Re: ISATAP scenario

[Christian Schild <schild@uni-muenster.de>]

On Tue, 2004-08-03 at 20:28, Pekka Savola wrote:
> > Still, it is a difference if a user has to install a software, configure
> > and maybe troubleshoot it, or if he just has to activate a service
> > (which e.g. is one of the beauties of 6to4 and probably made it a
> > success story).
> 
> IMHO, the point here is that if we could have an agreed-on
> (standardized) mechanism, the vendors could include such software
> automatically and enable it as well.  Then it would be dead simple.  
> Such an implementation could probably be userspace-only, making it
> even easier to implement and deploy. (All the ISATAP implementations
> I've seen have required kernel modifications, making the deployment a
> bit trickier.)

Right. And you had to do that, because KAME/USAGI hadn't tried to put it
in linux/*BSD standard tracks because of that apparently unfounded IPR
fears. 

> Why they haven't done so already is (I think) because there are about
> 5 different solutions, none of them sufficiently widely adopted or
> deployed; they want as few mechanisms as possible, ones which also
> have the potential to be provided by some ISPs, enterprises, or
> whatever.

Ok, aggressive question then:

Why run down the analysis/requirements/solutions road again, when we
have a running solution at hand that _is_ (was) already present in
vendors software and that probably just needs a little bit more
finetuning?

Christian