pritch@tut.cis.ohio-state.edu (Norm Pritchett) (04/26/89)
I would like to hear from individuals experienced in establishing a centralized electronic mail service for a large user base (4 figures or greater). Here at the Ohio State University we have a campus-wide token ring network interconnecting individually-administered departmental networks whose sizes range from a handful to hundreds. It is not very easy to provide a total count of hosts but it should be pretty close to a thousand. Some of our departments already implement one scheme or another for providing uniform addressing of mail for its users such that a sender need not be concerned with which particular machine to direct the message to. In these cases, the sender addresses the message to the department's Internet domain name (e.g. user@eng.ohio-state.edu or user@cis.ohio-state.edu) and the message is delivered to the recipient on his "home" machine. We would like to implement a similar scheme at the university-wide level where a sender could address a message to some-userid@ohio-state.edu and have the message delivered to the recipient on his home system. The major obstacle is with the "some-userid" part: we wish it to be representative of the recipient's real name (or actually be his real name) while at the same time have it uniquely identify him/her among the 75,000+ faculty, staff and students where there are numerous unresolvable name collisions. A format of Firstname.MI.Lastname which eliminates many collisions still leaves many remaining. If there is anyone who has experience in setting up a similar thing or has constructive advise, please correspond with me via mail at one of the following Internet addresses: pritch@cis.ohio-state.edu npritchett@osu-20.ircc.ohio-state.edu pritchett@eng.ohio-state.edu -- Norm Pritchett, The Ohio State University College of Engineering Network Internet: pritchett@eng.ohio-state.edu BITNET: TS1703 at OHSTVMA UUCP: pritch@sydney.columbus.oh.us CCNET: ENG::PRITCHETT (6172::PRITCHETT)
af%sei.ucl.ac.be@CUNYVM.CUNY.EDU ("Alain FONTAINE ", Postmaster - NAD) (04/28/89)
For what it is worth (two belgian cents = approx 0.0005 dollar..) : We have established an unified address scheme here. But we did not find any way to allow external correspondants to send mail to an individual when only knowing his name, *and* avoid clashes... This seems theoretically impossible. The sender must *know* and *specify* some more information to garantee uniqueness. So the addresses used are of the form : personal-identifier@unit.ucl.ac.be, where 'unit' is the standardized three or four letter sigle of the laboratory or service in which the person can be found. Of course, it is difficult for an external correspondant trying to contact somebody for the first time to guess the 'unit' to be used. On the other hand, clashes are a very low probability event, since units never count more than 50 persons. Implementation : the DNS would be a marvelous tool for this, since each unit could have and manage its own name server. Halas, (one of my favorite gripes), the arbitrary division of mail addresses into a local and a domain part makes it impossible to use the DNS down to the individual level. So the current situation is that one centralized machine contains a centralized database of mail routing information, and nearly all domain-addressed mail goes physically (uh, should we say that about zeroes-and-ones on wires and disks and ...) through that machine. Alain FONTAINE +--------------------------------+ Universite Catholique de Louvain | If your mail software barks at | Service d'Etudes Informatiques | my address, you may try : | Batiment Pythagore | | Place des Sciences, 4 | FNTA80@BUCLLN11.BITNET | B-1348 Louvain-la-Neuve, BELGIUM +--------------------------------+ phone +32 (10) 47-2625
mar@ATHENA.MIT.EDU (05/02/89)
We've been thinking of tackling this problem here at MIT. Our initial planning is as follows: * The full name of every member of the MIT community will be known to the mail hub. Mail sent to someone's full name will result in: 1) The mail is delivered if the name is unique and the person has a mailbox 2) An error response is generated saying "[full name] does not have an electronic mail address, please send mail to MIT Room ..., Cambridge MA 02139" 3) An error response is generated saying "[full name] is ambiguous, please choose one:" followed by a list of people giving the name, title, address, and a unique email identifier. 4) An error response saying "addressee unknown". * Every member of the MIT community will be given a unique identifier for email purposes. For most active email users, this will be their login name. For other people and those with name conflicts, it will be their initials and a number, similar to the NIC's whois database. This information will be kept up-to-date by Moira, the Athena Service Management System, and regularly updated on the mailhub. Users will be allowed to update some of their own information, and to become unlisted if they want to. Moira currently contains all of the necessary information for the students here at MIT, only the staff and remaining faculty must be added. The primary development effort will be modifications to the mail hub. -Mark Rosenstein MIT Project Athena Systems Development
