feedback-2019-06-04.txt

Overall, this is an interesting paper and provides useful guidance for
open science, open source communities of effort. It is interesting that
these rules are based on fairly young (few months - few years) and few
dozen participants. Perhaps the focus on newcomers is because that is
where the primary focus of these communities is right now. I would
encourage authors to also consider what happens when these communities
mature and what happens when they scale to 100 or 1000 participants? As
a community manager it would be even more useful to me if there was a
suggest 'best practice' for each rule or set of resources to explore to
make these rules more immediately implementable. It would also be
interesting if after this paper is published if communities that agree
'endorse' these rules and implement or share how they are fulfilling
each rule -- a community seal of approval?

Below are comments on each rule followed by suggestions for merging
rules and reordering rules:

Rule 1: Code of conduct is important, but the culture that it defines is
critical. This is the current trendy item to mention, but a code of
conduct is just one tool. Key questions here are: How do communities
create a culture of collegial, open, sharing? It isn't with just a code
of conduct at the start. It is also speaking up, if you see something
say something. It is modeling appropriate behavior. People participate
in open science or open source because there is something of personal or
professional value or gain for them. Would also suggest tightening up
the issues around reporting. Many communities do this badly and these
points are good, but a little fragmented. May also suggest terming this
community guidelines or rules of the road.

Rule 2 -- Make governance explicit is a good point, but muddied with
citation and licensing. This reviewer recommends a separate rule on
community acknowledgement (see notes on rule 10).

Would shy away from benevolent dictators as part of any governance
because the risk is that project then lives and dies with one person. If
this is steering communities towards best practices, identification of
how someone becomes a member of the project to have a voice, what
decisions are made collectively. Ideally, open source projects want the
entire community to feel shared ownership.

I might also suggest that simple and sloppy like the internet -- have
governance that can evolve as the organization grows and matures and
don't make it too complicated or heavy too soon. Borrow what you like
from other organizations and be transparent as your org evolves. It is
easy to let this be a critical stumbling block of not moving forward.

Rule 3 -- Be welcoming -- suggest this is swapped with rule 1. Be
welcoming should be first rule because that is critical to growth, Rule
2 -- have explicit governance and Rule 3 -- part of the governance is a
code of conduct. Onboarding guidance and role of veteran members is
great idea.

Rule 4 -- This rule should really emphasize, what's in it for the
newcomer. Newcomers must have a compelling reason to participate and
there must be more value in doing this collaboratively than alone.

Rule 5 -- See Rule 9.

Rules 6 & 7 potentially combine 6&7 shortening 6 with just paragraph of
making it easy to find.

Rule 8 -- May consider meet-ups with collocated larger events as an
option for projects. Many projects hold workshops/training/demos with
larger conferences where their attendees may be. This rule confuses
gaining new members as purpose with meeting face to face for governance
or strategic direction decisions.

There may be an element of marketing your project here -- are there
slide decks, stickers, postcards etc that explain simply what your
project does. What have successful communities done that help growth? I
think members are empowered to champion the community and the best way
to gain relevant newcomers is when their peers bring them in. What else
do you do to market and grow your community? Are there virtual ways to
do this?

Rule 9 -- Suggest combining rules 5 & 9. With Rule 5 taking the lead and
this being a subpart, installing could be an LPP.

Open source projects developing code to support open science are a
further niche to just open science projects that are not building
infrastructure or maintaining code. The former seems to be what this
paper is geared toward and this rule is emphasizing that. The former
also has a 'product', the later may not have a product and with no
product it is much harder for newcomers to find value or reasons to
contribute. May consider clarifying the scope of these rules to open
source, open science communities developing code or think of way to
broaden this rule.

Rule 10 -- Most of what is included here is important as part of Rule 3
-- be welcoming.

I would broaden this last rule to acknowledge ALL contributors to your
project.

Points about celebrating first contributions may fit in Rule 4 and
points about mentoring may fit in the welcome rule or rule 6 about
navigating the community documentation.

I would recommend ending this paper on individual and collective
acknowledgement. Citation of research objects is a new area where
software is leading with software citation, but open science communities
are generally not cited. Without a foundational community paper, the
community is not getting credit.

Further things to consider: How do you create a culture where peers
acknowledge each other, that individuals contributing get credit to
further their careers and that the project as a whole also increases its
value and the positive reinforcement draws in new members who want to be
associated with this work, want to use the tool, etc.

**Suggested reordering and combining of rules:**

Rule 3: Be welcoming.

Rule 4: Help potential contributors evaluate if the project is a good
fit. (Why contribute in the first place)

Rule 2: Make governance and licensing explicit. -- Remove citation and
include with Rule 10.

Rule 6: Make knowledge findable and keep it up to date.

Rule 7: Provide an easy, complete, and up-to-date guide to contributing.

Rule 1: Have and enforce a code of conduct. (once you have explicit
governance and easy to find guidance then make sure your code of conduct
is visible). This sets the stage for contributions and participation.

Rule 5: Develop forms of legitimate peripheral participation

Rule 9: Make it easy for newcomers to set up.

Rule 8: Use opportunities for in-person interaction, but with care. (Let
everyone feel like a project ambassador/owner);

Rule 10: Follow up on success. (Broaden to acknowledge ALL
contributions)

With this combination of rules, it leaves space for potentially a few
additions. I would suggest considering rules on failures in the
community or newcomer failures. The authors may also think about a rule
on evaluation of community? How do you know if you are successful? Is it
just \# of newcomers or sustaining membership beyond the newcomer phase
or sustaining the project. This is an evolving area of research, but
would be good to flag that more is not always better.