> I've got some other problems with the document  also - specifically
> the idea that link-local addresses are preferable to global addresses.
> Global addresses are always preferable because all applications
> can tolerate them.  At most, LL addresses should be used only
> when there is no other alternative, because some applications 
> need to treat them specially (this isn't as bad in v6 as in v4)
> But LL addresses should probably be used only when they are 
> explicitly specified.

Part of the issue here depends on how the application finds out addresses
of scope less than global. Link-local addresses don't make sense in
the DNS (and site-local don't make sense in the DNS except when a site 
is running a  two-faced DNS).
Thus if the application is going to explicitly be configured to use
a less-than global address, or explicitly use some link-local name resolution
protocol, this isn't likely to cause a problem in practise.

And for the applications, such as routing daemons, that communicate using
link-local (multicast) addresses the specification causes the right behavior -
a link-local source address will be used when the destination is link local.

  Erik

--------------------------------------------------------------------
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