From: "William C. Green" <w(_dot_)green(_at_)utexas(_dot_)edu>
X-Sender: green(_at_)mcl(_dot_)cc(_dot_)utexas(_dot_)edu
To: pem-dev(_at_)tis(_dot_)com
The attached exchange between anish(_at_)ctt(_dot_)bellcore(_dot_)com and
Jueneman(_at_)gte(_dot_)com is focused on the handling of nicknames. However,
the underlying assumption is that email addresses are the basic form
of identity used in the Internet.
On the Internet, yes. And I think it sucks. My university spends
countless consulting and course hours trying to explain to bright students
how the addresses work and why. As opposed to adapting the technology to
Sorry, I don't believe that. If the students are bright, a brief
written document should satisfy this need.
how people work in the real world, we have adapted the people to the
technology and software we could implement at the time (over a decade). Do
you really think the whole world is going to adapt? Why should we continue
to force them to?
Sure. The real world. You call someone by dialing a random variable
length string of digits with a variable length country code, assorted
"area code" like things, various prefix forms that you sometimes have
to strip and replace, sometimes an "extension", sometimes digits
represented by letters which letters respresent different digits on
dials in other countries (!), etc., etc. Of course things are simpler
in toy cases like a directory of a few hundred or thousand people that
have 3 or 4 digits phone numbers or can all be emailed to in a uniform
manner like firstname(_dot_)lastname(_at_)foo(_dot_) It's always easier to
organize/index a thousand things than a million things or a billion
things.
In all other interactions of my life, I do not use an email address to
identify myself.
Well, what do you mean identify? Most people just use their human
names. But if you want someone to contact you, you need to give them
more info like an 11 digit phone number or your email address or
enough information to look up this direct pointers. In my circles of
acquantence, the email is at least as important as phone number.
(Note: now that there is a mapping of all of the world's phone numbers
into DNS entries under the tpc.int domain, there is no particular reason
not to use that to map from your phone number to your email address. Since
email names map to DNS entries, the inverse is also true so you could
store a PTR from w\.green.utexas.edu to the node corresponding to your
phone number.)
In my view, the introduction of X.500 distinguished names has been a
very troublesome venture, and I see no evidence that things will get
better. Quite a lot has to happen before X.500 names are genuinely
useful as the basis for identity on the net.
I disagree. Observe the following DN in User Friendly Notation and
corresponding email addresses:
William C Green, vs.
green(_at_)wowbagger(_dot_)cc(_dot_)utexas(_dot_)edu
Computation Center, or
University of Texas at Austin,
w(_dot_)green(_at_)utexas(_dot_)edu
Texas,
US
Come on. To use your postal address to get anything useful out of any
existing directory system is a difficult artificial intelligence
problem.
One of these can be recognized and understood by people throughout the
world (looks kind of like what we have seen on correspondence for the past
century). The other can only be recognized by about 20,000,000 people in
the world and understood by even less. And the email addresses I use are
If someone is not Internet connected, who cares if they recognize an
email address? And why do they need to understand it? It is not
intended to provide the information that is elsewhere on your Business
Card.
pretty friendly compared to a lot that I've seen. The DN uniquely
identifies me at the University of Texas. The email addresses narrow the
field to three in the case of w.green, and 77 in the case of just green
(the Comp Center - cc - provides mail to the whole campus). The DN
provides organizational information -- the email address doesn't.
Even if email addresses were used, there would still have to be
certification hierarchies. The infrastructure for X.500 is all described
and being implemented for use in multiple applications -- not just PEM (our
plans are to use certificates for everything from unlocking dorms to buying
cokes). I believe you would run into the same problems and trade-offs in
implementing such an infrastructure with email addresses that have occurred
in the X.500 model -- only the solution would be single purpose. .
I disagree. There is exists a global distributed database trivially
extensible to include everyone's email address just waiting for
additional info to be stored in it. I just don't see what problems
you are talking about. If you can email someone from the Internet,
there must be at least an MX'ing gateway machine that could hold the
requisite DNS entries.
X.500 is complex, burdensome, and inflexible (I use more colorful
adjectives when my door is closed). Believe you me, I know-- I run an
80,000 entry DSA. But I would take a DN over an email address for business
correspondence any day of the week.
I see we mostly agree that X.500 is a loss. But for business
correspondence, why do you view it as a choice between a DN and email?
Surely you put what you think the other person needs: email if they
are sending email to you, your phone number if they might need to
phone you, your postal/shipping address if that is needed, and
explicit organizational/role information where appropriate.
If there were a good directory system, you would need to give only one
unique thing. It could be your phone number (short all numeric...),
your email address (a bit longer with some mnemonic structure...),
your DN (very long and complex but somewhat informatiave), or perhaps
other possibilities like your country number and personal ID number,
But there isn't a good directory system deployed.
-William
Donald