________
> > > Devl mailing list
> > > Devl at freenetproject.org
> > > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> > >
> >
> > -----BEGIN PGP SIGNATURE-----
> > Version: GnuPG v1.4.6 (GNU/Linux)
> >
> > iD8DBQFF7Jt7A9rUluQ9pFARAr7gAJ9iJHAur3dIsrClY+AmFSQgOy6bXgCfY2Po
> > kqDOiGlYjY/lZ0v66rlsip0=
> > =NcEY
> > -----END PGP SIGNATURE-----
> >
> > _______________________________________________
> > Devl mailing list
> > Devl at freenetproject.org
> > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> >
> 
> 
> -- 
> Founder and CEO, Thoof Inc
> Email: ian at thoof.com
> Office: +1 512 485 1970
> Cell: +1 310 593 3724
> AIM: ian.clarke at mac.com
> Skype: sanity
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
> 
-------------- 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/devl/attachments/20070306/07157e0a/attachment.pgp 

From toad at amphibian.dyndns.org  Tue Mar  6 16:42:51 2007
From: toad at amphibian.dyndns.org (Matthew Toseland)
Date: Tue, 6 Mar 2007 16:42:51 +0000
Subject: [freenet-dev] what about the automatic updater?
In-Reply-To: <20070306103051.GB4145@freenetproject.org>
References: <200703060712.50393.the_bishop@web.de>
	<20070306103051.GB4145@freenetproject.org>
Message-ID: <20070306164251.GB3992@amphibian.dyndns.org>

On Tue, Mar 06, 2007 at 11:30:51AM +0100, Florent Daigni?re (NextGen$) wrote:
> * Martin Scheffler <the_bishop at web.de> [2007-03-06 07:12:48]:
> 
> > does the updated code gets uploaded to freenet this time?
> > i had no luck updating the node by itself so far...
> > 
> > good byte
> 
> Toad hasn't announced the new build yet: it's likely he hasn't inserted
> it.

I have inserted 1018 but not 1017.
-------------- 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/devl/attachments/20070306/1ec7936a/attachment.pgp 

From freenet-devl at osndok.com  Tue Mar  6 18:03:40 2007
From: freenet-devl at osndok.com (Robert Hailey)
Date: Tue, 6 Mar 2007 12:03:40 -0600
Subject: [freenet-dev] Easier reference swapping
In-Reply-To: <20070306164222.GA3992@amphibian.dyndns.org>
References: <20070305174545.GA29567@amphibian.dyndns.org>
	<45EC5B42.1010701@sq7.org>
	<200703051809.41642.dbkr@freenetproject.org>
	<45EC5DD7.6080302@sq7.org>
	<20070305223643.GA12373@amphibian.dyndns.org>
	<823242bd0703060705s280801d4o217062ddfcf12817@mail.gmail.com>
	<20070306164222.GA3992@amphibian.dyndns.org>
Message-ID: <A3820F95-6AC9-45E6-A3FF-2CA48F8B6F71@osndok.com>


If I may suggest... In a password based connection establishment the  
user has to interact with the node anyway (either to set or view the  
password to convey it). As I understand it, an established node must  
already have a working NAT hole with it's external address and port  
known. So why not have the node supply this short (copy/pastable, IM- 
able, conversational) string?

e.g.
User enters "foriemdhs" as a new connection password, to which the  
node responds...
"Accepted, here is a url to your node: foriemdhs at 43.213.43.1:54533"

With the pinhole information, the new node can then send to the  
established node it's reference; yet the new node most likely does  
not know it's external address. However, wouldn't the source-address  
from the udp packets be the new node's external address 99% of the time?

Perhaps this would not work if the node's pinhole changes frequently.

--
Robert Hailey


On Mar 6, 2007, at 10:42 AM, Matthew Toseland wrote:

