> > > The real point is that some app developers have figured out that
> > > having an architected space makes it easier for them to know when
> > > connectivity will be broken,
> > 
> > I wish you'd stop making such blatently false statements.  It
> > is insulting to everyone in this group and it doesn't
> > contribute usefully to the discussion.
> 
> Please read the following:
> 
> On Wednesday, October 30, 2002 6:22 PM   Brian Zill wrote:
> > ...  Again, the solution
> > I've outlined is simple, practical, and works today.
> 
> Since I have first hand knowledge that Brian has actually done the
> development work to make several applications IPv6 aware, my statement
> is not blatently false.

that lends zero support to the statement you gave earlier.

and has already been discussed, there are lots of applications for
which SLs don't cause a problem, so the fact the numerous apps have
been ported to v6 without major problems is not a compelling case 
for keeping SLs.

apps CANNOT know when connectivity will be broken by use of SLs
even if they know whether those SL addresses came in from the same
interface, which in general they won't.   

if you insist that the apps filter SLs when relaying outside of
their scope then you're artifically breaking connectivity - the
referrer doesn't know whether two nodes with SLs could have 
communicated directly or not.

that and use of SLs doesn't tell the app whether the communication 
was administratively prohibited, any more than globals do.

Keith
--------------------------------------------------------------------
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