[rfc-i] Documents at AUTH48 which have been cleared by all authors

Bob Braden braden at ISI.EDU
Tue Aug 15 10:45:48 PDT 2006



  *> 
  *> I'd speculate this might be due to at least one of:
  *> 
  *>   a) RFC-editor hasn't context-switched back to that document yet, and 
  *> marking "all AUTH48 signoffs received" (e.g., removing 'AUTH48' from 
  *> queue designation) has no priority in the document handling, and/or
  *> 

Pekka,

The RFC Editor staff works only 5 days per week, and occasionally
illness or vacation may delay state changes by a few days.  But the
answer is generally (b).

  *>   b) Documents which refer normatively to other documents in the queue 
  *> which are also at AUTH48 don't get moved to any one other state until 
  *> all the documents get AUTH48 cleared because there is no intermediate 
  *> state past AUTH48 to move into [1].
  *> 
  *> Hence, for higher AUTH48 timing precision and greater visilibility of 
  *> who has the token on what documents, it might be better to have a 
  *> clearer delineation between "in AUTH48, waiting for authors" and 
  *> "ready to publish (but possibly waiting for something to happen to 
  *> other documents first)".
  *> 

This is a good idea.

There is a growing problem of managing the AUTH48 state.  We would like
to create a web-based interface for authors to sign off and to keep the
signoff status of each AUTH48 document.  That would include the
notation about normative references.  Whether it is worth adding a
whole new state is questionable.

The RFC Editor queue does show when there is an (explicit) unresolved
normative reference for a document in AUTH48.  I see that 15 of the
documents in AUTH48 are currently marked this way.  In fact, this is
incomplete -- we show only EXPLICIT dependencies, and there are often
several levels of indirection in the normative reference dependency
chain.  We have not yet tackled this level of complexity.

We appreciate your input.

 
  *> Btw, mailto URL in http://www.rfc-editor.org/queue.html is incorrect.
  *> 

OOOPS!

Bob Braden for the RFC Editor


More information about the rfc-interest mailing list