CURRENT_MEETING_REPORT_ Reported by Dave Crocker/Silicon Graphics Minutes of the Electronic Data Interchange Working Group (EDI) This group met as a BOF in Seattle. The meeting was the first face-to-face gathering of a group that had been conducting e-mail discussion for several months and was held to: o Introduce IETF participants to the EDI domain of discourse, o Review work from the mailing list, o Assess continued IETF interest in working in this domain and the priorities by which the mailing list has been functioning, and o Initiate new work. In particular, the MIME content type definitions were reviewed, and a preliminary discussion about the EDI-over-Internet Usage document was conducted. Attendance was substantially higher than expected (42 people) for a topic thought rather arcane to the IETF. A content-type naming contest was initiated, and completed the following day. There was a clear consensus for continuing this work as a formally chartered working group. Introduce IETF Participants to the EDI Domain of Discourse We had an extensive discussion about EDI background and some about Internet background. A bit of the mailing list history was reviewed, in terms of its representing what I called an ``aculturation process.'' This was a polite way of describing the head-butting and name-calling which resulted from many participants having plenty of experience to suggest that they held the correct view of how to do things when, in fact, we are trying to merge two, quite different ways of doing things and needed to find a basis for collaboration and compromise. Hence, my own assessment of the recent explanation for the complete list inactivity was that we were all catching our breath. (As well as waiting for me to produce the revised draft specification.) Review Work from the Mailing List We had previously identified four work items that are reasonable for near-term work, but agreed to defer the second two. Hence, our work is to: 1. Define MIME content types for carriage of EDI 2. Develop a ``usage'' document to assist the EDI community in understanding reasonable ways of doing EDI over the Internet The two items deferred for effort by a later working group (under the belief that work is best done serially): 3. Specification of EDI ``routing'' information (a la X.435) 4. Specification of mappings between Internet-based and X-400 based EDI Assess Continued IETF Interest Both the attendance and the participation in the BOF, as well as the extensive e-mail list history serve as a clear indication that this area is interesting and relevant to the IETF community and that there is productive work that is feasible within a reasonable timeframe. Initiate New Work We looked briefly over a new draft of the MIME-over-Internet document, being released as an Internet-Draft in parallel with these minutes. The three MIME content types it defines are EDI-X12, EDIFACT, and EDI-other. It was felt by the group that the last term lacked a certain je ne sai quois and should be replaced by a better term. This latter type is intended as a generic tag for all non-X12 and non-edifact which may be sent between trading partners having prior agreement. As an expedient, a contest was declared, for one day, soliciting suggested alternative names. The winning entry was EDI-consent. Initial discussion about the scope and structure of the usage document was begun, to be carried forward to the list. The presentation slide used to introduce this discussion follows the minutes. Dick Brooks raised a concern about the ability of MIME to carry binary EDI data. This has been a repeated concern on the EDI list, so it is worth mentioning, again, that MIME has a standard solution to the requirement, called Content-Transfer-Encoding:bin64, which converts the binary data into a form that can be carried over text-only links (i.e., the usual e-mail carriage mechanisms). Dick also solicited participation in a pilot project to field and test EDI over the Internet using the draft specification. This should be pursued over the IETF-EDI mailing list. Since the IETF only standardizes that which has been implemented, tested and used, such a pilot is an essential step along the standards path. Attendees Kevin Altis altis@ibeam.intel.com Perkins Bass bass@eskimo.com Scott Bradner sob@harvard.edu Lloyd Brodsky lbrodsky@rocksolid.com Dick Brooks d.brooks@ieee.org Rong Chang rong@watson.ibm.com David Crocker dcrocker@mordor.stanford.edu Dante Delucia dante@usc.edu Cheri Dowell cdowell@atlas.arc.nasa.gov Richard Everman reverman@ka.reg.uci.edu Erik Fair fair@apple.com Roger Fajman raf@cu.nih.gov Tina Feick feick@bnamf.blackwell.com Louis Fernandez lff@sequent.com Ned Freed ned@innosoft.com Shawn Gillam shawn@timonware.com Terry Gray gray@cac.washington.edu Jill Hanson jhanson@wsipc.wednet.edu Barbara Jennings bjjenni@sandia.gov John Klensin Klensin@infoods.unu.edu Andrew Knutsen andrewk@sco.com Paul Lambert paul_lambert@email.mot.com Barry Leiner leiner@nsipo.nasa.gov Edward Levinson levinson@pica.army.mil Laura McCarty lmccarty@pmel.noaa.gov Michael McLay mclay@eeel.nist.gov Mark Needleman mhn@stubbs.ucop.edu Michael O'Dell mo@uunet.uu.net Tim O'Reilly tim@ora.com Brad Parker brad@fcr.com Manny Pasetes ekp@utu.prememos.com Les Pennington les@cac.washington.edu Karen Petraska-Veum karen.veum@gsfc.nasa.gov Francois Robitaille francois.robitaille@crim.ca Jim Romaguera romaguera@netconsult.ch Michael Sorsen c02420MS@wuvmd.wustl.edu Einar Stefferud stef@nma.com Peter Sylvester peter.sylvester@inria.fr Phil Trubey ptrubey@netcom.com John Veizades veizades@wco.ftp.com