moore@cs.utk.edu (Keith Moore) (07/19/90)
In article <40713@think.Think.COM> barmar@think.com (Barry Margolin) writes: [concerning X.400 O/R addresses] >The days are long gone when we should expect ordinary users to type >"addresses" into their mail system user interfaces. Well, we still expect ordinary users to write postal addresses on envelopes, and it will be a few years more before we obsolete the current postal system. >The actual form that the user sees will depend on their mail system and >its general user interaction style. A simple interface would display >"Country:", "Last name:", "Computer name:", etc. prompts (with suitable >defaults, of course). This does seem to be what the X.400 developers have in mind. Whether this is an effective user interface is another question entirely. One of the big problems that ordinary users have with the current collection of email networks is that there is no universally accepted notation for addresses. "user@domain" appears to have the widest acceptance, but many users have to translate such an address into whatever their local mail system accepts, be it "domain!user" or gateway::"user@domain" or whatever. X.400 O/R addresses do not solve this problem, and in fact they make it worse. An analogy: if you want to make a telephone call, and you know the telephone number already, you simply dial the number. If you don't know the number, you can call directory assistance. Using X.400 O/R addresses with standard attributes is like calling directory assistance every time you want to place a call. If you know the exact set of name-attribute pairs to supply to specify a unique recipient, it's not a problem -- it only takes longer. If not, you might have to continue specifying additional information until you get it right. My guess is that ordinary users will regard this process as yet another computer-related hassle that impedes their normal work flow. Even if addresses like moore@cs.utk.edu are a bit cryptic (what does "utk" mean, anyway?), it's at least easy to type and easy to remember. Of course, X.400 addresses (and the whole X.400 suite) have more ambitious goals than the domain naming system and e.g., Internet mail. Any addressing scheme that attempts to uniquely name everyone in the world is going to have scaling problems, and it is difficult to construct addresses that are both terse and meaningful to humans. I, for one, will miss domain-style addressing when it's gone, but I have the feeling it will stay around much longer than X.400 developers anticipate. -- Keith Moore Internet: moore@cs.utk.edu University of Tenn. CS Dept. BITNET: moore@utkvx 107 Ayres Hall, UT Campus UT Decnet: utkcs::moore Knoxville Tennessee 37996-1301 Telephone: +1 615 974 0822