The aim of this document is to describe the culture, procedures, and baseline expectations for the admin team of the LGBTQ in Technology slack for general consumption and context, to clear up possible misconceptions about the power and authority admins have and how we exercise them.
There are some admins who are less technical and some who are more technical. By default, we are set as owners of the slack when admitted to the admin team, and added to the private channels #admin and #admin-signups. Owners of the slack have far-ranging powers including the ability to lock user accounts, both permanently and temporarily, and to delete, but not edit, other members' messages.
#admin is a normal private channel where we discuss admin business. It has the same message retention policy as any other channel in the slack.
#admin-signups is a Wheelie-driven channel where admins can see and approve or decline signup applications and referrals.
#announcements is a public channel every member listens to, but only administrators can post to.
#admin-log is another public channel every member can listen to (opt-in) but only administrators should post to. It is for any details associated with formal admin notifications about policy or member interventions. Admins may delete any non-admin posts to this channel and redirect discussion to #meta at their option.
When a new admin decides they are ready, they can ask and will be added to our more technical assets, which include:
- The GitHub Team
- The GitHub Repositories
- This includes, crucially, the GitHub Pages site that drives the slack website, but also includes codebases for most of our customized functionality.
- The private admin Google Drive folder
Note: Some admins may not have access to the full set of resources because of different issues like international access, country specific firewalls and surveillance, privacy issues, and other social, political, personal, and technical limitations, but across the whole team of admins, many admins will have access to each resource.
The LGBTQ in Tech Code of Conduct to the extent possible, binds all members of the slack, including the admins. The admin team does not hold itself above the code, and we enforce it with membership. This administrator policy and procedure document is an additional policy that applies to the admin team above and beyond the Code of Conduct. But we follow the Code of Conduct as well.
Admins frequently observe incidents that we do nothing about, largely because non-admin members are already doing heavy lifting, which we very much appreciate, discussing an incident in light of the CoC, and giving community and personal feedback about how the behaviors should be avoided, the language refactored, and, if needed, amends made. In cases where this happens, we silently cheer and stay out of it, because community correction is better in the long run, sometimes than when we show up and intervene. That said, we always try to be on duty, and stand ready to assist. Use the /admin command or DM an admin directly if you want assistance. We are here to help.
Although admins have far ranging technical ability and power, the admin team uses a collective, collaborative, discussion-based process for decision-making and recommendations. An admin who would go rogue and act alone is one we hope we wouldn't add to the team in the first place, but if one slipped through, acting on major issues solo without a very good reason they would likely be expelled from the admin team.
For any operation that's not simple and trivial, we discuss issues, incidents, user behavior, and any suggested or planned intervention with each other thoroughly in the private #admin channel before moving forward with interventions.
The role of admin requires good judgment and careful consideration. We generally do not make quick decisions and we are very careful to discuss major interventions and policy changes as a group before taking large-scale actions that will affect the entire slack.
There are situations where we must make quick decisions, even unilateral ones, and where we have to, we will, and it tends to make each one of us uncomfortable to make moves entirely unilaterally.
With all this said, we make it a goal to intervene with member behavior issues within 24 hours when necessary.
As a team we strive to reach consensus. However, there are practical limitations on our availability, and available time and energy. For major issues like Code of Conduct (CoC) changes, other major policy and procedure changes, and community changes, our current quorum threshold is 5 admins (which is a majority). For CoC violations by individual or small groups of users our current quorum threshold is 3 admins at the minimum. These limits may change as the composition and availability of the admin team changes.
A general outline of our response to an incident that needs intervention:
- Issue is reported via the
/admincommand (which sends a message to the#adminchannel) or admins observe it individually or collectively. - Admins that are logged in and present discuss it on
#adminimmediately and determine whether it's urgent enough to take action on immediately. - Assuming the issue can bear waiting, we wait and try to get as many admin perspectives, opinions, and recommendations as possible, and if possible, we try to come to consensus or quorum at minimum.
- Once we come to a majority or consensus decision, we collaborate on a response plan, including any writing we need for posts to
#announcementsor#meta. - It depends on context whether we collectively talk with any user(s) involved in the issue or whether one of us volunteers to talk with the user(s) about the issue and intercede or advise.
- Then one of us volunteers or is chosen to post anything relevant to
#announcementsand/or#meta. (Note that admins generally do not announce or discuss minor interventions with individual members. If member intervenions inspire policy or code changes, admins will handle these as separate issues.)
For smaller issues and requests, individual admins can do the work and then we talk about it in the #admin channel or log it in our centralized log of issues.
Examples of smaller requests are:
- Requests to remove spam.