cfe+@ANDREW.CMU.EDU ("Craig F. Everhart") (05/03/89)
The CMU installation of the Andrew system, andrew.cmu.edu, supports a name space of 8500 users. For an installation of this size, I believe it to be difficult or impossible to make somebody's unique ID correspond in a predictable way to their full legal name. (Some smaller installations, such as CMU's Computer Science department (cs.cmu.edu) with maybe 3000 users, feel that they can make the unique ID be the preferable Firstname.I.Lastname, so Andrew also supports that canonical format for such installations.) We tackled the problems of mapping name probes to the space of all names in a distributed manner, and came up with Andrew's White Pages, a name lookup service that can match probes using abbreviations, phonetic heuristics, and the like. The service runs via AFS on any workstation, not simply on a mail hub. We use it for mail delivery, as well, with results such as: - mail delivery to the named user - error response generated if the name probe was unique but only a fuzzy match - error response generated if the name probe was ambiguous; possible matches listed if there aren't more than a given number of them - error response generated if no match could be found. This has been in place for two years or more. In progress is a mechanism whereby people can update aspects of their own White Pages entries automatically, with optional administrative approval. Integrating larger lists of names, with optional mail deliveries such as paper campus-mail delivery, is a cute idea. We haven't pursued it very hard, but we think it could be fun. Craig Everhart Andrew message system
pritch@tut.cis.ohio-state.edu (Norm Pritchett) (05/04/89)
I'd like to thank all those who answered my query regarding the subject of this posting. For those who wanted me share what I found, that will be forthcoming -- I still have messages coming in at a steady rate and I'd like to wait for them to trickle off before I share. From the collective responses I got I was able to devise a pretty good scheme. I won't share it yet because some ideas are still being hashed out among some fellow networking folks on campus but if you are familiar with DND there's a lot of similarity to that. In my original posting I (intentionally) didn't present an accurate idea of the size of userbase we had to address because I didn't want to disuade some people from responding just because they thought their system wouldn't work for us. I mentioned 4 figures or larger in my message -- what we really need is a scheme that will comfortably handle a population in excess of 75,000. If some of you have thought about trying to develop such a system this large but have been disuaded for some reason or another (I've heard from a few such places) I think we've got something for you... stay tuned. -- Norm Pritchett, The Ohio State University College of Engineering Network Internet: pritchett@eng.ohio-state.edu BITNET: TS1703 at OHSTVMA UUCP: pritch@sydney.columbus.oh.us CCNET: ENG::PRITCHETT (6172::PRITCHETT)
mullen@itd.nrl.navy.mil (Preston Mullen) (05/04/89)
The Andrew Message System at Carnegie-Mellon University has a component
called "White Pages" that employs a fuzzy name recognition mechanism.
According to the author, Craig Everhart <cfe+@andrew.cmu.edu>, it
"matches name variants to people's names reasonably well without any
pre-identification of the possible variants of everybody's names."
(By the way, the + in his address is a flag that bypasses the smart
name recognition.) The Andrew Message System is built on top of the
Andrew File System, but the White Pages name recognition component is
easy to separate out.
When I asked about this in October, I was told that the software is owned
by IBM and that the licensing policy had not yet been determined.
I had hoped (and still hope) to use this kind of name matching in a
general approach like that recently suggested by <mar@ATHENA.MIT.EDU>
in his message to tcp-ip of Mon, 1 May 89 13:28:20 EDT. One might
want to set things up so that an address with exactly one match on
the wrong component (e.g., first name only) would result in a response
similar to the one sent for ambiguous names; in such a case, it might
be better to force the sender to confirm the intended addressee than
to deliver to the wrong person.
Preston Mullen
Laboratory for the Study of Human-Computer Interaction (Code 5530)
Naval Research Laboratory
Washington DC 20375
P.S. There is probably a better mailing list than tcp-ip for this topic,
but which one?