[hydra-dev] Networking stack (NAT, Firewall, IPv6)

[Tim Panton]

On 8 Jun 2010, at 10:03, Nick Lewis wrote:

>> I agree that it's not widely implemented yet, but tests do show that it
> solves the problem that it was created to solve.
> Some of the problems solved by STUN TURN and ICE seem a bit contrived
> (or do not actually work e.g. RFC3489). RFC5626 represents a return to
> reality with workable solutions for nat traversal and redundancy so this
> seems a good place to start 
> 
>> If it's needed for Hydra or not is another issue. I might be alone, but
> I still don't have a clear picture of the hole we're trying to fill with
> Hydra, what the product is.
> I think we are all (both digium and the community) suffering from v2
> syndrome and I do not think that the heads down on zeroc stuff is
> necessarily helping the product definition activity. It would be
> fantastic if digium could dedicate more resources to product definition
> and produce a detailed stalking horse that the community can then
> discuss
> 

I totally agree that the zeroc stuff is a distraction at this stage.

By now we should have prototypes and proof-of-concept code floating 
around - thats how good open source projects get momentum.

We seem to have been bogged down in process issues, 
legal and infrastructure.

T.

Tim Panton - Web/VoIP consultant and implementor
www.westhawk.co.uk