- Requests to announce channels.
- Requests to change channels from public to private.
- Requests to deactivate an account (from the account itself).
There are two general classes of interventions:
- Policy, Code, or Culture changes
- Member interventions for CoC violations
In the case of Policy, Code, and Culture changes, admins collectively identify challenging patterns in the current interpretation of or enforcement of the current code of conduct, culture, and other community expectations and rules. Then they discuss the issue and possible remediation, collectively come up with proposed changes, and announce the proposed changes to the slack community, usually with posts to #announcements and #meta, encouraging discussion on #meta. Since most documents of policy, code, and culture live on the lgbt.technology website, admins usually invite feedback both via the /admin command or through GitHub Pull Requests.
In these cases, admins usually announce the effort multiple times over several days, to try to draw as many eyes and as much criticism and discussion as possible, so the changes can truly be said to be a community effort.
For member interventions the process depends very much on context. Depending on the severity of the issue, a member may start out getting warnings or other interventions prior to any temporary or permanent ban. In some cases admins may privately and trivially try to provide course corrections and other warnings before any formal process begins.
A list of possible member interventions:
- One off warnings from individual admins, trivially (e.g. "Hey, we don't use that term here." and a slackbot response about it - which members can and should use with each other, of course, or "Can you please refrain from doing X?").
- A formal intervention and warning (see below for an example).
- A private group discussion with the member and any number of admins who volunteer to participate.
- Other possible recommendations (see Recommendations below).
- A temporary ban or account lockout for a specific amount of time, usually 24 or 48 hours, possibly escalating if the admin team tries more than one temporary ban as a form of intervention.
- A permanent ban or account deactivation.
The goal for these communications are to present the member with an accounting of actionable behavior, admin expectations, and a social contract regarding warnings and other possible penalties. For example, a member might receive a direct message like:
Member, multiple people have reported your behaviour of the past few days as against the Code of Conduct and we admins agree.
You have used your unacknowledged privilege to tone police POC, and made it the mission of the #unlearn-privilege channel to be your personal soapbox to passive-aggressively explain how all of your enemies are wrong.
You also used your influence in the community to try to get away without criticism with lording your status over other people and questioning their own identities and labels.
You have also outed multiple members, doxxed them, and harassed and insulted them after having been clearly and repeatedly asked to stop.
The first line of the CoC states:
LGBTQ in Technology prioritizes marginalized people’s safety and well-being over privileged people’s feelings and comfort.
Given the community and admin concern, we are removing you from the community and your account will be permanently banned.
Note: This is an example. The actual structure of infractions and penalties will differ with each individual and each situation.
Also note: There will be some cases where the member is hostile or uncooperative with the warning or intervention and will not agree with the notification or penalty. Admins reserve the right to implement the penalties anyhow, even if the member does not agree.
Possible penalties:
- Official warnings
- If admins decide the user has a discrete number of warnings remaining, admins will tell them how many.
- Temporary bans
- A temporary ban is a time-limited period that a member's account is logged out and disabled from logging in again.
- Permanent bans
- A permanent ban is a permanent logging out and disabling of an account from further access and use.
An admin intervention may also recommend certain amends-making actions, depending on the situation. Possible recommendations (these are examples, not a comprehensive list):
- Public apology
- Private apologies to particular users or channels
- Fact correction
- Joining appropriate
#unlearnand#antichannels appropriate to the issue and emotional labor the member ought to follow up on - Leaving problem channels where the member tends to especially display the problematic behavior
- A policy of disengagement from particular topics, users, or groups of channels
- A policy of avoiding certain behaviors, like posting advertisements, or links to particular websites (especially if without context or care)
A warned member may choose not to follow admin recommendations. If so, this may factor against them if the member commits further CoC violations and requires more interventions.
Here's how interventions and notifications map:
| Intervention Type | Source | Notifications to Wider Slack Community |
|---|---|---|
| Policy Change Discussion | admin team | Summary to #announcements, pointing to summary and thread discussion to #meta. |
| Policy Change Discussion | member(s) | Message to #meta. Contact admins if you want an announcement in #announcements. |
| Member warning for CoC and culture violations | admin team | No notification. |
| Member temporary ban for CoC and culture violations | admin team | Notification to #admin-log channel including CoC code references and summaries of violating behavior. Note: temporary bans will not name the user to avoid dogpiling on them when they return. |
| Member permanent ban for CoC and culture violations | admin team | Notification to #announcements with member's username. Notification to #admin-log channel including CoC code references, summaries, and possibly links to code violating behaviors. |
| Admin team changes | admin team | Notification to #announcements. Back-end change to ?admin slackbot response and private administrative records. |
Example notifications:
Policy Change Discussions - admin source (#meta):
Hi everyone. To acknowledge your concerns, we're looking for feedback on what and how we should notify the community when we have to intervene with anyone committing CoC violations and warn, temporarily, or permanently ban a user or set of users.
