On 18 sep 2008, at 19:07, The IESG wrote:
The document SHOULD use values reserved for examples where such
assignments exist (e.g. BCP 32, RFC 3330, RFC 4735, and RFC 5156)
and when
the necessary semantic can be communicated clearly enough. If
unassigned
codepoints are desired it is RECOMMENDED that those codepoints be
assigned
or registered. If assigned codepoints are desired, it is
RECOMMENDED that
the authors get approval from the current codepoint holder.
First of all: note that this issue goes well beyond the traditional
reach of the IESG. For instance, I've written books and taught
training courses where obviously examples had to be used.
For domain names I used example.com et al, operating under the idea
that anyone who registers such a domain can't complain that it's used
in examples.
However, IP addresses are often a concern. IPv4 only has 192.0.2.0/24
which is completely insufficient for decent examples. IPv6 has
2001:db8::/32 which is big, but still not useful in all cases because
it's often necessary to clearly show that two address ranges are
different so they must be different prefixes. A common solution is to
use RFC 1918 addresses for examples but I don't like it in general and
it's not really workable in certain cases because these addresses
carry a special meaning. (I.e., an RFC 1918 address in a 6to4 example
would create the impression that this is a workable combination which
isn't the case.)
So I suggest reserving three or so additional IPv4 and IPv6 global
unicast prefixes for documentation purposes, that look sufficiently
different from each other.
_______________________________________________
Ietf@xxxxxxxx
https://www.ietf.org/mailman/listinfo/ietf