Re: [IPv6] URL solution for identifying scopes ? Re: Fwd: I-D Action: draft-carpenter-6man-zone-ui-00.txt

[Toerless Eckert <[e-mail]>]

Thanks a lot, Soni

Seems like i may want to ask the question about the uri idea on art.

>From the little i try to understand, the "auth" URI as long as it applies to
the indicated destination URI does not have such a strong argument as i think the
"route:" URI i was suggesting, because the auth information is logically part
of the destination URI and probably origin concept, whereas the "route:"
part is not - because it would only describe logically destination independent
network parameters to use for the connection. Like i think we have a bunch of
protocol URN to parameterize other protocols...

Cheers
    Toerless

On Tue, Feb 20, 2024 at 11:42:55AM -0300, Soni "They/Them" L. wrote:
> 
> 
> On 2024-02-20 11:34, Toerless Eckert wrote:
> > Btw: nobody commented on my question whether a different form of encoding for the scope of an LLA
> > than rfc6874 would solve the URL problems of rfc6874. And not having seen any such discussions
> > in prior rounds was at the core of my concern of the process that previously happened.
> > 
> > On Tue, Feb 20, 2024 at 04:43:40AM +0100, Toerless Eckert wrote:
> > > For example, would something like a new URL schema be better ?
> > > >    route:en0:http://...
> > > > ?
> > > > I mean we have enough cases where one URL encodes another URL as a
> > parameter,
> > > it can't be that difficult to explain the most simple desirable way to do this
> > > for the link selection problem.
> > 
> > Aka: prepend a new type of URI specifying local connection parameters (such as zone)
> > before the target destination ("origin") URL. This new type of URI would only be handled
> > locally by the application (browser, ...) and never be sent across the network.
> > 
> > If this makes sense, it could be defined extensible to include other local socket parameters
> > that might make sense to be able to indicate explicitly. DSCP, SRH path comes to mind.
> > 
> > With appropriate routing setup on multi-homed hosts, this would also make the zone
> > identifier easily useable with any type of host (address or DNS name) in the following URL.
> 
> this looks vaguely familiar...
> 
> https://mailarchive.ietf.org/arch/msg/art/CK4VsoVw2Mi9QR_GncEhnVXbz0E/
> 
> https://mailarchive.ietf.org/arch/browse/art/?q=composable
> 
> > Cheers
> >      Toerless
> > 
> > --------------------------------------------------------------------
> > IETF IPv6 working group mailing list
> > [e-mail]
> > Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> > --------------------------------------------------------------------
> 

-- 
---
[e-mail]