Initializing the v2 COSEP, to get ready for the 2018 white papers

[drphilmarshall]

Dear v1 @LSSTScienceCollaborations/observing-strategy-white-paper-contributors ,
v1 @LSSTScienceCollaborations/observing-strategy-white-paper-editors
and anyone else interested,

As @rhiannonlynne pointed out on community.lsst.org here, the 2018 call for cadence optimization white papers anticipates that "the performance evaluation components of the white papers ... will ... be added to the COSEP, to provide a comprehensive reference point for survey strategy
evaluation.” This PR is to get us started on the path to a version 2 of the community observing strategy evaluation paper, by adopting the term "COSEP" and preparing the paper's structure to absorb the 2018 white paper submissions.

I suggest we make the v2 COSEP in 3 parts:

1. Simulations - including the old cadexp2 chapter, and a new one on the sims made for the 2018 WP call. OpSim has evolved a fair bit since v1, so it'd be good to describe that briefly here.

2. Science Cases and Performance Metrics - with chapters on AGN, Variables, Cosmology etc as in v1, to capture the WP's sections on performance evaluation.

3. Proposed Observing Strategies - a new part, with chapters organized by WFD, DDF etc., to provide a summary of the WP recommendations.

In this scheme, the different sections of each white paper would get split up and redistributed between parts 2 and 3 - but would retain their author lists. Then, the SAC chapter summarizing the white paper content and recommending the needed properties of the next (final!) round of simulations would go at the end of part 3 (and be the final chapter of the v2 COSEP).

Assigning this to @michaelstrauss and @rhiannonlynne to review, but comments welcome from all.

Best wishes,

Phil

[michaelstrauss]

I like what you have written. I would suggest that we not simply copy Chapter 2 (cadexp2) of the previous COSEP into V2, as it will just confuse people who want a clean story about the status of the simulations. The earlier version had a lot of "aspirational" words about, e.g., rolling cadence, that now is implementable with the feature-based scheduler, and I think we can tell a cleaner story without all the detours that were taken at the time V1 was written.
In Section 3, I could imagine we would include both the suggestions on cadence found in the white papers, and the summary and discussions of next steps that the SAC would produce based on them. The latter will be a public document, I believe.

[drphilmarshall]

Thanks @michaelstrauss ! Excellent suggestions. I agree we can make Chapter 2 much more up to date and relevant; most of the v1 analysis only used the minion_1016 baseline cadence simulation, and the others can be included via a very terse description (or even by reference to v1 or something). And it'll be great to have the bulk of the public SAC report included, too. I think the v2 COSEP will have a very useful "compendium" feel to it.