The summary of what happened...

Stan Barber, Co-Chairperson and Editor

Prologue

The draft came out in time to be processed by Internet Drafts prior to the meeting. There was 300 messages posted to the list over the course of the time between the publishing of the list and the meeting. This was 1/5 of the traffic on the list since the group started. There was good attendance at the WG meeting, but only three active participants: Stan Barber and Ned Freed as co-chairpersons and David Riley from software.com. This was unfortunate, but the work of the group was conducted and those results are discussed in this document. While it is hard to call this consensus, the chairpersons agree that these decisions are now made and it's time to move on to other matters.

Summary of Current Work

The Base draft has been updated and published by Internet Drafts. There needs to be an update to the authentication draft. Ned will query the current author to see if he will do an update. If not, Ned suggests that the draft be reassigned. I say that I will await work from Ned before seeking another to work on this draft.

Clarification of the wildmat format

There were two sets of text submitted for consideration other than that which is in the current draft. Clive Feather adds  "\u" and "\U" to make it possible to do raw encoded strings. Russ Alberry offered the documentation he is writing for the version of wildmat he is refining for INN. The group decided that Russ' text should be used as a basis for the next draft. 

Clarification of wildmat usage in various commands

There has been some discussion about how to deal with compound wildmats versus multiple wildmats. The group decided that commas can be used to link together wildmats to form a compound wildmat. A space would separate multiple wildmats. Spaces can be embedded in a wildmat if preceded by a backslash. If a command can have multiple wildmats (or compound wildmats), the evaluation of each wildmat is treated independently (i.e. a logical OR).

MODE READER

There was no consensus here. The group discussed that it could be used to partition commands between a "reader" server and a "feeder" server. This does parallel certain vendor products. David Riley is supposed to propose some text on the list for further discussion.

Replacing me as editor

There was no support for removing me as editor. I will release the next draft by the end of August and will do my best to release a new draft each month in which there is evidence of substantial changes.

Other business

Effective immediately and until the work of this group is completed, suggested changes to any of drafts (other than those that have to do with spelling and grammar) must be accompanied by replacement text. The text does not have to be super polished, but it must be clear enough to support the point-of-view of the advocate(s) for change. Expressions of support for changes must be explicit as well as expressions of support for no change. The final decision on what gets changed remains that of the chair. 

The draft relating to current practices will be considered by the IESG real soon now.

The charter of the group is primarily to clarify RFC977, not to invent a new protocol. Additions, even those that are codifying current practice, have to be added only as a last resort. These things can always be proposed as extensions.

David Riley believes that CHECK/TAKETHIS is important and should be considered for the base draft or as a specific extension. He will work on some text.

Respectfully submitted,

STAN BARBER