Re: Soliciting feedback: starting a satisfaction survey for the RFC Production Center and Publisher

[Michael Richardson <[e-mail]>]

Heather Flanagan <[e-mail]> wrote:
    > In order to get a better sense of how well the RFC Production Center and
    > Publisher are serving the needs of authors, I would like to start sending out
    > a short survey to authors immediately after the publication of their RFC
    > starting on February 1, 2019. Participation in the survey is entirely
    > optional, but the expectation is that requesting feedback immediately after
    > publication will be more effective than the current process of randomly
    > selecting a percentage of authors who have published in the previous year and
    > asking them to recollect what did and did not work for them.

I appreciate this effort, and I don't want to take away from it.
I think that we have a bigger production problem that involves bigger
documents.  I was recently involved in the ~3month AUTH48 effort around
RFC8415 (DHCPv6bis)...  The RFC Production Center did a totally awesome job.

The question is really: should we be producing such long documents?
At 154 pages, it's not even close to what I understand NFSv4 was.
Are we any good at this?  Are there any processes/proceedures/tools
that we could learn from the fiction publishing business?

Could we start some of the editing process earlier?
Would it help during IESG review?

    > The proposed questions in the survey are:

    > 1 Did the editing service provided by the RFC Editor improve the quality of
    > your document? [Yes|No]

But, to your question, I suggest that you ask a few extra questions.

* How long was it from WGLC to the beginning of AUTH48?
  (You know how long AUTH48 is, and also how long it was from IESG/ISE
  approval and AUTH48, but not how many cycles at IESG there were)

* Were there outside issues that made it hard to respond to AUTH48 in
  a timely fashion?  For instance, if the document requires a lot of
  mental context, was it hard to "return" to the document to finish it?

* Did the document remain in MISSREF for a long duration?

* Were you aware of where the document was in the queue?  Did you look?
  Did you get adequate notice as to when AUTH48 might occur?

* Did you have difficulties reaching all authors for approval?

--
Michael Richardson <[e-mail]>, Sandelman Software Works
 -= IPv6 IoT consulting =-