Re: [Rfced-future] Model proposal

[Eric Rescorla <[e-mail]>]

On Mon, Jul 6, 2020 at 9:19 PM Martin Thomson <[e-mail]> wrote:

> I do recall several lengthy discussions about the number of authors on a
> document, but as these were recurrent and inconclusive, I am of the view
> that these are perfect for a working group to strategize on.
>

I'd like to focus in on this example for a moment, as it also consumed
a fair amount of IESG time, and I think it shows some of the
complexity here.

Authorship conventions vary fairly significantly even within
genres. For instance, in computer security alone, I am familiar
with at least three primary author ordering rules:

- Students first, ranked from most to least contribution, followed by
  professors (sometimes ranked loosely from least to most
  supervision/engagement). [a lot of Security]

- Students first, but loosely ranked by increasing need to find a job
  as well as contribution, then professors as above
  [Systems-influenced Security]

- Alphabetical [Crypto]

If you go outside computer science into broader paper publishing
you will also see the following (among others):

- Giant author lists [especitally particle physics [0]]

- Detailed taxonomies of what contributors/authors did
  [medicine [1]]


My point here is that authorship policies depend on what values the
community in question is trying to promote (recognizing people's
contributions, advancing their careers, precisely delimiting what
everyone has done, having a single point of contact, etc.). And when
we discussed these topics in the IESG, we discussed similar
considerations. Thus, while these discussions certainly do benefit
from understanding the perspectives of other communities deciding on
the right policies is more about community objectives than it is about
publishing expertise.  Indeed, it seems possible that we might develop
different policies for different streams depending on which objectives
are most important to those streams.

-Ekr


[0]
https://www.nature.com/news/physics-paper-sets-record-with-more-than-5-000-authors-1.17567

[1]
https://www.elsevier.com/authors/journal-authors/policies-and-ethics/credit-author-statement