Dear TF members,
Here are the draft minutes from our last call.
Cheers
Ant
***
Draft Minutes – CoC TF Eighteenth Call
5 November 15:00 (UTC+2)
Attendees: Denesh Bhabuta, Athina Fragkouli, Antony Gollan, Brian Nisbet, Salam Yamout
Scribe: Kjerstin Burdiek
Summary: The CoC TF discussed the draft document prepared
by Antony and Athina which outlined both the CoC process and the
CoC Team formation. The TF discussed the relationship between the
CoC Team and the WG chairs, agreeing that the CoC Team had
authority over declaring violations, while the chairs oversaw
conduct in their own MLs and WGs. Beyond this, it was agreed that
they should not be too prescriptive; the two groups would need to
liaise to deal with any overlaps that arose. Regarding the overall
process, there were three parts, one concerning report management,
one concerning issuing consequences, and one concerning the
formation of the CoC Team. The TF discussed specific steps in each
section and added steps where needed. The TF also considered how
consequences would be determined and issued. They agreed that
consequences did not necessarily have to escalate before
high-level punishments were given and that the RIPE Chair might
need to be informed in the case of such a serious consequence.
Finally, Athina shared suggestions on how to protect the CoC Team
from legal damages, such as including a waiver in the RIPE Meeting
terms and conditions where participants consent to the CoC
process. There was an action on Antony to revise the draft after
the meeting in preparation for sharing it on the ML.
1. Minutes
The minutes from the last meeting were approved, noting one
correction Leo had sent to the list – update the introduction to
reflect that the action for him was to create a draft and share
this with him and Athina, which he had done.
2. Agenda
There were no changes to the agenda.
3. Actions
- (Ongoing) TF to think about how the CoC Team and the WG
Chairs would work together to manage discussions on mailing
lists. Including tools, e.g. Mattermost channel
Antony shared the document he and Athina had been working on, which was based on Leo’s initial draft. So far, they had focused on what sections were needed and how they should be ordered. Regarding this action item, Antony shared a section about the relationship between the WG chairs and the CoC Team. The draft stated that all community members were responsible for making people aware of appropriate conduct, but only the CoC Team was responsible for investigating reports and making determinations. WG chairs were not part of the CoC Team or its deliberations, but they were responsible for their WGs in a proactive way.
Salam said she did understand why they included this in the
document. Perhaps they could say that the role of WG chair had
already been defined elsewhere, and this document would not alter
it. However, the CoC Team might consult with the chairs in the
case of an incident on their ML or in their WG.
Antony asked if there might be cases where it would be better for
people to contact the WG chair rather than the CoC Team (for
example, if someone was spamming a ML the chair might just decide
to moderate them).
Salam said if that was the case, the CoC Team could reach out to
the WG chair to determine whether the chair would like a formal
case opened or would handle it themselves.
Athina said that people may be uncertain about the duties of the
CoC Team and the WG chairs. The WG chair role was indeed already
defined, but they could clarify the scope of the CoC Team. The CoC
Team could tell the person making a report if this was within
their scope or if it was within the WG chair’s. There used to be
separate CoCs for their meetings and for the mailing lists. Per
the RIPE ML CoC, the WG chairs could issue bans and other
consequences if needed. The CoC TF needed to clarify if WG chairs
still had these abilities and if the CoC Team needed the WG
chair’s permission to ban someone on a ML. The TF needed to
determine which party would apply any consequences.
Denesh asked if they needed to clarify who had more authority in
CoC breaches. There might be cases where the CoC Team did not
consider a certain act to be a breach, but the reporter then went
to the WG chair for a second opinion. This could become circular.
Brian said the process would be complex. There was an analogy
here between the WG chairs and the Programme Committee, where
defining their scope had taken a long time. They had found a
reasonable model, but it had changed a lot from how it started.