> Well, if you are exchanging electronically, you are probably able to
> move the latter as easily as the former, no?
>
> Sadly no, because of the [ insert favourite expletive ] NATs which  
> kill
> DCC. But you can move a small file with IM clients, surely? It's  
> moving
> a file and then double clicking on it or even executing it in place,
> versus copying a line of text manually.
>
> So it would be easier for IRC and other stuff crippled by NATs.  
> Would it
> be easier for anything practically used by semi-geek (rather than
> ubergeek) users?
>
> Also you can write it down. However, because it would probably have to
> be a one-time password (for security when dealing with newbies; if  
> it's
> a two way exchange, you could maybe have more permanent passwords,  
> at a
> cost in security), you can't remember it and write it down. And
> even if it wasn't one-time, you probably can't write it down anyway if
> you don't have a semistatic IP - but a lot of people *do* have
> semistatic IPs...
>
> Anyway, it would help in some cases, sure. So we should probably have
> both.
>
> On Tue, Mar 06, 2007 at 09:05:56AM -0600, Ian Clarke wrote:
>> You don't understand why:
>>
>> 43.213.43.1:54533:foriemdhs
>>
>> ...is easier than...
>>
>> identity=MZHvIrRsHRLqqTaEN-S5lLpeD8yk572zFfPFxsCsS-M
>> location=0.16978598221719554
>> testnet=false
>> myName=SanityMBP
>> lastGoodVersion=Fred,0.7,1.0,1009
>> sig=1fd90af28fddd96d0b159c7bd93658f9267b84dbacef75bc400074258eabe81e, 
>> 433cac175114111647b4d1e0af84448c34620adb6accf307e14daf72960a8048
>> version=Fred,0.7,1.0,1009
>> dsaPubKey.y=Xw1RbImKwC5EYlfOZtvV0nUz- 
>> ujplXfW6IHq125LfajUe2m6sA~IHwvu75wtCvdp~TgjNlgddQMhVBtOkhvSIr1cdiWFYq 
>> SaeULGFDeFJdON8PZj1Apt7IN2xIVh17WUA99JoeZhCKUH9d- 
>> nKubllovx0qFyVpezkPaluWhQrOx0JCYEYJOmah5xIUlvdPfG7bkSWGnWDq5VfJAQTIW~ 
>> XW2CBjtXh1jbsfOAE1l9k35JY44jNJ4Ehq- 
>> qlPLYO1gqpHE~HsikKxYFGX3FcByxRIYCTxVTSfO9bx6vAxzaBQnDIC6zthhzKJAezRJE 
>> M~YZ4zZA0q39~0Y8eKK1yIrSOQ
>> physical.udp=sanity1.dyndns.org:14505;76.167.194.83:14505
>> dsaGroup.g=UaRatnDByf0QvTlaaAXTMzn1Z15LDTXe-J~gOqXCv0zpz83CVngSkb-- 
>> bVRuZ9R65OFg~ATKcuw8VJJwn1~A9p5jRt2NPj2EM7bu72O85- 
>> mFdBhcav8WHJtTbXb4cxNzZaQkbPQUv~gEnuEeMTc80KZVjilQ7wlTIM6GIY~ZJVHMKSI 
>> kEU87YBRtIt1R~BJcnaDAKBJv~oXv1PS-6iwQRFMynMEmipfpqDXBTkqaQ8ahiGWA41rY 
>> 8d4jDhrzIgjvkzfxkkcCpFFOldwW8w8MEecUoRLuhKnY1sm8nnTjNlYLtc1Okeq- 
>> ba0mvwygSAf4wxovwY6n1Fuqt8yZe1PDVg
>> dsaGroup.q=ALFDNoq81R9Y1kQNVBc5kzmk0VvvCWosXY5t9E9S1tN5
>> dsaGroup.p=AIYIrE9VNhM38qPjirGGT-PJjWZBHY0q- 
>> JxSYyDFQfZQeOhrx4SUpdc~SppnWD~UHymT7WyX28eV3YjwkVyc~-- 
>> H5Tc83hPjx8qQc7kQbrMb~CJy7QBX~YSocKGfioO- 
>> pwfRZEDDguYtOJBHPqeenVDErGsfHTCxDDKgL2hYM8Ynj8Kes0OcUzOIVhShFSGbOAjJK 
>> jeg82XNXmG1hhdh2tnv8M4jJQ9ViEj425Mrh6O9jXovfPmcdYIr3C~3waHXjQvPgUiK4N 
>> 5Saf~FOri48fK-PmwFZFc-YSgI9o2-70nVybSnBXlM96QkzU6x4CYFUuZ7- 
>> B~je0ofeLdX7xhehuk
>> ark.pubURI=SSK at 1CA0TRYSSEnK~Tj7- 
>> OqVajmsqS85iUVhjmuz8EU9HQs,MuvO8pTCCPCTqUIMpkEXUBu55nV2DgiGjhSgr~9FY0 
>> I,AQABAAE/ark
>> ark.number=128
>> End
>>
>> I think it speaks for itself.
>>
>> Ian.
>>
>> On 3/5/07, Matthew Toseland <toad at amphibian.dyndns.org> wrote:
>>> I don't understand why a password and IP address is easier than a
>>> one-time reference. I suppose it has the advantage of being able to
>>> write it down - but for it to be secure it would need to be a one- 
>>> time
>>> password; you'd need to generate a new one every time ...
>>>
>>> Hmmm. Maybe we should provide both mechanisms?
>>>
>>> On Mon, Mar 05, 2007 at 01:13:43PM -0500, Colin Davis wrote:
>>>> It solves #2- Don't run things you get in e-mail..
>>>>
>>>> Instead of requiring a Noderef, allow someone to connect with  
>>>> just a
>>>> password, and the IP address. This is something you can TELL  
>>>> someone, or
>>>> say in an IM, no file transfer required.
>>>>
>>>> Dave Baker wrote:
>>>>> On Monday 05 March 2007 18:02:42 Colin Davis wrote:
>>>>>> I know it's less secure, but what about simply allowing people to
>>>>>> connect to your machine if they know a passphrase? The  
>>>>>> passphrase would
>>>>>> take the place of the Key, but be user-settable, and short.
>>>>>
>>>>> That doesn't solve either problem though, surely?
>>>>>
>>>>> my 2p on #freenet:
>>>>>
>>>>> [17:48] <dbkr> as far as both-way-adding goes, I think that's  
>>>>> where we reach a
>>>>> tradeoff with security, which is one of the main challanges for  
>>>>> Freenet.
>>>>> [17:49] <dbkr> I'm not convinced the whole difficulty of  
>>>>> exchanging refs isn't
>>>>> a red herring - everyone can handle emailing a file.
>>>>>
>>>>> I'm definately in favour of the ability to burn a CD with an  
>>>>> installer on it
>>>>> that installs a node with your reference pre-bundled, although  
