On Wed, Apr 26, 2006 at 11:19:23AM +0000, NextGen$ wrote:
>
> > I see your concerns and hence I checked the SoC2005 website to see how much
> > work have been done by other projects in 2005. For instance, there were two
> > projects for gaim. One student implemented STUN/SIP/Simple and another
> > studenten implemented UPnP for gaim. So, I guess it should be enough work for
> > a Soc2006 project to add STUN/UPnP to freenet.
> > However, could you please provide me a hint where I can find more information
> > about ARKs. Probably I should consider implementing ARKs as well? I like the
> > basic idea of ARKs, however, I think my understanding of ARKs is not deep
> > enough to make a decision.
>
> imho Up&p and stun are useless without multi-homming support.
>
> Concerning STUN, in .5 we were using a dns root server to detect our external IP
> . That's what STUN will provide, right?. See https://emu.freenetproject.org/svn/branches/legacy/unstable/src/freenet/node/IPAddressDetector.java and oldDetect().
No, we weren't. We were creating a UDP socket pointing to the root
server, and reading off our end's IP address, without sending any
packets. This DOES NOT WORK behind a NAT, and is simply a hack to get our
IP if we're running on a JVM that doesn't support NetworkInterface.
>
> And concerning UP&P ... you know my thought.
I think UP&P is a good idea, but it will require that we ask the user
whether they are on an untrusted LAN.
>
> NextGen$
--
Matthew J Toseland - toad@???
Freenet Project Official Codemonkey - http://freenetproject.org/
ICTHUS - Nothing is impossible. Our Boss says so.
