It appears to me that we are making some rather weird assumptions here,
based on history in other environments (like current domain names, phone
numbers, USMail addresses, and so). It is not so clear to me that there will
always be a one-to-one mapping between the destination one wants to reach,
and some name like vixie.sf.ca.us. Eventually such strict bindings will hurt
us even more than they do today. What do you really want? Well, I want to
email Joe Smith? No way, there are too many of those, give me more
information? So, eventually may be we have to deal with more levels of
indirection and some data base support that allows us to zoom in, and, like
it or not, perhaps introduce some margin of error ("sorry, wrong number,
find another Vixie").
I think in our discussions we have ways too much a historical Internet
with strict bindings in mind, and names and addresses that are supposed to
last for lfetimes. This will not work. Ther are too many people, too many
companies, too many mapping strategies out there. Whether I email to "Vixie,
the bind dude," or "Vixie, the networking person living in Northern
California," may be there should be algorithms and facilities in place that
map those to the same person. Perhaps even in ways the information (such as
email) follows him, rather than him having to chase after his email (e.g.,
remap his "current" location when/after he moves).
What names or addresses things eventually map to should be fairly irrelevant
in terms of perceived complexity, as you'd alias Joe Smith@whereever to some
short version in your personal alias/nickname file anyway. Or is that just
mapping information for a data base supported translation server?