>>>>> I think
>>>>> leaving the installer out for an emailed-version means it's  
>>>>> nothing the user
>>>>> couldn't do themselves.
>>>>>
>>>>>
>>>>> Dave
>>>>>
>>>>>
>>>>>>
>>>>>>
>>>>>> If that were in place, you could send an e-mail saying:
>>>>>>
>>>>>> Hey Jon, I just found this cool new thing called freenet,  
>>>>>> which lets you
>>>>>> get to all sorts of sites which aren't on the normal web! It's
>>>>>> anonymous, and free, you should check it out. It works by  
>>>>>> connecting
>>>>>> through each other's computers, but I'll let you connect to me  
>>>>>> to get
>>>>>> started.
>>>>>>
>>>>>> Go to FreenetProject.org and download it, then give it my  
>>>>>> hostname,
>>>>>> which is XXXXXXX and give it the connection passphrase  
>>>>>> "IamNotEvil".
>>>>>>
>>>>>> Don't give anyone else that information, or it won't work.  
>>>>>> It'll only
>>>>>> allow one connection.. After your up, you can connect to other  
>>>>>> friends,
>>>>>> and everyone's connection gets faster.
>>>>>>
>>>>>> I'm on IM if you want to talk about it.
>>>>>> -Person you Know.
>>>>>>
>>>>>> Matthew Toseland wrote:
>>>>>>> We will only get a darknet if it is really easy to swap  
>>>>>>> references with
>>>>>>> your friends - opennet or no opennet.
>>>>>>>
>>>>>>> The original idea for Freenet 0.7 reference swapping was that  
>>>>>>> you:
>>>>>>> - Go to your node, and ask it to create a bundle.
>>>>>>> - Send the bundle to your friends.
>>>>>>> - They unzip it and run it to install Freenet.
>>>>>>> - The bundle includes your noderef.
>>>>>>> - It also includes a one-time key that allows the node to  
>>>>>>> automatically
>>>>>>>   connect to yours despite yours not having their noderef yet.
>>>>>>>
>>>>>>> There are two big problems with this:
>>>>>>>
>>>>>>> 1) Everyone and his dog is behind a NAT. This means in order  
>>>>>>> to connect
>>>>>>> you must have already exchanged references, full stop. THIS  
>>>>>>> SUCKS. It
>>>>>>> also affects connectivity for newbies in a bad way (which is  
>>>>>>> important
>>>>>>> IMHO).
>>>>>>>
>>>>>>> 2) Generally people shouldn't run programs that they receive  
>>>>>>> in emails!
>>>>>>>
>>>>>>> Solution to the first one - and to newbie connectivity issues  
>>>>>>> - is to
>>>>>>> implement UP&P and hope that routers implement it properly in  
>>>>>>> future -
>>>>>>> is this a realistic hope?
>>>>>>>
>>>>>>> Solution to the second one is to just send the noderef and a  
>>>>>>> link to the
>>>>>>> website, and only use full bundles when e.g. giving somebody  
>>>>>>> a CD-R
>>>>>>> (which we should make really easy).
>>>>>>>
>>>>>>> Plugins for e.g. IRC clients, IM clients, have been suggested  
>>>>>>> but I'm
>>>>>>> not sure how well this would work for newbies, and in any  
>>>>>>> case I set up
>>>>>>> a darknet-tools list for people to talk about this and nobody  
>>>>>>> has even
>>>>>>> talked about it since a few days after it was set up, let  
>>>>>>> alone done
>>>>>>> anything.
>>>>>>>
>>>>>>> <_ph00> so the basic problem is "how to safely exchage refs",  
>>>>>>> and the
>>>>>>> solution "eliminate ref exchanging by implementing  
>>>>>>> opennet"?!? Am I the
>>>>>>> only one to think that's very stupid?
>>>>>>>
>>>>>>>
>>>>>>> ---------------------------------------------------------------- 
>>>>>>> --------
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Devl mailing list
>>>>>>> Devl at freenetproject.org
>>>>>>> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
>>>>>> _______________________________________________
>>>>>> Devl mailing list
>>>>>> Devl at freenetproject.org
>>>>>> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Devl mailing list
>>>>> Devl at freenetproject.org
>>>>> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
>>>>
>>>> _______________________________________________
>>>> Devl mailing list
>>>> Devl at freenetproject.org
>>>> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
>>>>
>>>
>>> -----BEGIN PGP SIGNATURE-----
>>> Version: GnuPG v1.4.6 (GNU/Linux)
>>>
>>> iD8DBQFF7Jt7A9rUluQ9pFARAr7gAJ9iJHAur3dIsrClY+AmFSQgOy6bXgCfY2Po
>>> kqDOiGlYjY/lZ0v66rlsip0=
>>> =NcEY
>>> -----END PGP SIGNATURE-----
>>>
>>> _______________________________________________
>>> Devl mailing list
>>> Devl at freenetproject.org
>>> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
>>>
>>
>>
>> -- 
>> Founder and CEO, Thoof Inc
>> Email: ian at thoof.com
>> Office: +1 512 485 1970
>> Cell: +1 310 593 3724
>> AIM: ian.clarke at mac.com
>> Skype: sanity
>> _______________________________________________
>> Devl mailing list
>> Devl at freenetproject.org
>> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
>>
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl



