[CUWiN] the future... ?

David Young dyoung at pobox.com
Fri Oct 8 00:00:39 CDT 2004


On Tue, Oct 05, 2004 at 11:08:25AM +0100, Ian White wrote:
> Can you tell me where CUWiN is heading ? The frontpage is out of date
> by a couple of months, and the technical docs appear to imply a year's
> worth of work ahead.

It's not quite that long.  Our routing software is due Jan 1, 2005.

> I'm coming from the the direction of MeshAp (ADOV) and Locustworld, but have
> concerns about the open-sourceness of that project.
> 
> I'm looking for the kind of functions that LW offer but with a true open-source
> project with people able to take the project forward.
> 
> I see that the software is pretty open, and I'm not sure how the routing etc
> works, but I'm looking for
> 
> 1) Cheap PC solutions, capable of having repeater and multi-gateway meshes

By multi-gateway meshes, I take it that you want multiple connections
to the Internet?  I.e., a "multi-homed" network?  We are a ways off from
solving that problem satisfactorily.  There are some problem statements
and proposals in our source tree, including a design for a protocol
to help a router choose its gateway.  See the files starting with
'trunk/src/doc/local/gateway-discovery'.

> 2) Login Security - Nocat appear pretty standard, and I've got no
> issues with hacking it

NoCat+PERL is too big for the devices we're targetting, but these days
there is NoCatSplash!  We'd be pleased if somebody would integrate it
with our system.

> 3) Network security ? IPsec between nodes, and new nodes requiring a
> 'mesh' cert to be able to join the mesh, and stop rogue nodes.

We are not treating security problems in this phase.  It will be hard for
us to strike a good balance between a free and open network, and security.
Also, this network technology has to run in the absence of a public-key
infrastructure.

Something that is missing in the security area is a threat model for CWNs.
The threat model would be an enormous contribution, if you would undertake
to create it.  One way to get started is to identify and interview people
around the world who are building CWNs.  But what is best, I think, is
to interview some "subscribers" to CWNs.  The engineering should follow
the threat model.

> 4) VPN support. Many clients are using VPN to connect to their office
> networks.

Our network should be transparent to VPN software.  Do you have some
special requirements?

In sum, everything you mention is something we either want to, do, or
will support.  We can get there with your help!  Download our development
environment and "play."  It is available through anonymous CVS and SVN.
Here are Zach's instructions:

  Anonymous CVS is hosted by Source Forge:

  Information: http://sourceforge.net/cvs/?group_id=89823
  Web CVS Viewer: http://cvs.sourceforge.net/viewcvs.py/wireless

  To checkout use these commands:

  cvs -d:pserver:anonymous at cvs.sourceforge.net:/cvsroot/wireless login
  cvs -z3 -d:pserver:anonymous at cvs.sourceforge.net:/cvsroot/wireless \
      co cuw-trunk

  Don't forget that you should use "cvs update -dP" when updating to
  make sure you get the right directories in your tree.

  Anonymous SVN is hosted on cuwireless.net:

  Browsable URL: http://svn.cuwireless.net/svn/cuw
  svn checkout http://svn.cuwireless.net/svn/cuw/trunk

Dave

-- 
David Young             OJC Technologies
dyoung at ojctech.com      Urbana, IL * (217) 278-3933


More information about the CU-Wireless mailing list