An editorial process for the REMARK directory

[sbenthall]

Historically, we have included many "potential REMARKs" in the REMARK directory, even when these are not up to what we might consider the REMARK standard.

This is partly because the standard is a moving target. That problem would be fixed with a release process. See #104

The other reason for this problem is a conflict of interest in the construction of the REMARK directory. It is serving multiple roles. In some of those roles, we have wanted to include as many REMARKs as possible. In other roles, we have wanted to raise the standard of quality. The former tendency has historically won out over the latter.

This is preventing the REMARK directory from being what it is intended to be, which is a collection of vetted computational artifacts demonstrating best practices in computational economics. This is an institutional issue, not a technical issue.

A solution to this problem would be to separate the interests within the REMARK system by introducing a robust editorial process. Most journals have editors and reviewers that are different from the journal article authors.

This is the editorial process of the Journal of Open Source Software (JOSS). It is more elaborate than we need at this stage, but it's presented for comparison.

https://joss.readthedocs.io/en/latest/editing.html