From toad at amphibian.dyndns.org  Tue Mar  6 19:52:43 2007
From: toad at amphibian.dyndns.org (Matthew Toseland)
Date: Tue, 6 Mar 2007 19:52:43 +0000
Subject: [freenet-dev] Beginnings of new low-level spec
Message-ID: <20070306195243.GA16228@amphibian.dyndns.org>

MRogers posted the following on Frost:

http://wiki.freenetproject.org/NewPacketFormat
http://wiki.freenetproject.org/NewTransportLayer

Any feedback would be welcome. These are the new proposed formats for
the packet transport layer and the congestion control and retransmission
layer respectively.
-------------- 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/devl/attachments/20070306/73469430/attachment.pgp 

From toad at amphibian.dyndns.org  Tue Mar  6 21:22:14 2007
From: toad at amphibian.dyndns.org (Matthew Toseland)
Date: Tue, 6 Mar 2007 21:22:14 +0000
Subject: [freenet-dev] Freenet 0.7 build 1018
Message-ID: <20070306212213.GA504@amphibian.dyndns.org>

Freenet 0.7 build 1018 is now available. Your nodes should automatically
download the new update for you; please upgrade. If auto-update is not
working, please tell us.

There are very many changes in build 1018, far too many to list here,
but the main thing is that Fred's memory usage should be significantly
reduced. A lot of work has gone into this area, including a 2-line patch
which dramatically cuts the datastore's memory use, and major work