Please respond with reactji for your recommendations. Of course, other feedback and ideas are welcome as well.
1️⃣ A vague post to #announcements that mentions an issue, but generally tries to reinforce community values without naming names.
2️⃣ A post to #meta that specifies CoC codes violated and summaries of what behaviors admins interpreted as violations without naming names.
3️⃣ A post to #meta that specifies CoC codes violated, with links to examples of the user(s) committing violations.
4️⃣ A post to #meta that names the specific user(s) committing the violations and which CoC violations were committed, with links to examples of those violations
5️⃣ A post to #meta informing that an intervention was made, with details (including naming names) in a separate public log channel.
6️⃣ No announcement. Just do the moderation work.
7️⃣ Other options that I will explain in my comments.
Policy Change Discussisons - admin source (#announcements):
We're seeking poll votes, feedback, and discussion on how best to notify you in the future when we have to intervene with members on CoC violations (in
#meta): https://lgbt.slack.com/archives/C038JGM02/p1612278147293300
Member Temporary Ban (#admin-log):
Hello. We've issued a temporary ban for a member who violated these CoC codes:
- No harassment.
- They harassed members and admins on 1/1/2020 and did not disengage when requested and prompted to do so.
- No incitement of violence towards any individual, including encouraging a person to commit suicide or to engage in self-harm.
- They suggested that members should engage in self-harm as a response to criticism and pushback.
We gave them a 48 hour time out and they will be allowed to return to the community after their account is reenabled.
Member Permanent Ban (#announcements):
The admin team permanently banned member
@examplefor serious and repeated violations of CoC codes. For more detail, see the message in#admin-logwith more details (including a link).
Member Permanent Ban (#admin-log):
More details regarding the permanent ban of member
@example
They repeatedly and unrepentantly violated the following CoC codes:
- No racist, sexist, cissexist, ableist or otherwise oppressive behavior is allowed, casual or explicit. This includes any harmful language, behavior, or action toward people of color, trans folks, disabled and other marginalized identities in our community. These are violations of the Code of Conduct.
- An example link to this behavior: (provide link)
- Summary providing relevant interpretation.
- No harrasment: Deliberate intimidation
- An example link to this behavior: (provide link)
- Summary providing relevant interpretation.
Note: These are examples and may not be entirely correct in detail or interpretation or severity, but are provided as an outline of expectations.
If interpersonal issues develop between individual members and admins, most admins will recuse themselves from decision-making and enforcement with those members. If uninvolved admins decide there is an intervention required, a recused admin may give opinion or advice, but they generally tend not to.
Note: there is an escape clause here. If a member or group of members gets along with no admins, some admins may still have to intervene. If possible, admins will be careful about the impacts and conflicts of interest inherent in such a situation, but admins must be able to and must be allowed to act in any case, even if interpersonal difficulties would normally lead to ethical recusal and inaction.
Every admin is community and consensus-oriented, and has a good grounding in social justice. At minimum they must be antiracist and they must understand, support, and enforce the CoC, including, for example, how we prioritize the experience of marginalized people over the feelings of privileged people.
If you want to discuss admin policy, process, behavior, if something seems off or inequitable or unjust, and it's not any particular admin but the team as a whole, or cultural expectations, the right place to take this for community and admin discussion is probably the #meta channel.
The admin team serves you and the community as a whole and it's our priority to make the slack as inclusive, welcoming, and equitable as we can. So please let us know if we're making mistakes, and talk about it!
Ways to negotiate communication with admins and organizing beyond #meta if you feel there's a clear issue and you think other admins can help:
- You can take your concerns to the entire admin team (but without public comment) by using the
/admincommand. This will be the quickest way to get admin team collective attention. (If for some reason the command fails, try it again. The command lives somewhat virtualized, and can fail if it's spun down.) - You can create a private channel and invite the admins you trust to it. You do not have to ask our permission first to do this.
- You can communicate with DMs with any individuals in the admin team you trust, and we will take your concerns seriously, respect the privacy you specify, and bring concerns forward, when you permit us to.
If an admin seems to have gone rogue or has violated these procedures, please notify the entire admin team with the /admin command, or any individual admin you're comfortable reaching out to. The admin team takes these procedures very seriously, and collective discussion and coordinated action is extremely important to us. Any admin taking solo action on any non-trivial issue is not tolerated and grounds for removal from the admin team.