Regarding this document, they should say that the CoC Team and the
WG Chairs must liaise where there is an overlap. An act might not
be a breach of the CoC and yet still merit some kind of
consequence. An example from a different event was when someone
had worn a sword as party of their costume. This wasn’t a breach
of the CoC but was a breach of the convention policy, which was
the Operations Team’s responsibility. When needed, the two teams
convened to discuss how to handle certain incidents. Similarly,
spamming a ML might not be a CoC breach but could be handled by
the WG chair issuing a ban. However, the CoC Team should perhaps
have the power to overrule the WG chair on what constituted a
breach, and this might be where you brought in the RIPE Chair Team
to mediate and discuss. But generally, they had different areas of
authority, and the WG chairs should want to promote the CoC.
Finally, a separate document for the CoC Team responsibilities was
not necessary because the first document had outlined them.
Salam agreed with Brian. The document should state what WG chair
responsibilities were and that they should liaise with the CoC
Team in case of an overlap.
Antony asked if they should remove this information from the
document.
Brian said parts were redundant and could be removed. It was good that the document didn’t prevent by default someone who was a WG chair from being on the CoC Team. He was wondering whether there needed to be a formally appointed liaison or if that could be informal.
Antony asked what would happen if he misbehaved in a WG session. The WG chair might ask him to calm down or change his behaviour, which made sense as they were responsible for that WG. He asked if a CoC Team member would have the same authority.
Brian said it depended on whether a complaint had been made.
Antony asked if that meant the role was purely reactive.
Brian said yes. However, that person on the CoC Team might raise
the issue, and he would hope they would discuss this with the
chair. He thought a single CoC Team member should not be the one
ordering someone to stop certain actions, the wider team should be
involved.
Antony asked if they should make that point in this part of the
document or leave it out.
Brian said to leave it out.
Athina said they should revisit it. During the process, they
might decide to rethink the relationship between the WG chairs and
CoC Team during investigations. But if the case concerned
behaviour on the ML, the CoC Team must liaise with the WG chair.
Salam said they should not reiterate what everyone’s role was, as
it would open up more ambiguity. If there was no complaint, the
CoC Team would not be involved. She said they could should revisit
this later.
Denesh suggested moving on and clarifying the wording later.
- (Ongoing) TF to think about the appropriate level of process
to allow further improvements to the CoC.
There were no comments, and this action remained ongoing.
- Leo to draft a working document for Athina and Antony to
update for discussion at the next meeting
Antony returned to the document and said they had restructured
this from Leo’s initial version. There were three large sections.
The first section outlined the process for managing reports and
the steps involved. The second, on consequences, was inspired by
the Python consequences ladder they had looked at on their
previous call. The final section covered the formation of the CoC
Team, from recruitment to dismissal. He suggested going through
the document with the TF.
4. Review Draft
Antony noted that this was still a rough draft. So far they had
focused on what they wanted to say broadly and the overall
structure.
Denesh suggested they go through each section and then could then
review this in more detail ahead of their next call.
Antony said for the reporting section, they would mention the
ways people could make reports. Then they would describe the ways
reports would be received and subsequently handled. They would
also cover how a reporter would be notified that their report was
under review. On-duty team members would look for conflicts of
interest and recuse themselves if needed. The Team would consist
of three people, two volunteers and one RIPE NCC member.
Brian said he was not in favour of automated messages about
reports being received, especially for these sorts of matters.
Antony said they could use both, first an automated response and
then a person responding again once the report had reached them.
Brian said the first message could state that the CoC Team had
received it, and then the Team could decide which member would
handle it.
Salam asked if they should describe what an on-duty team member
was, as the TF had not yet determined how these team members would
be chosen. This section outlined the ticketing system, so the
on-duty team member description could go elsewhere.
Antony asked if she thought this was too detailed.
Salam said she thought this information belonged in the third
section of the document, which could outline how the team was
formed.
Athina said she understood Salam’s point but said the section
titles were unclear. In the section referring to the report being
received, which described the period up to the investigation
beginning, it might be useful to mention that people with
conflicts of interest would not be involved in the investigation.
If they did not state that, it might give the impression that
those people would still be involved at the start of the
investigation and only later excluded.
Antony asked if there was actually another step in the process
and if that would constitute a new section.
Denesh agreed there was another step there – after the receipt of
a report.
Salam suggested the steps were: 1) submission of a report, 2) the
process between the report submission and the start of the
investigation. This second step was where team formation and
conflicts of interest would be covered.
