Hi Jim,

I'm still not sure I understand your position?

> > I'm not arguing that general-purpose apps should
> > do anything special to support link-local addresses.
> > For the most part, I doubt they'll ever see them,
> > since link-local addresses won't be in the DNS.
> 
> Exactly my point in last mail to Itojun.
> 
> > 
> > But then again, I don't think that most apps need to do
> > anything to discourage their use with link-local addresses. 
> 
> I agree.  I am not worried about that if they are not
> in DNS.

To me, this means that the IETF doesn't have to take any new action on
this issue.  So I'm not sure what it is you are advocating?

Let me try to clarify my position.  All I'm saying is that if an app
works fine with link-locals, and that's all the user has, why not let
the user use that app?  In situations where there is no router (which is
what I meant by "ad-hoc" by the way -- a no infrastructure situation:
could be one link or many links), all the user has is link-local
addresses.  If I'm out in the park working on my IPv6 laptop and a
friend drops by with another IPv6 laptop, shouldn't we be able to string
a firewire (or reversing Ethernet) cable between our two machines and
allow say telnet to work from one to the other?  And if I was dictating
into my laptop via my IPv6 Bluetooth microphone before my friend dropped
by, shouldn't that continue to function even after we plug in the
firewire (or Ethernet) cable?

IPv4 deals okay today with the telnet scenario above.  There are
potential problems with IPv4 if you throw in the second link
(represented by the bluetooth microphone in the above scenario) since
collisions could occur in the 169.254/16 space, and IPv4 doesn't have a
scope-id in the sockaddr_in to differentiate the links.  IPv6 solves
that problem.  This is a big selling point for IPv6.

Sure, there are an extremely small percentage of apps that do referrals,
or otherwise won't work with link-locals.  That doesn't mean we should
penalize the vast majority of apps that work just fine.  If we tell the
authors of telnet (to stick to that one example) that they have to add
special code to watch out for users typing in link-local addresses, and
then refuse to use them when the user does so (even though the app would
otherwise have worked just fine), what have we gained?  I think we'll
have gained an upset user who will go back to IPv4.

--Brian

--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page:                      http://playground.sun.com/ipng
FTP archive:                      ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to