Re: [IPv6] [Last-Call] Artart last call review of draft-ietf-6man-rfc6874bis-02

On Mon, Mar 27, 2023 at 4:24 AM Michael Sweet <[email protected]> wrote:

> Brian,
>
> > On Mar 26, 2023, at 10:33 PM, Brian E Carpenter <
> [email protected]> wrote:
> > ...
> >>
> https://datatracker.ietf.org/doc/html/draft-ietf-6man-rfc6874bis#appendix-A
> <
> https://datatracker.ietf.org/doc/html/draft-ietf-6man-rfc6874bis#appendix-A
> >
> >> It says "Simply use the 'IPvFuture' syntax left open in RFC 3986".
> >> You could do that, right? But maybe we have an accidental production
> here. That's fine too. Just say it, if that's the case.
> >
> > CUPS, I believe, does use IPvFuture internally, but that has no impact
> except on CUPS participants.
>
> Actually, it isn't just CUPS, it is all AirPrint printers...  In practice
> this isn't an issue since mDNS hostnames are normally used and the backup
> (but numeric IP addresses) can likewise usually rely on IPv4 addresses
> assigned via DHCP.  But on the off chance that you have an IPv6-only
> network and mDNS doesn't work, you have a situation where you aren't able
> to access the printer's web interface (for configuration) because browsers
> don't support it... :/
>

I tried to think of a really generic example, and thought of java.net.URI.
Is this draft going to break that? The answer seems to be no.

https://docs.oracle.com/javase/8/docs/api/java/net/URI.html

"Aside from some minor deviations noted below, an instance of this class
represents a URI reference as defined by RFC 2396: Uniform Resource
Identifiers (URI): Generic Syntax, amended by RFC 2732: Format for Literal
IPv6 Addresses in URLs. The Literal IPv6 address format also supports
scope_ids. The syntax and usage of scope_ids is described here." where
"here" is this page:

https://docs.oracle.com/javase/8/docs/api/java/net/Inet6Address.html#scoped

So, there are real usages that seem to match exactly what is in the draft.
But, on the other hand, Brian had to patch wget. I think the draft should
be much more direct. State that it is an unfortunate breaking change from
RFC 3986, and it might not always work. I guess the grammar should also be
checked against Java, .NET, etc. just to make sure it will actually work,
not out of principle.

thanks,
Rob

_______________________________________________
art mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/art