Brian agreed that the team should be selected before the start of
an investigation.
Antony suggested a new section for group formation. For the
report evaluation step, there would be three people evaluating:
two volunteers and a RIPE NCC professional to assist the
volunteers.
Salam said this repeated other sections. She also asked for
clarification on another step and whether it meant a report was
evaluated or a decision was issued.
Athina said they needed to agree on the term used here, either
determination, ruling, or verdict.
Brian said he liked decision or determination, as they were
determining whether there was a breech. Verdict sounded like a
judgment of the person rather than the breach.
Denesh said the evaluation section needed to be reworded in that
case.
Antony agreed and asked if there were any other comments.
Salam asked who would be notified of the decision.
Antony said there could be a notification step, perhaps for
staff.
Salam asked why this was not in step four, as it concerned
enacting a decision.
Antony said this depended on how they split up the sections.
Salam said fewer was better.
Brian said he preferred taking action as a separate step because
it was important and could have serious consequences. It should be
highlighted.
Salam said they could have these as two sections, especially if
there could be a process in between them, such as an appeal.
Denesh agreed that an appeals process was necessary to include
and that there should be a separate action step.
Antony asked if the determination was completed solely by the CoC
Team or if the RIPE Chair would also be involved.
Brian said that depended on what level of action needed to be
taken. A minor consequence did not need to be announced, but a
serious consequence like banning might need to be issued by the
RIPE Chair and possibly high-level RIPE NCC staff.
Denesh said this also depended on how much those people wanted to
know. The TF might need to ask them to what extent they wanted to
be informed.
There were no comments on the step for recording evaluations.
Antony moved on to the step on potential consequences. Leo had used the ladder concept from the Python framework; did they want to use an escalating ladder structure or just have a list? They did not want to imply higher consequences required crossing lower consequences as well.
Salam said the way it was currently worded suggested that ,
people might be banned from the meeting but could still go to
social events. She asked what happened in the case where someone
was grossly inappropriate (i.e. a sexual predator).
Athina said the section mentioned being banned from an event and
from future events. She agreed they might need to clarify.
Brian said the social event should be mentioned specifically.
There also should not require there to have been prior incidents
before consequences could happen. A single violation could be
severe enough to merit a ban. The TF should make it clear that
escalation was not required.
Salam said this was clear. She asked what should happen if there
were harassment at a venue. Would they be banned from social
events?
Brian said that points four and five clarified they could be banned temporarily or permanently.
Salam said they should change the wording to clear up the
confusion.
Brian said this partly depended on the RIPE NCC’s contract with
the venue. If they had booked the hotel, the event planners could
get the hotel managers to cancel the stay of violators. However,
it might be that they would have to just ban people from event
spaces. This step had more to do with implementation.
Denesh noted there was still work to be done on this and asked
Antony to share the draft so they could discuss it at the next
meeting.
5. Personal Risk
Athina said the evaluators should be protected from potential
legal damages. Leo had asked how they could protect these people
to prevent volunteers from being discouraged. They had dealt with
similar situations in the arbitration procedure and with the
Executive Board CoC Team. The participants of such an evaluation
must waive their rights to damages caused by the evaluation. The
CoC TF could get this waiver by including it in the terms and
conditions for participants at meetings, similar to previous
waivers about meeting conduct. For MLs, they could add a text
similar to the one about unsubscribing. This would remind people
of their obligation to adhere to the CoC and other related
documents, such as the CoC Team procedure. These methods would
limit the risk, but lawsuits were still possible. Another solution
could be having liability insurance, but they would need to
investigate this further.
Brian asked where that text would be and suggested they look at
what other communities did. He had not seen similar situations,
but in the case of RIPE attendees, it was true they could argue
their careers would be affected if they were banned. However, the
MLs were less of a concern.
6. AOB
Denesh suggested sharing the meeting recording with Leo.
Antony said he would share it with him.
Denesh said he would not be available for their next call on 19
November.
5. Actions
- (Ongoing) TF to think about the appropriate level of process
to allow further improvements to the CoC.
- Antony to edit the draft document to prepare it for sharing
on the ML.