Sender: minutes-request@ietf.CNRI.Reston.Va.US From: Stan Barber Date: Fri, 12 Jul 1996 00:35:59 CDT To: minutes@CNRI.Reston.Va.US Subject: MINUTES OF THE IETF BOF NNTP BOF Minutes Prepared by Stan Barber, with assistance from John W. Noerenberg [Please see the slides for the original contents presented by Stan Barber.] Stan Barber called the meeting to order and provided a basic agenda and overview of RFC977 and work that had gone on before. He presented the original description concerning this BOF and discussed a possible time line and deliverables should a working group be formed. These took the form of two documents. One would be a revisions of RFC 977 that would clarify any ambiguities in the original document and provide a specific mechanism to add standardized extensions in the future. There was some confusion in the group concerning exactly what Stan meant by this. His intent was to to establish both the mechanism of adding verbs for new standard extensions, provide the server with a mechanism to inform the client about which extensions (both standard and proprietary) it supports and suggest a mechanism for establishing such extensions as standards from the IETF perspective. From the discussion within the group, it appeared that there was a clear consensus that providing a mechanism for the server to inform the client about which extensions it supports was important. It was not clear that there was consensus on the need to establish such extensions from the IETF perspective and the mechanism for establishing such verbs was not significantly discussed. Additionally, Keith Moore, one of the Applications Area Directors, expressed his feelings that a revised RFC977 that did not include accepted current practice would probably not be acceptable to IESG and suggested that the two documents be merged into one. Stan said that he thought that using the second document to prove out the extensions mechanism described in the first document would lay a good foundation for future extensions work. There was not clear consensus for either approach, but since Keith said the IESG would not accept the document as proposed, Stan revised his proposal along the lines that Keith suggested and said that perhaps other new extensions like free-text searching, replication or streaming might be used as test cases to test the new extensions standardization mechanism. There was also some concern that the IESG may not permit the formation of the working group due to concerns that it may be hard to find consensus in the NNTP community. It was stressed that that the charter should be clear and the amount of work have be limited to realistically attainable goals. Stan proposed that a draft document be published in mid August and discussion about problems or revisions in the document be done on a new mailing list set up as part of the WG creation process. Drafts would be updated to reflect changes on which consensus had been reached and at the San Jose IETF, a document could be voted on that might move onto the regular IETF standards track. Keith said that would probably be a suitable approach, but it would basically have to start at the beginning of the process and would not really be advancing RFC977 as it currently exists. That would mean that interoperable independent implementations of the new protocol would not necessarily have to exist by December, but would need to exist by the IETF after that. The Internet Software Consortium, NetScape, the "reference implementation" for Unix and Microsoft are all interested in pursuing this process and so having such multiple implementations at the appropriate time would not be a problem. The group was asked if they felt there was a need to create a working group. There was clear consensus that there was. The group was asked about whom should be working group chair. Stan Barber volunteered, but Keith and Paul Vixie both said that would rather Stan served as a document editor. As there was no other candidate forthcoming at the meeting, this issue remained open. Until it is resolved, Stan Barber will provide coordination for the group. The charter of the working group would involve three goals: 1. Revise and publish a standards-track successor to RFC 977 that would remove ambiguities from the original document, add a mechanism for adding standards-track extensions to the protocol and providing a mechanism for the server to inform the client of which extensions it supports. 2. Include in the same document some reasonable group of existing commonly used extensions forming a new base functionality for NNTP 3. Upon completion of the RFC977 successor document and with the approval of the Applications Area Directorate, select a new standards-track extension for NNTP and test the newly established mechanism for adding extensions. A mailing list as been established for this work at ietf-nntp@academ.com. Subscription to the list can be done by sending a regular internet-style subsciption message to "ietf-nntp-request@academ.com". Archives of the discussion are available at this URL: http://www.academ.com/academ/nntp/ietf A general information page on this work is available at this URL: http://www.academ.com/academ/nntp/ietf.html Other archive sites for the mailing list will be announced at a later date. Respectfully submitted, STAN BARBER