Difference between revisions of "Harvard-2017:C3"

From CMB-S4 wiki
Jump to navigationJump to search
Line 4: Line 4:
Post talks here.
Post talks here.
*  [[File: s4-harvard-pubs-parallel.pdf | Discussion Slides ]]
*  [[File: s4-harvard-pubs-parallel.pdf | Discussion Slides ]]
*  [[File: s4-harvard-pubs-report.pdf | Report from Sessions ]]
*  [[File: s4-harvard-pubs-report.pdf | Report from Session ]]
== Notes ==
== Notes ==

Latest revision as of 11:00, 25 August 2017

Back to Harvard-2017 main page

Parallel Session C3: Publication Policy (Chair: Nathan Whitehorn) [Jefferson 256]

Post talks here.


What are the goals? 1) good science disseminated widely 2) appropriate credit assigned 3) protect younger people 4)somehow recognize critical dirty work (e.g., calibration) and don't overemphasize certain type of contributions

Q: Is there any commonality to exactly how paper splits are drawn (particularly difference between papers that need to be reviewed and those that don't)? The publications policy could end up discouraging mentioning CMB-S4 if it forces the paper to go through review.

A: Some collabs set the requirement by whether the paper uses unpublished collab data or private collab software. Maybe need window after public data releases.

Options on who is an author on a given paper: 1) Alphabetical opt-out, with corresponding author as pubs. committee 2) Alphabetical opt-out, with corresponding author as manuscript author 3) Alphabetical opt-in 4) Short list first authors + alphabetical opt-out 5) Some combo of the above, based on paper “type."

How do you construct the author list to begin with: 1) Every member is an author, on papers after ~6months, off after 1-2 years. Sometimes tied to a service obligation (shifts). Universally used in extremely large collaborations (HEP). 2) One list that is on in perpetuity (e.g. builders), other list (e.g. analyzers) that comes and goes with specific papers. 3) Authors justify themselves on each paper. (We should not do this).

Talk policy: Can you give credit to the appropriate people by having them give talks? For example, this is how HEP people do it. Flawed as well. This is just another way to address the “first author” problem. We need a talk policy.

What should the collaboration review process be?


— Want to encourage CMB-S4 collaboration papers soon. Need to have some collaboration review associated with it. We need some internal organization soon to do this. How to do this without discouraging publication in the near term?

— Need a CMB-S4 talks committee soon. Need a slide deck. Be a clearinghouse for conference invitations.

— Is there any precedent for talks committees to enforce quality of talks? Common to require distributing slides beforehand to collaboration.

— Combination of strictly alphabetical author lists and builder list leads to bulky author lists

— Is it possible to make up for the downsides of a given choice in author policy with a specific talk policy? All these decisions go together.

— There is some advantage in simplicity.

— AI: Make a speakers committee! Unanimous agreement.

Action items/Next steps

  • Recommend immediate formation of a CMB-S4 speakers' bureau
    • Collect and solicit conference invitations
    • Start giving talks specifically on S4 as "XXX for the CMB-S4 Collaboration"
    • Provide a set of slides/plots to use in talks
    • Review/distribution of S4 talks by/to the collaboration
  • Publication of papers with "CMB-S4" in the title encouraged immediately
  • Start setting up an authorship and publications policy soon, but not right now
    • Review improves quality of publications, awareness of results to wider collaboration
    • Publishing with full author list, in some form to be determined, demonstrates that CMB-S4 is a collaboration and that a lot of people care about it